Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Creating destruction: the political economy of zombie firms
(USC Thesis Other)
Creating destruction: the political economy of zombie firms
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Creating Destruction:
The Political Economy of Zombie Firms
Scott Wilbur
Dissertation for the Degree of Doctor of Philosophy
Political Science and International Relations Program
Faculty of the USC Graduate School
University of Southern California
May 2018
2
Contents
List of Tables and Figures 3
Abstract 4
Acknowledgments 5
Introduction 7
1. The Puzzle of Zombie Firms
State of Knowledge
Orienting Hypotheses
Research Design
15
16
25
35
2. Zombie Firms among Japanese Listed Firms, 1995-2008
Statistics on Zombie Firms among Japanese Listed Firms
Transparency
Bankruptcy Law
Conclusion
47
49
57
71
75
3. Zombie Firms among Japanese SME, 1994-2014
Statistics on Zombie Firms among Japanese SME
The Credit Guarantee System and Zombie SME
Drift in Japan’s Credit Guarantee System
Conclusion
80
82
86
101
114
4. Zombie Firms among Chinese Firms, 1998-2015
Statistics on Zombie Firms among Chinese Firms
Japanization Redux: Japanese Explanations for Chinese Zombies
Soft Budgetary Constraint as an Alternative Explanation
Conclusion
117
119
127
137
144
Conclusion
Main Findings and Contributions
Avenues for New Research
The Future of Zombie Firms
148
148
154
160
Appendix: List of Field Interviews 161
References
167
3
List of Tables and Figures
Tables
1.1. Operational Definitions of Zombie Firms
1.2. Case Studies of Orienting Hypotheses
1.3. Information for Case Studies of Orienting Hypotheses
3.1. International Comparison of Credit Guarantee Systems, 2014
3.2. International Comparison of Credit Guarantee Systems in Normal and Crisis Times
3.3. Central Government Budget Allocations to the Credit Guarantee System, 1999-2015
4.1. Chinese Zombie Firms by Industry, Listed Firms Database, 2013
4.2. Number of Bankruptcy Cases Disposed of by Chinese Courts, 2003 12
Figures
2.1. Zombie Firm Ratio among Japanese Listed Firms, FN Definition, 1988-2008
2.2. Zombie Firm Ratio among Japanese Listed Firms, CHK Definition, 1988-2008
2.3. Number of Zombie Firms by Measure of Financial Support (Excluding One-Shot Zombie
Firms), 1995-2008
2.4. Number of Zombie Firms among Japanese Listed Firms by Industry, 1995-2008
2.5. Zombie Firm Ratio by Industry among Japanese Listed Firms (Excluding One-Shot Zombie
Firms), 1995-2008
3.1. Zombie Firm Ratio among Japanese SME by Equity Capital, FN Definition, 1999-2008
3.2. Zombie Firm Ratio among Japanese SME by Debt Type, FN Definition, 2009-2014
3.3. Debt-Based Zombie Firm Ratio among Japanese SME, 1994-2014
3.4. Zombie Firm Ratio among Japanese SME by Firm Size, 2010-2014
3.5. Schematic Map of Japan’s Credit Guarantee System
3.6. Outstanding Loans and Outstanding Guaranteed Loans for Japanese SME and
Microenterprises, 1997-2014
3.7. Public Opinion on Issues Needing Government Action
4.1. Chinese Zombie Firms by Year, Listed Firms Database, 2000-2015
4.2. Chinese Zombie Firms by Year, Industrial Enterprise Database, 2000-2013
4.3. Chinese Zombie Firms by Province, 2000-2013
4.4. Zombie Firms among Chinese Firms by Ownership Type, 2000-2013
4.5. Profitability and Debt Ratios among Chinese SOE and Collective Enterprises, 1998-2013
4.6. Profitability and Debt Ratios among Chinese Private Firms, 1998-2013
4.7. Bank Nonperforming Loans to Total Gross Loans, 2010-2015
4
Abstract
Zombie firms—generally defined as unprofitable businesses that survive in the
marketplace through financial relief—have been widely characterized as harmful due to their
purported interference in the process of creative destruction after economic crises. Yet, while
many economists have studied the negative economic effects of zombie firms since the global
financial crisis, little research has been performed by political scientists on the political economy
of these problematic enterprises. What political conditions make zombie firms possible, and what
political conditions lead to their elimination? This dissertation develops preliminary theoretical
explanations to address these questions. Through three orienting hypotheses—transparency,
bankruptcy law and drift—the dissertation examines and elucidates political factors behind the
emergence and disappearance of zombie firms. It explores these hypotheses in two “extreme”
cases of zombie firms—Japan since the banking crisis in the 1990s and China since the global
financial crisis—utilizing a variety of information, including large-n statistics showing the
variation in the zombie firm ratio over time, secondary source material, government and
international organization reports, and 67 interviews with financial institutions, legal industry
professionals, government officials, and academics. The research uncovers evidence that is
largely consistent with these hypotheses in the case of Japan. In particular, lack of regulatory
transparency and an unconducive bankruptcy regime contributed to the rise of Japanese listed
firms with zombie status during the 1990s, until financial and legal reforms at the turn of the 21
st
century initiated a decline. By contrast, drift enabled the continuation of high coverage ratios in
Japan’s credit guarantee system for small and medium-sized enterprises (SME) well into the early
21
st
century, which perpetuated a considerably high zombie firm ratio among Japanese SME
throughout the 2000s and even into the 2010s. Research further establishes a connection
between bankruptcy law and zombie firms in China, and the existence of an alternative
mechanism—soft budgetary constraint—that has led to a considerable zombie presence among
Chinese state-owned enterprises. In finding support for these preliminary theoretical
explanations, the dissertation suggests hypotheses that might be further evaluated as additional
cross-national data with variation in the zombie firm ratio becomes available, as well as new lines
of inquiry related to the zombie firm issue in each case.
5
Acknowledgements
There is no such thing as a self-made man when it comes to finishing a dissertation and
Ph.D. program. Only with the help of many outstanding individuals and institutions was I able to
see these endeavors through to completion.
I would first like to acknowledge my dissertation committee: Saori N. Katada, Wayne
Sandholtz, David C. Kang, Gerardo Munck and Pierrette Hondagneu-Sotelo. Saori served as
my advisor since my first day in the program, and strongly promoted my academic development
with incisive feedback and constant moral support in my six-plus years at USC. She also ignited
my interest in the politics of SME policy, a subject that eventually led to my dissertation project
on zombie firms. Wayne meanwhile offered excellent research design advice as the project took
shape, and pushed me to make my research relatable to an audience outside my familiar niche of
Japanese politics. Saori and Wayne also deserve credit for encouraging me to apply for a
Fulbright grant during my fourth year, which ultimately enabled me to spend fifteen months in
the field in 2015-16 and gather information about zombie firms in Japan and China. I was
extremely fortunate to have them as committee co-chairs, and remain indebted to them for
cultivating me as a scholar in my time at USC.
Dave, Gerry and Pierrette also made key contributions as committee members. Dave
forced me to think about framing my project as a puzzle, and nurtured my understanding of
academia as a profession through countless activities at the USC Korean Studies Institute. Gerry
prompted me to think in more rigorous case study terms about my project as well as its
connections to larger theoretical debates, and broadened my knowledge of comparative politics
through his leadership of a comparative politics seminar throughout my time at USC. Pierrette
showed me the care and effort necessary to conduct research interviews, thereby deepening my
appreciation for the challenges and value of qualitative methods.
I would also like to acknowledge other individuals at USC who supported my Ph.D.
studies over the years. Many professors, including Jefferey Sellers, Diana Z. O’Brien, Anthony
Kammas, Pat James and Carol Wise, contributed to my learning. Program staff, including Cathy
Ballard, Veri Chavarin, Linda Cole, Indira Persad and Aurora Ramirez, assisted me in
important but often inconspicuous ways. Library staff, including Lourina Agnew, Mary Clark,
and Robert Labaree, made using the university’s library resources easy and enjoyable. Before
USC, four professors were integral to my academic growth and eventual decision to pursue a
Ph.D.: Kevin Doak, R. Bruce Douglass and James Lamiell at Georgetown University; and Philip
S. Hsu at National Taiwan University.
I have been fortunate to receive substantial institutional support over the years. At USC, I
received funding for my education and research from the Dornsife College of Letters, Arts and
Sciences, the School of International Relations, the Center for International Studies, the Korean
Studies Institute and the US-China Institute. I would like to specifically acknowledge the USC
Provost’s Ph.D. Fellowship Program, which was my primary source of institutional funding.
Outside USC, I was privileged to receive financial support from the Japan-U.S.
Educational Commission, whose many contributors—including Japanese and American
taxpayers—collectively enabled fifteen months of dissertation fieldwork through a Fulbright
6
Graduate Research Fellowship between September 2015 and December 2016. My host
institution in Japan, Waseda University, provided a wonderful base to conduct research, and my
host scholar, Shujiro Urata, graciously assisted my project despite an astounding workload. The
Commission also funded one month of fieldwork in China in fall 2016. My host institution in
China, the Zhejiang University of Technology’s China Institute for Small and Medium
Enterprises led by Renyong Chi, presented a very hospitable environment for gathering
information about zombie firms in China. Lastly, I would be remiss if I did not specifically thank
Matt Sussman and Keiko Toyama, whose work at the Commission was of fundamental
importance to my productivity in the field.
I incurred enormous intellectual debts during my time in Japan and China. The greatest
debt was to the 67 individuals who selflessly granted me interviews. Ironically, while these people
shared profound insights about many issues surrounding zombie firms, I am unable to publicly
thank them because of the anonymity of their interviews. I also owe a debt of gratitude to the
many scholars who helped me develop my project and kept me on the right track in the field:
Kyoji Fukao, William Grimes, Kathryn Ibata-Arens, Kenneth Mori McElwain, Hideaki
Miyajima, Gregory Noble, Gene Park, Paul Scalise, Kay Shimizu and Naoyuki Yoshino. Yasuo
Goto both wrote a treatise on SME in Japan that impacted my thinking about the subject, and
became a coauthor and close friend over the course of several meetings in Tokyo. In China,
Xiaohong Chen, Susan Finder and Douglas Fuller pointed me in the direction of the current
debates about bankruptcy law that helped me understand the law’s practice. Finally, Xiao Hua
and Fan Dong were amazing research assistants who helped me make the most of my short time
in country.
I also received tremendous support from friends in Japan to whom I am intensely
grateful: Izumi Inagaki, Naoko Kato, Masa Murasawa, Tetsuaki Nakao, Takao Nozaki, Yosuke
Onodera, Takashi Sakaki, Takamasa Uesugi and Eisaku Yamauchi. At Waseda’s Kikuicho
dorm, Matthieu Felt, Colin Jones, Kalle Pihlainen and Tyler Walker were resourceful neighbors
and treasured companions on many outings in Tokyo.
Another notable source of support were my peers at USC, especially Justin Berry, Fabian
Borges, Juve Cortes, Tyler Curley, Adam Feldman, Eric Hamilton, Tom Jamieson, Peter
Knaack, Corrina Liu, Matt Mendez, Mark Paradis, Simon Radford, Mariana Rangel, Seanon
Wong and Mingmin Yang. These fellow journeyers not only showed me how to survive the
process of getting a Ph.D.; they enormously enriched my life by making my time in Los Angeles
fun and interesting, by keeping my spirits up when I faced challenges, and by giving me
perspective. I owe them a special debt of gratitude.
Finally, I would like to thank my family for its love and support. My parents Susan and
Stuart never wavered in their encouragement of my graduate studies, even when my learning
took me on a circuitous route through Taiwan which lasted longer than anyone expected. The
worldliness of my grandfather, Leslie Wilbur, was a prime reason for my initial interest in
international affairs and politics, and my friendship with him continues to be one of the most
cherished relationships in my life. It is to him, with much affection, whom I dedicate this work.
7
INTRODUCTION
Scholarship in political economy typically assumes that economic crisis transforms state
policy. While not all political economists see crisis as automatically effecting new and different
policy directions—some argue that crisis induces critical junctures which then possibly result in
policy change (Gourevitch 1986; Blyth 2002; Capoccia and Kelemen 2007), while others note
that not all crises even generate critical junctures (Collier and Collier 1991; Kaufman 2017)—the
predominant belief is that crisis leads to adjustment (Stallings 1992; Drazen and Easterly 2001).
As Rodrik (1996, 26) describes, “If there is one single theme that runs through the length of the
political economy literature it is the idea that crisis is the instigator of reform.”
This dissertation takes up the opposite scenario. Instead of engaging crisis as a situation
where the deteriorating performance of status quo policy gives way to reform, it explores when
crisis is shadowed by the persistence of preexisting arrangements which impair recovery. In other
words, it shifts attention away from crisis as a catalyst of change, toward crisis as a revealer of
rigidity. Stated more bluntly, it is interested in policy constancy after economic shocks.
This dissertation examines this sort of contrarian crisis response in the realm of a recently
identified economic problem—zombie firms. In general definition, zombie firms are inefficient
businesses that remain in the marketplace instead of pursuing restructuring or bankruptcy. First
reported as a corporate phenomenon that undermined the recovery of Japan following its
banking crisis in the 1990s, zombie firms have been increasingly observed since the global
financial crisis in both high-income and developing countries. Their emergence has concerned
government policymakers in wide variety of countries, including the United Kingdom, the
8
United States, China, South Korea, and European Commission member states, and
consternated finance industry professionals, many of whom have been disturbed by these firms’
stifling impact on resource reallocation and macroeconomic growth. Speaking on zombie firms
in Europe, the global head of restructuring at a major professional services firm ominously
warned, “The fundamental tenet of capitalism, which holds that some bad companies need to fail
to make way for new and better ones, is being rewritten” (Stothard 2013).
Zombie firms have a harmful reputation because of their purported interference in
creative destruction, the process of corporate revitalization introduced in Schumpeter’s classic
work Capitalism, Socialism and Democracy. For Schumpeter, creative destruction—“the process of
industrial mutation […] that incessantly revolutionizes the economic structure from within,
incessantly destroying the old one, incessantly creating a new one”—was capitalism’s core
characteristic, a “perennial gale” that continuously provided innovation and energy for economic
development (Schumpeter 2006, 83–84). Economists subsequently invoking Schumpeter
emphasize creative destruction’s importance especially after crises, when growth is scarce and the
churn of market exit by inefficient firms, entry by new firms, and expansion by successful firms
can spur much-needed productivity (Cox and Alm 1992; Hall, Farber, and Haltiwanger 1995;
Gomes, Greenwood, and Rebelo 2001). While recessions may prompt layoffs and lower incomes,
they also promise a paradoxical “cleansing effect” beneficial to economic regeneration and
renovation (Caballero and Hammour 1994), and can be an optimal time for government policy
to concentrate on simultaneous job destruction and creation (Caballero and Hammour 1996).
Zombie firms, by contrast, represent the absence of such policy initiative, since their
survival depends on the perpetuation of a status quo bias that essentially tolerates widespread
financial distress and postpones restructuring in the corporate sector. More problematic still,
zombie firms are believed to impose a negative externality on healthy firms. Not only are
9
zombies themselves inefficient, but their distortion of market competition depresses prices for
their goods and increases labor costs, which lowers the profits and collateral that new and more
productive firms could earn, thereby impairing market entry and investment (Caballero, Hoshi,
and Kashyap 2008, 1944). These spillover effects contribute to deflation in post-crisis economies,
and can even turn viable firms into zombies (Hoshi 2006, 33).
This dissertation asks two basic and interrelated questions about zombie firms: What
political conditions make these problematic firms possible, and what political conditions lead to their elimination?
While a burgeoning economics literature has emerged around zombie firms (Ahearne and
Shinada 2005; Caballero, Hoshi, and Kashyap 2008; Fukuda and Nakamura 2011; Imai 2016;
Nie et al. 2016b; McGowan, Andrews, and Millot 2017; Acharya et al. 2017), little if any work in
political economy addresses their political background, or the policy actions that ultimately
achieve their market exit. The dissertation thus aims to build theory about political
configurations that make zombie firms likely to appear after crises, heeding Cohen’s call for
political economy scholarship that takes on broad questions and poses new theory (Cohen 2010,
889). In a nutshell, it seeks to understand the political economy of firms that are creating destruction
in post-crisis recessions by impeding what Schumpeter and many others see as the healthy
evolutionary cycle of capitalism.
Due to the newness of zombie firms as a topic in political economy, the dissertation
employs case studies to develop theory about political causal mechanisms behind these firms. It
primarily examines two national cases believed to sustain zombie firms: Japan in the “lost
decades” after the banking crisis in the early 1990s, and China after the global financial crisis.
The former case is generally considered the paradigmatic example of zombie firms in a post-crisis
economy; the latter case has garnered concern in recent years for possibly sustaining many
zombies and repeating the Japanese experience (Orlik 2011; Guilford 2014; Lewis, Mitchell, and
10
Yang 2017). China is not special in this regard, as many countries both inside and outside East
Asia have allegedly faced “Japanization” after 2008 insofar as their economies have supported
zombie firms, including South Korea, Eurozone countries, the United Kingdom, and the United
States. Therefore, the dissertation’s use of Japan’s encounter with zombie firms as a prism for
looking at China is meant to address the comparative question of whether the causal mechanisms
behind these firms are unique to Japan, and the extent to which they operate in other national
and temporal settings.
In choosing Japan and China—countries which represent “extreme” cases because of
their reportedly high zombie firm ratios—the dissertation is generally similar to other works that
intentionally focus on instances of policy failure (Scott 1998; Wong 2011), which puts it at risk of
criticism for selection on the dependent variable (Geddes 1990). The dissertation nevertheless
takes this approach because of the early stage of the research agenda and data limitations on
zombie firms, and the dissertation’s primary goal of providing heuristic value in suggesting
possible causal pathways and independent variables that future studies can test against once more
cases with variation on the dependent variable become available (George and Bennett 2005). For
the time being, it is first useful to refine initial hypotheses about zombie firms’ political
underpinnings, meaning that the dissertation’s arguments about causality are necessarily
preliminary, not definitive.
Political economy scholarship is sometimes derided for “peer envy” of the economics
discipline and the academic respectability of its hard-science models (Cohen 2009, 441). The
dissertation turns this notion on its head by harnessing economics research to inform its
investigation of zombie firms in its case studies on Japan and China. In particular, it utilizes
econometric studies of these firms among stock exchange-listed firms in Japan (Nakamura 2017),
small and medium-sized enterprises (SME) in Japan (Imai 2016; Goto and Wilbur Forthcoming),
11
and private and state-owned enterprises (SOE) in China (Nie et al. 2016b). Some of these studies
are in the native language of the target country (Nie et al. 2016b), while others are original efforts
by target country collaborators and the author (Goto and Wilbur Forthcoming). Together, the
studies depict the variation in the zombie firm ratio over time—the dissertation’s outcome
variable of interest—among these different corporate categories in the two countries. The
significance of their estimates is bolstered by their generally large number of observations, which
for several studies reaches into the tens of thousands. This quality lends representativeness to
their estimates of zombie firms’ presence, and helps the dissertation hone in on times when
zombie-supporting policy configurations are likely in place.
The dissertation’s exploration of zombie firms’ political background is guided by three
orienting hypotheses based on an interdisciplinary synthesis of theory in economics, law, and
political science. The first of these hypotheses builds on the finding in economics research that
troubled banks—banks with capital ratios near the required minimums and banks with
deteriorating nonperforming loan ratios—can have an incentive to practice forbearance toward
weak firms (Peek and Rosengren 2005; Fukuda, Kasuya, and Nakajima 2006), and proposes that
this zombie-inducing scenario is possible when politicians lack regulatory information about
banks’ health due to the politicians’ structural relationship with bureaucrats. The second
hypothesis is premised on the idea from legal studies that restructuring through bankruptcy can
be a safer, less-costly choice for distressed firms than enduring protracted indebtedness
(Berkovitch and Israel 1998), and argues that the incentive structure of prevailing bankruptcy law
is an important condition that motivates impaired firms to either pursue or postpone
restructuring, with the latter possibility being a possible reason why some indebted firms instead
seek financial relief and willingly become zombies. The third orienting hypothesis holds that the
moral hazard implicit in government policies to guarantee bank loans may not only stimulate
12
banks to lend to zombie-prone firms in crisis moments, but can outlast crises and continue to
contribute to zombie creation when the policies are not updated because of limited public access
to policymaking and concentrated stakeholder support for the status quo—a process known in
the political science literature on institutions as “drift” (Hacker, Pierson, and Thelen 2015).
The dissertation first examines these orienting hypotheses in the paradigmatic case of
Japan during the 1990s and 2000s. It inspects the first and second hypothesis among listed firms,
and the third hypothesis among SME. It makes this within-case distinction because of the
applicability of Japan’s credit guarantee system, which serves only SME and thus makes Japan’s
SME a unique corporate subject for investigating the third orienting hypothesis. After the initial
exploration of the three hypotheses in Japan, the dissertation then evaluates the extent to which
these three potential explanations account for the zombie firms observed after the global financial
crisis in China, one of many countries where concern about zombie firms and potential
“Japanization” has been substantial in since the crisis.
Briefly summarized, the dissertation finds considerable empirical support for each of its
orienting hypotheses in Japan, but less robust support for their explanatory power in China. For
China, the orienting hypothesis with the best predictive strength is the hypothesis on bankruptcy
law, while there is little evidence that is consistent with the other two hypotheses. Instead, an
alternative explanation that soft budget constraints facilitate zombie formation may account for
why zombies increased in China after the crisis, especially among the country’s SOE.
Collectively, these results clarify certain policy constellations that merit attention in future
research on zombie firms’ political origins, and hint at the degree to which the zombie-creating
arrangements in the model case of Japan have resurfaced elsewhere more recently.
The dissertation’s theoretical findings primarily contribute to the small body of political
economy scholarship on policy continuity after economic crisis, which has already shed light on
13
areas like exchange rate regimes (Cohen 2008), foreign direct investment (MacIntyre 2001), and
portfolio investment (Pepinsky 2014). They also contribute, though in a narrower way, to the
large literature on the relationship between political institutions and economic development that
is not exclusively concerned with post-crisis policy response (North 1990; Przeworski et al. 2000;
Rodrik, Subramanian, and Trebbi 2004; Acemoglu, Johnson, and Robinson 2005; Hicken,
Satyanath, and Sergenti 2005; Ang 2016).
Additionally, the dissertation’s individual case studies on zombie firms yield surprising
empirical findings that defy conventional theoretical expectations about political variables and
economic performance. For example, about listed firms in Japan, one would assume that
developmental state policies would lead to few zombies, since these policies have traditionally
been associated with vigorous growth (Johnson 1982). Nonetheless, zombie firms surged among
listed firms in the early years of Japan’s banking crisis in the 1990s, an outcome that disputes
certain positive features of the developmental state like bureaucratic expertise and cross-
shareholding. Similarly, the dissertation identifies a lingering pattern of protectionism toward
Japan’s SME underlying a substantial zombie presence during the 2000s, despite prominent
theory that majoritarian reform in Japan’s electoral system in the mid-1990s moved the country’s
public policy in a neoliberal direction (Rosenbluth and Thies 2010). Furthermore, the case study
on China finds that decentralization may have exacerbated the incidence of zombie firms among
SOE after the global financial crisis, in contrast to the notion that power devolution to local
governments enhances economic efficiency in authoritarian states (Montinola, Qian, and
Weingast 1995).
The dissertation is structured in the following way. Chapter 1 lays out the academic
debate about zombie firms; presents an analytical definition of zombie firms used in the
dissertation’s analysis; describes the three orienting hypotheses that guide the dissertation’s
14
investigation on zombie firms’ political causes; and outlines the research design. Chapters 2
through 4 cover individual case studies. Chapter 2 examines the first two orienting hypotheses in
the case of zombies among Japanese listed firms in the period 1995-2008. Chapter 3 inspects the
third orienting hypothesis in the lesser-known yet arguably more significant case of zombie SME
in Japan in the years 1994-2014, a period that extends well beyond the immediate aftermath of
Japan’s banking crisis. Chapter 4 then explores the three hypotheses for zombies in Japan in the
more recent case of China in the period 2000-2013, and investigates an alternative hypothesis for
China’s zombies related to soft budgetary constraint. The Conclusion recaps the dissertation’s
findings, discusses their broader implications for the study of zombie firms, and suggests related
areas for future research.
15
CHAPTER 1.
The Puzzle of Zombie Firms
In recent years and especially since the global financial crisis, zombie firms—generally
defined as insolvent businesses that remain in the marketplace instead of pursuing restructuring
or bankruptcy—have become increasingly widespread. From British hotels and catering
companies, to Korean shipbuilding corporations and Chinese enterprises in smokestack
industries like coal and steel, zombie firms have made media headlines around the world
(Stothard and Giles 2012; Stothard 2013; Jung-a Song 2015; Wildau 2016b; Gough 2016). They
have also been criticized by economists, who argue that zombies harm macroeconomic health by
inefficiently using financial resources, postponing restructuring, and lowering their more
productive competitors’ potential market gains.
While journalists and economists have been quick to respectively label and critique
zombie firms, political economists have produced less research on the political reasons why these
problematic businesses emerge and endure after crises. To help rectify this gap, this dissertation
explores political conditions that enable zombie firms to exist, as well as conditions behind the
decrease of these troubled firms. Because of the nascent stage of the research agenda on this topic
and limited data availability, the dissertation employs a case study research design to refine
several orienting hypotheses about the political mechanisms that make and unmake zombie
firms. To do so, it scrutinizes two “extreme” cases where zombies have generally thought to have
occurred in outsized amounts: Japan’s “lost decades” in the 1990s and 2000s, and China in the
period 2000-2013.
16
This chapter outlines the dissertation’s theoretical arguments and research design. The
“State of Knowledge” section reviews the literature on zombie firms and their conceptualization,
justifies an analytical definition used in the dissertation’s subsequent analysis, and identifies an
important gap in the discussion about zombie firms, namely, an absence of political explanation.
The “Orienting Hypotheses” section posits three conjectures about political causes of zombie
firms. The “Research Design” section describes the dissertation’s rationale for using a case study
design to refine these hypotheses, and for selecting Japan and China as cases. It also explains the
nature and limitations of the data for these cases, and presents a mixed-methods research
approach that uses large-n descriptive statistics and qualitative data, including interviews, for
understanding the prevalence of zombie firms and the mechanisms that sustain them.
State of Knowledge
Zombie firms have captured the popular imagination in recent years, corresponding to
the broader cultural fascination with zombies in the early 21
st
century. Generally connoting a
business enterprise that “is kept alive by lenient creditors and low interest rates even though it is
too weak to invest or expand” (Financial Times n.d.) or that “needs constant bailouts in order to
operate, […] can manage to pay the interest on its debts but not reduce the actual debt, or […]
appears healthy from the outside but is in fact teetering on the edge of ruin” (Golub and Lane
2015, 47), the term “zombie firm” has surged in usage since the global financial crisis, when
many slumping businesses have seemingly avoided restructuring or liquidation because of
supportive financial policies.
In journalistic assessments, zombie firms are exemplified by well-known distressed
American companies such as Citigroup, Bank of America, General Motors and Chrysler that
17
survived the crisis thanks to bailouts by the Troubled Asset Relief Program (Cooley 2009). But
detection of zombie firms has not been limited to the United States. Indeed, zombie firms have
been reported in numerous developed and developing countries, including the United Kingdom
(Jones and Giles 2012), France (Bird 2015), Greece, Spain, Italy (Stothard 2013), Portugal (The
Economist 2013), South Korea (Lee 2015; Jung-a Song 2015), India (MacAskill and Kumar
2014), and China (Miller 2016). According to leading international business media like the
Financial Times, The Economist and Reuters, they are now a widespread phenomenon in the global
economy.
While reports on zombie firms have grown substantially in recent times, likely due in part
to the term’s sensationalistic quality, references to this class of corporate entity often take its
meaning for granted, rather than explicitly defining it. As a result, popular discussion on zombie
firms typically suffers from the problem of concept stretching (Sartori 1970), whereby observers
claiming to see zombie firms in fact perceive businesses with different attributes which do not
constitute the same kind of enterprise. Idiosyncratic conceptualization may not matter for
individual instances of zombie firms, but it complicates comparative inquiry and hinders
cumulative knowledge building, since it conflates disparate phenomena and obscures the
empirical reality of zombie firms’ presence in both specific economic cases and the universe of
economies.
Therefore, a serious examination of zombie firms must start from a precise analytical
definition of these firms’ constitutive features that includes certain characteristics and excludes
others. For this task, the descriptor “zombie” itself is not necessarily helpful, since it conjures up
fantastic imagery of “the living undead” who have died and been brought back to life, whereas
businesses never truly experience death like zombies because they always retain some assets.
Financially distressed firms may undergo restructuring, or they may recover profitability and
18
return to viability, but they cannot completely lose value prior to revival, because in such
circumstances they would be de facto new firms. Perhaps for this intrinsic reason, there exists no
official or legal definition of zombie firms, since such corporate status is, strictly speaking,
ambiguous if not impossible, not to mention pejorative as well.
1
Further complicating any definition of zombie firms is the fact that determining corporate
value involves a degree of subjectivity. While insolvency is often straightforwardly defined as the
condition of having more liabilities than assets, sometimes an asset’s valuation is not its real
valuation. For instance, a firm that is financially distressed may be distressed because it borrows
money to spend on research and development, which makes its balance sheet show negative
equity. Moreover, the value of the new product created by this expenditure may be uncertain.
However, if investors are reasonably confident in the firm and this product, the firm will still be
viewed as viable (Interview 17). This pattern is frequently observed in startup firms, which may
endure significant periods of virtual insolvency before turning profitable. Well-known examples
of firms like this include Amazon and Tesla. But older firms that experience negative profits for
years can remain viable, too (Interview 52), as can firms that engage in highly leveraged
transactions that expose them to debt but are ultimately beneficial for tax reasons (Andrade and
Kaplan 1998). Because assessing corporate value is complex, finance industry and bankruptcy
professionals often use case-specific criteria as well as precise measures like return on assets and
leverage ratios to gauge firm health, and eschew vague terms like “zombie” in practice, even if
1
According to Nie et al. (2016a, 63), at a State Council executive meeting on December 9, 2015, Chinese Premier Li
Keqiang said that businesses which run losses for three straight years and fail to make structural adjustments must be
“cleaned up” (chuqing) through measures such as asset restructuring, transfers of property rights, and closure. Some
observers interpreted Li’s statement as being an official Chinese definition of zombie firms as firms that have
negative profits for three straight years. Because the Chinese government’s use of this definition in policymaking is
unclear, however, it is questionable whether it is truly an official definition.
19
they have a general idea of its connotation (Interview 31).
2
In other words, not only is there no
legal definition of zombie firms, there is no industry standard either.
Rather than in the worlds of law and business, zombie firms first became a distinct
phenomenon in academia, particularly in the discipline of economics. The earliest known
reference to zombie firms was made by Kane (1987) regarding insolvent savings and loan
companies in the United States. Kane broadly defined zombie firms as “any institution with
negative capital and income according to generally accepted accounting principles,” and warned
that such firms possessed deleterious characteristics like fictitious zombies, because they “wreak
havoc in financial markets by feeding on the markets of the living, thereby turning competitors
into zombies, too.” More specifically, zombie savings and loan companies posed a policy
problem, Kane argued, because their survival both forced healthy competitors to pay higher
deposit rates and lessened their ability to issue loans, and made them pay higher premiums to the
Federal Savings and Loan Insurance Corporation, thereby depressing overall industry profit
margins (Kane 1987, 78).
Following the initial appearance of zombie firms in economics research in the late 1980s,
the concept reemerged in diagnoses of Japan’s prolonged economic slump in the 1990s and
2000s, a period now widely known as Japan’s “lost decades” due to the country’s chronically low
annual GDP growth of around 1 percent. In the late 1990s and early 2000s, journalists and
economists including Federal Reserve Chairman Alan Greenspan began pointing to zombie
firms as both a symptom and a cause of Japan’s steep economic downturn after the burst of an
2
For example, a 2001 report by a multinational bank on corporate health in Japan maps out four general statuses
according to a two-by-two typology table with “leverage” on the horizontal axis and “RoA” on the vertical axis.
Firms with high RoA and high leverage are labelled “daring”; firms with high RoA and low leverage are labelled
“strong”; firms with low RoA and high leverage are labelled “weak”; firms with low RoA and low leverage are
labelled “plodding” (Feldman 2001, 33).
20
asset bubble in 1991 (Henry 1997; Chandler 1999; Landers 1999; Fackler 2003; Katz 2003a;
Koo 2008, 35). While economists at the time generally failed to specify what they meant by
“zombie firms” and did not always use this precise term, they commonly identified an disturbing
growth in indebted and inefficient businesses that stayed operational through various forms of
financial support, including low interest rates and forbearance lending (Katz 2003b; Sekine,
Kobayashi, and Saita 2003; Ahearne and Shinada 2005).
3
A few economists suggested that these
troubled firms likely received credit because of banks’ underlying nonperforming loan problems
(Peek and Rosengren 2005), but the general thrust of the discussion blamed the firms for
problematically lingering in the market, which observers saw as thwarting creative destruction
and hindering Japan’s revitalization.
Zombie firms’ role in Japan’s stagnation was most explicitly and influentially asserted by
the economists Hoshi and Kashyap in several economics journal articles starting in the second
half of the 2000s (Hoshi 2006; Caballero, Hoshi, and Kashyap 2008; Hoshi and Kashyap 2010,
2011). In particular, a 2008 coauthored article by Caballero, Hoshi, and Kashyap (henceforth
“CHK”) in the flagship journal American Economic Review transformed the academic study and
policy debate around zombie firms because of the article’s clear conceptualization of zombie
firms and the magnitude of its findings. The article provided a parsimonious criterion for
identifying zombie firms—it defined zombie firms as companies whose interest payments are
lower than the hypothetical risk-free interest payments (Caballero, Hoshi, and Kashyap 2008,
1948)—and examined the prevalence of enterprises meeting this criterion on the Tokyo Stock
Exchange (TSE) between 1981 and 2002 with data from the Nikkei Needs Financial dataset. It
uncovered a zombie firm ratio that rapidly expanded from slightly more than 5 percent of firms
3
Other economists around this time noted the increase in highly-indebted firms supported by forbearance lending,
but did not explicitly call them “zombie firms” (Sekine, Kobayashi, and Saita 2003).
21
in 1991, to nearly one-third of the Japanese firms it sampled in the late 1990s and early 2000s
(Caballero, Hoshi, and Kashyap 2008, 1945).
The substantial amount of zombie firms among exchange-listed businesses suggested
strong evidence for CHK’s contention that zombie firms were a major culprit behind Japan’s
post-bubble malaise. Because banks would only allow below-market interest payments from
financially-struggling firms, CHK theorized, the sizeable number of firms making discounted
interest payments implied that weak firms comprised a large segment of Japan’s economy and
“congested” the marketplace by failing to exit, thereby hurting the remaining healthy firms.
CHK tested this hypothesis and found that firms identified as zombies not only defied the normal
patterns of creation and destruction that happen after economic shocks, but also depressed
productivity in sectors where their presence was higher and lowered their more productive
competitors’ investment and employment growth. Zombie firms thus caused deflationary
pressures in the wider economy, and even led some healthy firms to become zombies as well
(Caballero, Hoshi, and Kashyap 2008, 1944–46). In short, zombies cumulatively weighed down
Japan’s potential recovery and prolonged its low growth after the bubble.
The publication of CHK’s article gave the concept of zombie firms much greater
prominence and shaped the subsequent scholarly and policy debates about the causes of Japan’s
lost decades. Many economists cited CHK’s zombie firm argument as a persuasive and at least
partial explanation for Japan’s sustained downturn (Griffin and Odaki 2009; Kobayashi 2009,
339; Garside 2012, 90–94). Warnings about zombie firms’ negative effects, including reference to
CHK’s article, also appeared in official reports and working papers about Japan’s need for policy
reform by international organizations like the OECD and International Monetary Fund (OECD
2015, 23; Lam and Shin 2012, 3; Arbatli et al. 2016, 8), and in policy research by leading
22
economic think tanks like Bruegel that admonished other countries to avoid Japan’s example
(Darvas 2013, 7).
Since CHK’s article, the idea that zombie firms have been responsible for Japan’s
protracted slowdown has become a tenet of “Japanization,” a popular disparaging term for
Japan’s painful and seemingly unending predicament of secular stagnation that has served as a
cautionary tale for other economies trying to rebound after the global financial crisis (Kawai and
Morgan 2013, 11; Hendrickson 2013; Pesek 2014). For example, the Bank of England’s
November 2012 Financial Stabilization Report referenced CHK in warning that the United
Kingdom should draw several lessons from Japan’s lost decades, including understanding the
risks of extending credit support measures to weak firms over long periods of time (Bank of
England 2012, 29–31). While not citing CHK per se, former US Treasury Secretary Larry
Summers discussed the threat of Japan-like zombie firms in a major 2012 speech about the post-
crisis American economy, demonstrating a belief that such businesses contributed to Japan’s
malaise and could stymie the US recovery if not prevented (Summers 2012, 72).
Apprehension about zombie firms also appeared in continental Europe around the same
time. Speaking in early 2013, Sony Kapoor, head of the Brussels-based think tank Re-Define,
revealed, “Behind the scenes there is an emerging policy debate about how many zombie
companies there are and how bad the problem really is. This is an issue both the European
Central Bank and the [European] Commission are concerned about” (Stothard 2013). More
recently, in October 2015, Liu He, vice chairman of China’s National Development and Reform
Commission and the top economic advisor to President Xi Jinping, proclaimed the Chinese
central government’s desire to accelerate the elimination of zombie firms as an important step in
productivity-oriented economic reform (Zhongguo Xinwen Wang 2015). Liu’s announcement
was quickly echoed by influential pro-reform voices like Caixin Media editor-in-chief Hu Shuli,
23
who stressed that China should learn about Japan’s suffering from zombie firms during the lost
decades (Hu 2015).
4
Despite increasing global apprehension about zombie firms augmented by CHK’s study
and concern about Japan’s notorious experience, some scholars questioned the correctness of
identifying zombie firms on the sole basis of discounted interest payments. Most notably, Fukuda
and Nakamura (henceforth “FN”) countered in several works that below-prime rate interest rates
are insufficient to indicate zombie firms, because Japan adopted a quantitative easing monetary
policy in the 2000s that enabled many firms, including healthy ones, to borrow at very low
interest rates (Fukuda and Nakamura 2011; Nakamura and Fukuda 2013; Nakamura 2017).
5
To more accurately gauge the presence of zombie firms, FN proposed the addition of two
conceptual criteria, profitability and evergreen lending, to CHK’s interest payment criterion.
Profitability is important, FN argued, because firms with earnings before interest and taxes
(EBIT) exceeding their hypothetical risk-free interest payments are generally healthy, and such
firms’ good financial standing may in fact account for why their interest payments are lower than
what the prime rate would stipulate. Evergreening lending is also relevant, since financially-
troubled firms sometimes request new loans from banks to pay the interest on existing loans when
the firms are indebted. In the case of post-bubble Japan, evergreen lending was particularly
prevalent among overleveraged banks that provided debt relief to distressed firms to hide the
value of the banks’ nonperforming loans. Evergreen lending thus enabled weak firms to make
interest payments at prevailing market rates without concessions, an outcome that CHK’s
4
Hu’s voice may be considered influential due to the large and powerful readership of the biweekly business
magazine Caijing of which she is founding editor. Caijing’s Chinese and English websites, which are loosely resemble
nytimes.com, reportedly draw around 3.2 million unique visitors each month. The print version is widely read by
China’s governmental, financial and academic elites (Osnos 2009, 56).
5
Japan’s adoption of a zero-interest rate policy in early 2000s also intensified competition between banks, leading
many banks to offer interest rates below the prime rate to attract borrowers (Interview 21).
24
Table 1.1. Operational Definitions of Zombie Firms
A firm is a zombie if it… CHK FN
pays interest rates below the hypothetical prime rate X X
is unprofitable (EBIT are below the hypothetical interest payments) X
receives evergreen lending (borrowings increase in current period t after
total external debt exceeds half of firm assets in period t-1)
X
zombie firm definition omits because it only focuses on firms’ payment below the hypothesized
risk-free rate (Fukuda and Nakamura 2011, 1126–27; Nakamura and Fukuda 2013).
Since FN offered their revised conceptualization, it has quickly gained traction among
economists investigating zombie firms in Japan and elsewhere (Kwon, Narita, and Narita 2015;
Imai 2016; Nie et al. 2016b; Tan, Huang, and Woo 2016). This trend suggests that a consensus
may be emerging around the relative merit of FN’s more strenuous definition.
6
It also implicitly
calls into question the zombie firm explanation of Japan’s lost decades that has been largely
inspired by CHK’s research, as well as the policy recommendations which it has spawned. At the
same time, FN’s work has been faulted for selectively looking at listed firms (it uses a similar
sample universe of TSE-listed firms as CHK, but with data from Corporate Financial Databank
from the Development Bank of Japan), whereas most businesses in Japan and other economies
are small and medium enterprises (SME) that do not publicly trade their stock (Kwon, Narita,
and Narita 2015; Imai 2016).
Because profitability and evergreening are essential to the concept of zombie firms, this
dissertation prefers the FN definition of zombie firms instead of the CHK definition that relies
6
This is not to say that the CHK definition is no longer in use. For example, a recent OECD working paper
employs the CHK definition without referencing the FN definition in a cross-country analysis of zombie firms’
effects on investment and employment in the period 2003-2013 (McGowan, Andrews, and Millot 2017). Other
studies agree that the FN definition presents a conceptual improvement over the CHK definition, but caution that it
may still misidentify non-zombie firms as zombie firms if the non-zombies undergo temporary profit decreases (Imai
2016; Nie et al. 2016b).
25
solely on discounted interest payments. While CHK and FN differ in their respective zombie firm
definitions, however, it bears noting that their work similarly focuses on identifying zombie firms
and analyzing their economic effects. Both CHK and FN, as well as more recent studies of
zombie firms utilizing their respective definitions (Kwon, Narita, and Narita 2015; Imai 2016;
Saruyama and Yan 2016; McGowan, Andrews, and Millot 2017), do not explain the political
conditions that cause these controversial firms to emerge in the first place. This omission is
noteworthy, since zombie firms have been widely depicted as malefic due to their allegedly
negative impact on economic growth, and since policymakers around the world have shown
strong interest in suppressing them. This dissertation thus attempts to addresses this gap in the
research on zombie firms by approaching them from a political economy perspective.
Orienting Hypotheses
At first blush, zombie firms and politics might appear unrelated, since these firms’
defining characteristics—unprofitability, below-prime rate interest payments, and
evergreening—may be purely financial matters determined by firms’ individual business
operations and their lending institutions. For instance, businesses may lose profitability due to
any number of economic reasons tied to their revenue and expenses, such as consumer
preference changes, technological shifts, and domestic or foreign demand swings, which have
seemingly little directly to do with politics. Similarly, banks’ decisions to grant interest rate relief
or roll over loans may be a rational economic choice based on the belief that recipient firms’
health might eventually recover and enable the banks to keep benefiting from the lending
relationship (Peek 2009). Simply put, the connection between zombie firms and politics is not
obvious, and it may seem questionable that they are connected.
26
Nonetheless, important conditions with political origins can underlie zombie firms’
constitutive features. For one, banks’ willingness to assist weakly-performing firms by discounting
interest payments and practicing forbearance may be motivated not only by the belief that the
firms will eventually recover repayment ability after a temporary downturn, but by the desire to
postpone the realization of bad loans. This latter motive may be especially apparent during
serious recessions, when banks have accumulated vast amounts of debt whose repayment seems
uncertain, and the banks themselves have trouble finding new borrowers and attracting
capitalization from other lenders. Lacking both additional revenue streams and capitalization to
make new loans, banks facing the unattractive possibility of sizeable loan defaults have an
incentive to perform concessional lending to weak firms as a lesser evil, especially banks with
capital ratios near the required minimum (Peek and Rosengren 2005). In the alternative scenario,
banks would have to write down the unpaid debt and accept large equity losses, which could lead
remaining creditors of the banks to withdraw their money and precipitate the banks’ failure.
Foreclosing on delinquent borrowers en masse would also be suboptimal from the banks’
perspective, because simultaneous collateral liquidations, sometimes called “fire sales,” would
severely depress the price of the collateral (Jaskowski 2015, 193).
Zombie firms may thus be a sort of epiphenomenon of significantly indebted banks or
“zombie banks,” a type of bank loosely defined as “an insolvent financial institution whose equity
capital has been wiped out so that the value of its obligations is greater than its assets” (Onaran
2012, 2).
7
Undercapitalized banks’ survival is contingent on their ability to postpone
acknowledgement of their nonperforming loans as they continue to finance borrowers, including
7
Onaran (2012, 22) writes that Kane (1987) coined the term “zombie bank.” Since Kane referred to distressed
savings and loan corporations as “zombie firms,” however, this dissertation treats Kane as the founder of the term
“zombie firm.” Zombie firms and zombie banks are conceptually similar in being distressed corporate entities that
survive because of supportive financial policies.
27
distressed firms, in the hope that they will eventually repay their debts and aid the banks’ revival.
This absence of debt enforcement begs the questions: Under what circumstances are such weak banks
politically possible, particularly when they exist on a wide scale? How can such troubled banks continue to lend to
weak and possibly nonviable borrowers when loan repayment appears dubious, and when such behavior is
considered detrimental to the health of the wider economy?
In general, banks provide an essential economic function by serving as financial
intermediaries that connect creditors and borrowers, influence the amount of currency in
circulation, and provide settlement services to businesses. Banks are also crucially important
because their failures can be contagious—one bank’s failure can lead depositors in other banks to
withdraw their deposits simultaneously, causing an extensive liquidity crisis (Amyx 2004, 28). For
these reasons, banks are usually subject to strong government supervision, and large and
prolonged debt accumulation is rare because it is usually detected early and dealt with swiftly.
In other words, zombie firms’ prevalence in large numbers over substantial periods of
time constitutes a puzzle, because governments should be on guard against heavily indebted
banks.
Orienting Hypothesis 1. Zombie firms increase when government regulators are not transparent about bank
debt after macroeconomic shocks
One possible hypothesis is that weak banks provide financial support to zombie firms
when government regulators have asymmetrical power that enables the regulators to avoid
transparent disclosure of the banks’ debt after major macroeconomic shocks. This means that the
emergence of zombie firms requires two conditions, each of which is necessary but insufficient on
its own. The first condition is the occurrence of a financial crisis which results in large amounts of
bad assets. This condition makes it hard for corporate borrowers to service obligations to lenders
28
because asset values have depreciated and leaves banks stuck with high debt levels, which
become an incentive for the banks to engage in forbearance behaviors like discounting interest
rates and evergreen lending to postpone realization of the debt and equity losses. The second
condition is that governments postpone actions to address this debt, such as the use of public
funds to recapitalize the banks or the creation of asset management companies to accelerate
corporate reform and dispose of assets. In the absence of government measures to fix indebted
banks, the banks follow their incentive to provide credit relief to distressed borrowers through
means like discounted interest payments and forbearance lending in the hope that the borrowers
will eventually recover and repay the loans, thereby rescuing the banks themselves. This process
gives birth to zombie firms.
While the first of these conditions is basically exogenous to political factors, the second is
inherently related to how governments manage and oversee the banking sector. As mentioned
above, banks’ economic function makes them critically important to society, and politicians have
a strong interest in seeing that banks operate smoothly to ensure general macroeconomic health
and stability for the public benefit. Therefore, politicians have an incentive to ensure banks run
in ways that avoid the risks associated with low capitalization, insolvency, and the potential
creation of zombie firms. However, politicians themselves may lack the time and technical
expertise necessary to monitor banks for signs of these problems, given the multiple and
competing pressures which politicians face and the practical complexities of the banking industry.
Conventional wisdom says that politicians deal with this challenge by delegating
regulation implementation to bureaucratic agencies with better resources and more complete
information about the area of regulation. Delegation grants bureaucrats some discretion over
regulatory outcomes, but bureaucratic influence is typically constrained by legislation that
delimits exactly how bureaucrats can exercise regulation (Huber and Shipan 2002). This suggests
29
that a rough balance of power exists in the division of labor between politicians and bureaucrats,
and that regulators’ authority over their policy areas is both limited and responsive to elected
officials who design the laws which bind the regulators’ activity.
In practice, the influence of regulators is not always so limited. A case in point lies in
systems where a single political party is dominant for a long time and feels secure about its future
control of government. Electoral stability can alleviate incumbent politicians’ need to strictly
control bureaucrats, since binding rules may interfere with the politicians’ flexibility and
bargaining vis-à-vis the bureaucrats (Moe 1995; de Figueiredo 2002; Muramatsu and Scheiner
2009). In such scenarios, arrangements that politicians establish to check bureaucratic authority
while they are out of power—requirements for gathering and reporting information, monitoring
agency actions, and evaluating agency decisions (Moe 1995, 138)—may be considered less
necessary, because lack of electoral competition leads politicians to believe that they will continue
to engage with bureaucrats over time, and that informal channels are thus sufficient to learn
about policy implementation. A high level of independence may in turn incentivize bureaucrats
to form policy expertise, i.e. mastery of the policy process and substantive policy details
(Gailmard and Patty 2007, 874), and provide fertile autonomy for the development of distinct
institutional agendas (Barnett and Finnemore 1999).
In these situations, politicians may be at a structural disadvantage in overseeing
regulatory matters and have difficulty obtaining a full picture about policy implementation from
bureaucrats who have little formal obligation to share information. This informational imbalance
may less consequential in good times when the regulated policy area functions normally, but it
poses a challenge when trouble occurs and bureaucrats may be even less willing to disclose
information, especially negative information that harms their reputations. In such crisis
moments, politicians may be hindered in understanding and addressing policy problems by the
30
lack of transparency, and the relatively insulated preferences of unrestrained regulators may be
ascendant in the government’s response, even if the regulators’ modus operandi was partly
responsible for creating the problem in the first place.
An initial hypothesis about the political cause of zombie firms would thus be that the
structural supremacy of weakly checked regulators impairs transparency about bank sector health
following crises, which delays political countermeasures for repairing undercapitalized banks and
allows banks to act on their incentive to buy time for their own recovery by giving credit support
to weak firms which then become zombies. This potential explanation may be novel as a theory
of zombie firms, but it is not entirely new as an argument about slow government responses to
debt crises. Amyx (2004, 33) argues that regulatory institutions populated by exclusive informal
networks may prefer to play “organizational defense” and tightly manage information about the
health of the banking sector after financial crises, rather than adopt formal and transparent
monitoring that would show the true extent of banks’ solvency. Vogel (2006, 48–50) similarly
contends that government regulatory regimes are path dependent and may perform poorly
during crises when regulators have unrestricted authority, which enables them to pursue their
normal regulatory approach which may be unworkable when all banks hold bad debt.
While these existing works highlight the key role played by insulated bureaucrats in
prolonging banking crises, they do not illuminate the political context wherein regulators hold
ascendency vis-à-vis politicians and can take their preferred passive approach to addressing weak
banks in crises in the face of politicians’ likely calls for action. These works’ absence of attention
to the status of would-be challengers is particularly noteworthy, since regulatory agencies, like
other institutions, should be viewed as dynamic sites of conflict given that their behavior entails
distributional outcomes with political significance (Streeck and Thelen 2005, 9; Mahoney and
Thelen 2010, 8). Thus, identifying the alignment of power between bank regulators and
31
politicians, especially any power asymmetry favoring the former over the latter, should provide
more complete insight into how regulators can be resilient in pursuing a lenient approach to
insolvent banks, and ultimately allow banks to propogate zombie firms.
Orienting Hypothesis 2. Zombie firms increase when existing legal frameworks obstruct corporate
bankruptcy after macroeconomic shocks
While lack of transparency about indebted banks is one potential explanation for the rise
of zombie firms, it is also possible that the prevailing bankruptcy code eases or exacerbates
distressed firms’ incentives to pursue restructuring, and thereby influences the level of zombies.
For example, the famous Chapter 11 of the US Bankruptcy Code offers indebted firms the ability
to keep control of their business operations as debtor-in-possession, a status that affords the firms
privileges like the power to cancel or reject contracts, and automatic stays against creditor
collection and litigation attempts. These features of Chapter 11 are generally thought to
encourage weak firms to reorganize at an early date before their financial conditions substantially
worsen and they are forced to sell core assets, a scenario that could ultimately cause the firms to
shut down.
Some legal systems outside the US also compel firms to file for bankruptcy when they are
insolvent or likely to default. These systems induce indebted businesses to promptly restructure,
which diminishes the possibility of their seeking credit assistance from banks as a form of life
support. Debtors in these more rigorous jurisdictions may also be liable to creditors and
criminally punished for noncompliance, which further motivates them to move toward
restructuring and initiate bankruptcy proceedings at an early stage (Kinoshita 2013, 23–24).
By contrast, some legal systems provide few reasons for firms to quickly file for
bankruptcy. For example, some bankruptcy rules may allow distressed enterprises to postpone
32
restructuring so long as they pay for their operational needs, even if they carry substantial debt
commitments. Other bankruptcy rules may also dissuade firms from reorganization, such as rules
that require the government to assign civil and criminal responsibility to managers of
corporations whose failure is believed to result from intentional damage or gross negligence, and
rules that make it difficult to lay off regular employees prior to filing for bankruptcy, which
increase the threat of dismissal among employees once bankruptcy procedures are taken and may
subsequently lower corporate performance (Kinoshita 2013, 25–26).
Though this basic comparison between bankruptcy frameworks is elementary, it suggests
how differences in legal rules may affect distressed firms’ decisions about handling their
indebtedness. In the absence of conducive or rigorous legal procedures, firms and their managers
may have less incentive and obligation to pursue bankruptcy, and find it preferable to ask
creditors for debt relief and thereby acquire zombie status. It is therefore plausible that zombie
firms are likely to emerge in settings where such bankruptcy regimes are in place in the time after
economic shocks have depressed asset values and many firms are indebted.
Orienting Hypothesis 3. Zombie firms increase when government policy secures private lending to firms with
marginal creditworthiness
While the first orienting hypothesis proposes that banks’ forbearance toward distressed
firms can be motivated by banks’ capitalization and nonperforming loan levels, it is also possible
that banks are incentivized to provide financial assistance to troubled firms because the banks are
highly confident about recouping the value of their loans. In other words, zombies may arise not
only when banks both want to and can disguise their bad debt, but also when banks underwrite
weak firms likely to have loan repayment difficulty—and subsequently request financial relief
33
from the banks and take zombie status—because the banks have solid reason to believe that the
loan value is ultimately recoverable.
Ceteris paribus, banks should be reluctant to lend to corporate borrowers with marginal
profitability for fear of loan default. Therefore, banks should generally avoid financing weak
firms. However, if extraordinary assurances give banks confidence about repayment from
borrowers with high potential for repayment trouble—assurances presenting what economists
call a “moral hazard” (Saito and Tsuruta 2014)—the banks will have an incentive to extend
credit, because the banks will believe that they can both recover their capital and profit from the
borrowers’ interest payments. An example of such an assurance would be a government policy
that gives full or near-full financial support to banks in providing credit to certain businesses,
which would shift the default risk from the banks to the government, and embolden the banks to
lend to ordinarily unappealing customers and reduce the banks’ reluctance about granting
forbearance that turns these businesses into zombies.
The budgetary costs of government financial intervention for taxpayers, as well as the
questionability of lenient credit’s effect on improving weak borrowers’ business operations,
should generally lead to low political support for such policies and make them rare, restricted in
size, and short-lived. However, it is hypothetically possible for this kind of policy to endure over
time if its effects are insensitive to its context, if its design does not facilitate updating in the face
of shifting circumstances, and if it is not hard to obstruct such updating—in short, if its operation
satisfies political conditions theoretically associated with institutional drift (Hacker, Pierson, and
Thelen 2015, 180).
Drift is a mechanism whereby decision-makers do not revise formal policies when new
developments change the social effects of the policies in ways that are known to at least some
political actors. Though conceivably there may be alternative policies to mitigate against the
34
transformation in the original policy’s effects—i.e., the change in the policy’s effects is
rectifiable—attempts to amend the original policy are not made or are stymied, leading to the
policy’s continuation (Hacker, Pierson, and Thelen 2015, 184).
Theoretically, there are several possible causes for the status quo bias of drift. On the one
hand, the interaction of decision-making institutions, such as the separation of powers and
supermajority requirements to overturn vetoes and filibusters, may foster distinct institutional
preferences or present formal obstacles conducive to stalemate. Drift may also be an outcome of
limited public access to policymaking, which enables well-organized interests to impede potential
moves toward policy alternatives by exerting influence on expert committees that develop
legislative proposals. Finally, drift can occur even when the political environment is not
antagonistic, such as in cases where there is uncertainty about the possible effects of policy
adjustment, or collective action barriers and costs to forming coalitions around reform (Hacker,
Pierson, and Thelen 2015, 187–88).
While banks can be motivated to create zombie firms when the banks need to disguise
their own debt, they may also have a separate business rationale for birthing zombie firms if
assurance policies give the banks confidence that they can profit from loans to weak firms
regardless of the firms’ potential to default and demand for financial relief associated with zombie
status. However, the extension of such policies over time should only be possible if there is
political support that defies the likely opposition from taxpayers and those who doubt the
policies’ effect on improving recipient firms’ business conditions. The mechanisms of drift
represent hypothetical ways that this support may win and preserve the policy status quo, thereby
leading to resiliently high levels of zombie firms.
35
Research Design
This dissertation uses a case study research design to explore the causal mechanisms
proposed by these three orienting hypotheses. It adopts this research strategy for two reasons.
First, because the research agenda on the political economy of zombie firms is incipient, it is
necessary to establish preliminary conjectures about zombie firms’ political origins in the absence
of extant theory. Intensive investigation through case studies is better suited to this task than
large-n cross-case approaches, since the latter require the relevant variables and outcomes to be
more refined (Mahoney 2007; Gerring 2007, 39–43).
Second, cross-national data on zombie firms is currently in short supply. In particular,
access to large-scale, firm-level datasets capable of showing the zombie firm ratio in the economy
over time—the dissertation’s outcome variable of interest—is both difficult and expensive to
acquire even for one country, let alone multiple cases. Very recently, cross-national studies on
zombie firms have appeared in working papers issued by the OECD (McGowan, Andrews, and
Millot 2017) and the Sustainable Architecture for Finance in Europe (SAFE) Research Center
(Acharya et al. 2017), suggesting that this kind of research is now becoming possible. However,
these studies appeared too late to be included in the dissertation’s analysis, and furthermore their
usage of the CHK definition of zombie firms makes them problematic from a conceptual and
comparative standpoint. Therefore, for practical reasons of current data accessibility and
conceptual consistency, the dissertation pursues a case study approach, despite the tradeoff that
this entails for external validity.
36
Case Selection
The dissertation explores its orienting hypotheses about political causes of zombie firms
across two national cases: Japan after the banking crisis in the 1990s, and China after the global
financial crisis. It takes up these two cases because economics scholarship suggests that they
possibly represent “extreme” cases with notably high values on the outcome variable—the
zombie firm ratio—relative to the univariate distribution of zombie firms, and that they are thus
suited to hypothesis generation (Gerring 2007, 89). Because cross-national studies on zombie
firms have only recently begun to uncover what this distribution looks like, it should be
emphasized that the representativeness of the mechanisms identified in these two cases is
currently hard to state with much certainty, and that these mechanisms should be further
evaluated as more comparative data becomes available.
For the Japanese case, the dissertation looks at both firms listed on the Tokyo Stock
Exchange which tend to be large-sized firms, and small and medium-sized enterprises (SME). It
looks at SME separately because of their extensive utilization of a credit guarantee system that, as
a government policy institution specifically for SME, is unavailable to large Japanese firms.
Through this system, the Japanese government has facilitated lending to SME by supplying
guarantees on loans issued to SME by private finance institutions, i.e. banks, meaning that the
system has potentially embodied the mechanism for creating zombie firms outlined in the third
orienting hypothesis. For the Chinese case, the dissertation does not make a similar distinction
between listed firms and SOE, though it does note substantial differences between SOE and
private Chinese firms.
The following discussion previews each of these cases, and explains which of the orienting
hypotheses are investigated in each case study.
37
Listed Firms in Japan, 1995-2008
The dissertation first inspects listed firms in Japan during the lost decades of the 1990s
and 2000s, the episode that first brought popular attention to zombie firms as a concept.
Conventional wisdom holds that many listed Japanese firms became zombies following the onset
of Japan’s banking crisis around 1991. This crisis was precipitated by an asset boom in the wake
of the yen’s appreciation in the 1985 Plaza Accord, and indirectly enhanced by excess savings
and financial deregulation in the 1970s and early 1980s (Hoshi and Kashyap 2001; Katz 2003b,
91–92). The eventual rupture of land values left many borrowers with large debts, particularly
the numerous firms that took out loans to acquire highly-priced assets with real estate collateral
whose market worth subsequently imploded. These collateral-based loans became the basis of a
large-scale nonperforming loan problem in Japan’s banking sector.
Evidence suggests that Japan’s primary bank regulator, the Ministry of Finance (MOF),
was slow to react to the banking crisis. Rather than take decisive action to address banks’
problems, such as by publicly writing off bad debt, calling for higher capital requirements or
encouraging banks to foreclose on debtors, MOF officials attempted to cover up the damage by
using questionable accounting methods and encouraging banks to rollover their loans to weak
borrowers, thereby disguising the reality of the banks’ losses. Meanwhile, Japanese politicians did
not immediately intervene to demand banking reform, with the consequence that the level of
nonperforming loans generally increased throughout the 1990s. Though financial leniency
toward overleveraged firms may have saved the employment of 3 or 4 million Japanese workers
(i.e. 5 to 7 percent of the labor force), it also precipitated a debt buildup that amounted to ¥42
trillion ($339 billion)—10 percent of all loans—by March 2002 (Katz 2003b, 82, 86).
Around this time, Japan implemented insolvency law reform. In April 2000, a new law
called the Civil Rehabilitation Law, which introduced a debtor-in-possession procedure, came
38
into effect. It was followed by the implementation of amendments to the Corporate
Reorganization Law in April 2003. In the wake of bankruptcy reform, insolvency filings among
large corporations quickly increased in the early 2000s (Steele 2006). Well-known listed firms that
underwent restructuring soon after the activation of the Civil Rehabilitation Law include the
supermarket chain Mycal and the department store chain Sogo (Nakata 2009).
Given these political economy factors related to listed firms in Japan during the 1990s
and 2000s, the dissertation first explores Orienting Hypothesis 1 (“Transparency”) and Orienting
Hypothesis 2 (“Bankruptcy”) with respect to zombie firms in this firm category.
SME in Japan, 1994-2014
Most studies on zombie firms in Japan’s lost decades look at listed firms (Katz 2003b;
Ahearne and Shinada 2005; Hoshi 2006; Caballero, Hoshi, and Kashyap 2008; Fukuda and
Nakamura 2011), without equivalent concern to zombies’ possible existence among non-listed
firms. The imbalance is notable for SME, a corporate category of considerable importance to
Japan in terms of percentage of firms in the economy (99.7 percent), national employment (70
percent), and contribution to GDP (50 percent) for whom issuing stock is less common. Partly to
address this bias and provide a more comprehensive picture of zombie firms’ presence in Japan’s
lost decades, the dissertation examines the zombie issue among SME.
The dissertation also emphasizes SME for a specific methodological reason. Initial studies
of zombies among SME suggest that this corporate category had a high zombie firm ratio
throughout the 2000s, especially among microenterprises with capitalization lower than ¥10
million ($100,000) (Imai 2016). This high zombie ratio is noteworthy because Japan had
consecutive years of moderately positive economic growth between 2003 and 2007—annual
national GDP growth averaged around 2 percent, though certain individual prefectures grew
39
more slowly—and nonperforming loans largely disappeared after the early 2000s, reducing
banks’ rationale to extend debt relief to weak firms to disguise bad debt. Moreover, the high
zombie ratio among SME persisted even after the realization of bankruptcy reform at the turn of
the 21
st
century, casting doubt on another potential mechanism behind the creation of zombie
firms. Furthermore, the widespread presence of troubled SME supported by government policy
poses an empirical puzzle, since prominent political science theory suggests that Japan underwent
dramatic economic restructuring beginning in the mid-1990s due to changes to its electoral
system, and that Japanese voters pushed the government to enact market liberalization that
presumably would have attenuated such weak firms (Rosenbluth and Thies 2010).
One possible cause of the enduringly high proportion of zombie firms among SME has
been Japan’s credit guarantee system, a policy institution specifically serving SME which the
government expanded in 1998 in response to Japan’s domestic credit crunch and the Asian
financial crisis. This system insured private bank loans to SME up to 100 percent of the loan
values, and continued largely unchanged throughout the 2000s, despite requiring substantial
public funding outlays to cover its annual deficits.
The dissertation examines Orienting Hypothesis 3 (“Policy Drift”) in relation to the credit
guarantee system’s impact on zombie firm creation among SME in Japan. Exploration of this
policy institution’s relationship with zombie SME might clarify why zombies have apparently
been a more substantial presence among this firm category than among listed firms throughout
the lost decades, which would add a new angle to a scholarly controversy and major public
discourse about Japan’s recent economic history. It would also indirectly suggest restraint against
the excessive zombie firm portrayals common among observers of Japanese listed firms, which is
one kind of research contribution (King, Keohane, and Verba 1994, 17).
40
Private Firms and SOE in China, 1998-2015
After the case studies of zombie firms in Japan during the lost decades, the dissertation
turns to a non-Japanese case—China after the global financial crisis. While China is typically
absent from comparative research because of concerns that it is too unique (Kennedy 2011; Tsai
2017), it has arguably been the most widely-discussed economy sustaining zombie firms in the
recent years since the crisis. It is therefore an important site for research on these firms’ political
causes, not least because it presents a methodological opportunity to investigate whether
mechanisms evident in Japan’s experience with zombie firms are not exclusive to Japan.
In recent years, domestic Chinese journalism, foreign media, international organizations
like the IMF have all brought attention to the zombie issue, and Chinese political elites including
Premier Li Keqiang and National Development and Reform Commission Vice Chairman Liu
He have announced that cleaning up zombies is a major public policy goal for the country,
especially within the SOE firm category (He and Zhu 2016; Miller 2016; Wildau 2016b; Donnan
and Mitchell 2016a; International Monetary Fund 2016b). Many Chinese observers and
academics have further noted similarities between China’s encounter with zombie firms and
Japan’s experience in the lost decades (Hu 2015; He and Zhu 2016), suggesting that the political
mechanisms which once underlay the emergence of these troubled firms in Japan might be
present in other national and temporal contexts as well.
The dissertation examines its three orienting hypotheses in China to understand the
extent to which these possible mechanisms are evident in a non-Japanese instance of zombie
firms. Because China’s status as a capitalist economy is debated, however, the dissertation also
explores an alternative orienting hypothesis for zombie firm creation related to soft budgetary
constraint (“Soft Budget”), a form of government financial support for businesses in socialist
systems described by Kornai (1992). While CHK’s seminal study on zombie firms suggested that
41
Table 1.2. Case Studies of Orienting Hypotheses
Orienting Hypothesis Case
1. Transparency Listed Firms in Japan
2. Bankruptcy Listed Firms in Japan
3. Policy Drift SME in Japan
1. Transparency
2. Bankruptcy
3. Policy Drift
Alternative Hypothesis: Soft Budget
Private Firms and SOE in China
zombies’ congestion effect is generally similar to what occurs when a socialist economy makes a
transition to a market economy (Caballero, Hoshi, and Kashyap 2008, 1972), the dissertation is
interested to know whether the political mechanisms of zombie firms in China are the same as
their earlier counterparts in Japan, or whether they are influenced by China’s socialist legacy and
thus causally distinct.
Data
Outcome Variable
The dissertation utilizes economics research with large-N quantitative data to show the
outcome variable—the zombie firm ratio—in each of its case studies. For the case study on listed
firms in Japan, it uses data analyzed by Nakamura (2017) from the Development Bank of Japan’s
Corporate Financial Databank between 1995-2008. For the case study on SME in Japan, it uses
data analyzed by Imai (2016) from the Tokyo Shoko Research database between 1999-2008, and
data analyzed by Goto and Wilbur (Forthcoming) from the Japan Ministry of Economy, Trade
and Industry’s Basic Survey of Japanese Business Structure and Activities between 1994-2014.
For the case study on private firms and SOE in China, it uses data analyzed by Nie et al. (2016b)
42
from the Listed Company Database between 1998-2015, and the Chinese Industrial Enterprise
Database between 1998-2013 (excluding 2010).
Independent Variables
The dissertation examines several independent variables related to the dissertation’s three
orienting hypotheses. For the first orienting hypothesis, the independent variables are the debt
levels in banks which are predicted to activate banks’ incentive to engage in forbearance lending,
and the structural relationship between the legislative and executive that affects how regulatory
authority is delegated and monitoring abilities are developed towards the banking sector. For the
second hypothesis, the independent variable is the nature of bankruptcy law, specifically the rules
within the law that affect distressed firms’ decisions about undergoing restructuring through
bankruptcy proceedings. For the third orienting hypothesis, the independent variables are the
presence of an assurance policy that incentivizes banks to expand lending to firms with
precarious business conditions, and political support for the policy whose resilience may be
explained by limited public access to policymaking and strong influence from key policy
stakeholders favoring the status quo.
The dissertation draws upon a mixture of secondary sources, government and
international organization reports, and semi-structured interviews to assess these multiple
independent variables and refine the causal mechanisms proposed by the orienting hypotheses.
This information was primarily collected during fifteen months of fieldwork in Japan (14 months)
and China (1 month) between September 2015 and December 2016, during which time the
author was mainly based in Tokyo, Japan, and Hangzhou, China, but widely travelled
domestically to understand potential regional features of the conditions behind zombie firms in
each country.
43
Table 1.3. Information for Case Studies of Orienting Hypotheses
Orienting Hypothesis Case Information
1. Transparency
2. Bankruptcy
Listed Firms in Japan Outcome variable: Corporate Financial
Databank, Development Bank of
Japan,1995-2008
Independent variable: Secondary sources,
interviews
3. Policy Drift SME in Japan Outcome variable: Basic Survey of Japanese
Business Structure and Activities, Japan
Ministry of Economy, Trade and
Industry, 1994-2014; Tokyo Shoko
Research database, 1999-2008
Independent variable: Secondary sources,
government and international
organization reports, interviews
1. Transparency
2. Bankruptcy
3. Policy Drift
Private Firms and
SOE in China
Outcome variable: Listed Company
Database, 1998-2015; Chinese Industrial
Enterprise Database, 1998-2013
excluding 2010
Independent variable: Secondary sources,
interviews
Because interviewing was particularly important for obtaining information about the
mechanisms behind the emergence of zombie firms, the process of how the interviews were
conducted bears elaboration. All interviews were conducted solely by the author, on the
conditions of informed consent and interview subject anonymity. For subjects who agreed to let
their interviews be recorded, the author subsequently transcribed the interview before deleting
the audio recording. Six of the 67 total interviews were transcribed by an undergraduate student
research assistant at USC, who deleted both the interview recordings which she transcribed, and
her own transcriptions after she transferred them to the author. A complete list of the author’s
44
interviews and interview attempts appears in the dissertation’s appendix. Among the 67
interviews, 41 were conducted in Japanese, 19 were conducted in Chinese, and 7 were conducted
in English.
The interviews were conducted with subjects from a variety of professional backgrounds
in Japan and China. None of the subjects was known before entry to the field. Instead,
connections to the subjects were made through direct request (“cold calling”), or through face-to-
face request or follow-up contact at or after initial interviews to obtain additional interview
subjects (“snowballing”). A common pattern was an initial cold-call interview with subject in
academia, who would then become a resource for supplying additional interview subjects in his
research area who were often practitioners such as small business owners, bank managers, and
operators in various aspects of Japan’s credit guarantee system. A related pattern occurred in
interviews with government officials. Once a senior official was interviewed, he would often
willingly introduce colleagues in the same agency as additional interview subjects.
A separate and noteworthy entrée for engaging interview subjects occurred during
fieldwork in China. Following email correspondence with a researcher on China’s legal system in
summer 2016, the author joined several large-size discussion forums on the Chinese social media
application WeChat where Chinese scholars and legal professionals discussed bankruptcy law.
Because forum members hailed from different regions around China and were accessible to
personal contact through the WeChat platform, the author was able to quickly arrange
interviews with numerous subjects—mostly lawyers knowledgeable about bankruptcy law—in
various locations in China, which was practically important due to the limited time in the
country (30 days), and the research’s interest in geographic variation in the occurrence of zombie
firms. Interviews with subjects met through the WeChat discussion forums comprised the bulk of
45
the interviews in China, though subjects were also met through cold calling and snowballing on
initial interviews with Chinese academics, as well as through colleagues’ personal introductions.
Despite the range in interview subjects’ backgrounds, the interviews themselves shared
certain commonalities. For one, the interviews generally lasted around 45 minutes, unless the
subjects requested shorter interviews, though longer interviews also happened. While interview
questions were tailored to the individual subjects’ professions and expected knowledge of
particular factors related to zombie firms, the question order followed a general pattern. Subjects
were first asked about the breadth of their professional experience to gauge their proximity and
relationship with the specific aspects of the zombie firm causal story about which they were
subsequently asked. Due to the preliminary nature of the project’s study of zombie firms, many of
these subsequent questions were probing questions, though interviews conducted at a later stage
in fieldwork also included confirmatory questions aimed at verifying responses which were
anticipated based on prior insight.
Though the interviews included a broad range of subjects including financial and legal
professionals, businessmen, bureaucrats, politicians, and economists, the limited number of
interviews compared to the immenseness of the zombie firm problem—hundreds or thousands of
firms in each case covering more than a decade in elapsed time—means that the subjects were
unrepresentative of the people related to zombie firms in Japan and China. Furthermore, the
interview arrangement techniques of cold-calling and snow-balling also involved a certain degree
of bias, since the ability to hear from the subjects was affected by non-random factors such as the
public availability of the subjects’ contact information and some subjects’ relationships with other
subjects. Nonetheless, because of the esoteric quality of certain factors behind the creation of
zombie firms, the information gathered through the interviews was considerably valuable for
clarifying the mechanisms behind these firms’ emergence.
46
The following three chapters examine the political underpinnings of zombie firms
proposed by the three orienting hypotheses, first in Japan in the lost decades, then in China after
the global financial crisis. The next chapter begins with the transparency and bankruptcy
hypotheses in the case of listed firms after Japan’s banking crisis since the mid-1990s, which is
widely considered the prototypical instance of zombie firms.
47
CHAPTER 2.
Zombie Firms among Japanese Listed Firms, 1995-2008
This chapter examines the case of zombie firms among listed firms since Japan’s banking
crisis in the 1990s.
8
Exemplified by well-known troubled firms like Yamaichi Securities and the
discount retailer Daiei (Henry 1997; Wehrfritz and Takayama 2002), this corporate class has
been the traditional focus of scholarly concern about Japan’s zombie firms since they were first
identified in the post-bubble period (Katz 2003b; Sekine, Kobayashi, and Saita 2003; Ahearne
and Shinada 2005; Hoshi 2006; Caballero, Hoshi, and Kashyap 2008). Today, many economists
and pundits still point to large zombie firms as a reason why the Japanese economy remains
sluggish two decades after the banking crisis (ANU TV 2013; Chan 2015; Prestowitz 2015; Smith
2016). The possibility that these firms continue to impede Japan’s growth underscores the
significance of understanding their political causes, not only for Japan itself, but also for the many
other countries that have become alarmed about zombie firms in the wake of the global financial
crisis and have been keen to learn from Japan’s experience.
This chapter shows that the zombie firm ratio among listed Japanese firms escalated in
the 1990s, but fell markedly after 2001. This finding calls into question conventional wisdom
about zombie firms’ persistence throughout Japan’s lost decades, and suggests that the zombie
8
In this chapter, the terms “listed firms” and “large firms” are used interchangeably. Firms listed on the Tokyo
Stock Exchange are required to have a minimum market capitalization on listing day of ¥25 billion to be on the
exchange’s first section, and ¥2 billion to be on the exchange’s second section (Japan Exchange Group 2015). By
contrast, the legal definition of small and medium enterprises (SME) in Japan is any manufacturing firm with no
more than ¥300 million in capital, or any wholesale, service and retail firm with less than ¥100 million in capital.
While there is no numerical employee requirement for firms to be listed on the TSE, it is fair to assume that firms
offering at least ¥2 billion in market capitalization have several hundred if not thousands of employees. It is with this
general size contrast to SME in mind that the chapter uses “listed firms” and “large firms” synonymously.
48
firm problem, at least among listed firms, was resolved much earlier than commonly thought.
Correspondingly, it signifies that lessons about Japan’s policy response to zombie firms are
indeed ascertainable, since events around the start of the 21
st
century clearly reduced the
prevalence of these troubled enterprises after their initial rise in the 1990s. At the same time, it
implies that Japan’s stagnation during the 2000s stems from other political economy reasons than
zombie firms, which the chapter notes as areas for future research.
The chapter proceeds in two main parts. First, it presents descriptive statistics on the
zombie firm situation among listed Japanese firms based on large-n quantitative data. These
statistics illustrate the zombie firm ratio among listed firms—the outcome variable of interest—
and certain characteristics of these firms, including their survival duration as zombies and their
comparative existence across different industries. These statistics are primarily drawn from a
recent study by Nakamura (2017), who analyzes a large sample of listed firms from the
Development Bank of Japan’s Corporate Financial Databank during the period 1995-2008.
Second, the chapter examines two political economy conditions—transparency and
bankruptcy law—which the dissertation’s orienting hypotheses propose may affect the rise and
fall of large zombie firms as evident in these statistics. On the former condition, the chapter
revisits the economic mechanism behind the creation of zombie firms analyzed by Peek and
Rosengren (2005)—the decision by undercapitalized banks to provide financial assistance to
impaired borrowers to hide the banks’ undercapitalization—and postulates that this mechanism
should be unlikely to exist in the presence of banking regulation that effectively alerts politicians
to banks’ debt levels and enables timely political intervention to fix weak banks. It then poses a
hypothetical arrangement whereby the absence of political turnover discourages such regulation,
and presents evidence showing how the historical legacy of one-party rule by the Liberal
Democratic Party (LDP) made Japan’s banking sector vulnerable to weak political supervision
49
because politicians certain of keeping power deemed it strategically unnecessary to create formal
information channels with the primary bank regulator, the Ministry of Finance (MOF). Evidence
suggests that the nonexistence of these channels delayed the response by Japanese politicians to
the banking crisis in the 1990s, and that in the absence of swift political action to address banks’
debt levels, MOF allowed undercapitalized banks to extend support to troubled borrowers in the
hope of their eventual recovery and loan repayment, thereby turning numerous firms into
zombies.
Aside from banking regulation, the chapter also explores how Japan’s bankruptcy laws
affected the zombification of listed enterprises during the banking crisis. Starting from the
assumption that bankruptcy law can promote the early recovery of troubled firms by facilitating
orderly dispensation to creditors via restructuring, the chapter investigates the orienting
hypothesis that restructuring through bankruptcy could be an alternative for helping distressed
businesses, including potential zombie firms, to avoid the prolonged indebtedness associated with
zombie status. The chapter then shows that in the case of Japan during the 1990s, bankruptcy
proceedings were onerous for both firms (debtors) and banks (creditors), and surrounded by the
threat of criminal involvement. The chapter thus finds that Japan’s bankruptcy law was
unconducive to the restructuring of insolvent firms until legal reform in the early 2000s, which
may explain why these firms were more inclined to request financial relief from their banks and
their banks were more likely to provide it and thus produce zombie firms.
Statistics on Zombie Firms among Japanese Listed Firms
Since the conceptual reappraisal of the CHK zombie firm definition by FN (2011),
scholarship has generally coalesced around the FN definition of zombie firms as businesses that
50
Figure 2.1. Zombie Firm Ratio among Japanese Listed Firms, FN Definition, 1988-2008
Source: Nakamura (2017, 12)
are unprofitable and receive financial relief in the forms of discounted interest rates and
evergreen lending. This is principally because the more strenuous FN definition eliminates the
possibilities of misidentifying healthy businesses that receive below-prime interest rates as
zombies, and misidentifying troubled firms that use evergreen lending as healthy companies
(Kwon, Narita, and Narita 2015; Imai 2016; Goto and Wilbur Forthcoming).
A recent study that uses the FN definition to examine zombie firms among listed firms in
Japan is by Nakamura (2017), who analyzes a large sample of listed firms from the Development
Bank of Japan’s Corporate Financial Databank in the period 1995-2008. Its results contrast with
prevailing wisdom about the prevalence of zombie firms in Japan’s lost decades. Most
importantly, the study finds that, within its firm sample, the zombie ratio surges following the
outbreak of the crisis in the early 1990s, peaks in 2001 at approximately 17 percent, and quickly
declines in the next two years before stabilizing at roughly 5 percent of listed firms between 2003
and 2007 (Figure 2.1.).
2.2.2 Features of the Modified Criteria and Exclusion
of “One Shot Zombie Firms”
Similar to CHK, firms that are listed or used to be listed on the Tokyo Stock
Exchange (TSE) excluding TSE Mothers (market for emerging companies) and
whose primary business belongs to the manufacturing, construction, real estate,
retail, wholesale (excluding nine general trading companies), and service industries
constitute the sample universe for this study. The sample period ranges from 1995
to 2008. The firm-level non-consolidated financial data are collected from the
Corporate Financial Databank, compiled by the Development Bank of Japan.
Zombie firms are identified every year based on the aforementioned modified
version of CHK’s criterion. In short, the dataset used in this chapter is the same as
that of Nakamura and Fukuda (2013).
Figure 2.2 depicts how the ratio of zombie firms changed based on these
modified criteria from 1995 to 2008. From 1995 to 2001, the modified zombie ratio
exhibits similar features to that of CHK, although it is always far smaller.
Regardless of the criteria, the zombie ratio increases substantially in the late 1990s
under the prolonged recession and banking crisis. On the contrary, after 2002, it
declines dramatically from 2002 to 2004 and remains low until 2007 corresponding
tothedeclineinthenon-performingandsub-performingloanratiosduringthesame
period, while that of CHK keeps increasing and hovers at nearly 40 % as shown in
0%
5%
10%
15%
20%
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Sum of non-performing and sub-performing loan ratio of major banks
Non-performing loan ratio of major banks
Zombie ratio based on the modified criteria
Fig. 2.2 Zombie ratio based on the modified criteria. Notes 1. “Modified criteria” refers to the
methodadoptedinthisbooktoidentifyzombiefirms,whichintroducedthe “profitabilitycriterion”
and “evergreen lending criterion” as additional criteria to CHK’s criterion. 2. “Sub-performing”
refers to loans classified as “need attention” but not classified as “special attention” by the
self-assessment of assets by banks
12 2 Evolution and Recovery of Zombie Firms: Japan’s Experience
51
Figure 2.2. Zombie Firm Ratio among Japanese Listed Firms, CHK Definition, 1988-2008
Source: Nakamura (2017, 11)
By comparison, an extension of the conceptually problematic CHK definition shows a
zombie ratio that is both considerably higher in absolute terms throughout the 1990s, and still
rising in the 2000s to cover more than one-third of listed Japanese firms (Figure 2.2.). In other
words, Nakamura’s study shows that the magnitude of zombie listed firms is less than previously
thought, and that, among this corporate category, zombies ceased to be a major presence by the
early 21
st
century. The accuracy of the FN definition’s estimate is also corroborated by the
nonperforming and sub-performing loan ratios at major Japanese banks, whose change over time
between 1998 (when Japan’s Financial Supervisory Agency was created and began to record
data) and 2007 generally parallels the trajectory of the FN-estimated zombie ratio.
In addition to uncovering a more moderate zombie firm ratio among listed firms in the
1990s and a substantial decline in the ratio after 2001, Nakamura makes several discoveries
about the characteristics of listed firms with zombie status. For one, of the roughly 2,300 zombie
firms in the sample between 1995 and 2008, 47.6 percent recover from zombie status in the next
CHK’s criterion. One is the “profitability criterion.” Under this criterion, firms
whose earnings before interest and taxes (EBIT) exceed the hypothetical risk-free
interest payments are excluded from zombies.
4
Low leveraged firms whose total
external debt is less than one-fifth of their total assets from zombies are also
excluded. Healthyfirms are unlikely to have negative pre-tax profits after deducting
non-operating income, while unhealthy firms are unlikely to be low leveraged.
Therefore, excluding these firms from zombies also reduces the likelihood that the
modified criteria misidentify healthy firms as zombies.
The second criterion is the “evergreen lending criterion.” Under this criterion,
unprofitable and highly leveraged firms with increasing external borrowings are
included in zombies. Specifically, the evergreen lending criterion categorizes as
zombiesfirms whose EBIT is less than the hypothetical risk-free interest payments,
whose total external debt exceeds one-fifth of their total assets, and whose bor-
rowings have increased from the previous year. Firms with negative pre-tax profits
and large external debt are rather unlikely to take out a fresh loan. Therefore, by
categorizing such firms as zombies, the modified criteria are less likely to
misidentify unhealthy firms as non-zombies.
0%
5%
10%
15%
20%
25%
30%
35%
40%
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Sum of non-performing and sub-performing loan ratio of major banks
Non-performing loan ratio of major banks
Zombie ratio based on CHK's criterion
Fig. 2.1 Zombie ratio based on CHK’s criterion. Notes 1. Zombie ratio based on CHK’s criterion
refers to the result of the reproductive calculation applying the criterion of Caballero et al. (2008)
to the dataset of this book. 2. “Sub-performing” refers to loans classified as “need attention” but
not classified as “special attention” by the self-assessment of assets by banks
4
Sinceinterest paymentsaresmallerthanthe hypotheticalrisk-free interestpaymentsunderCHK’s
criterion,onlythosefirmsthathadpositivepre-taxprofitswereexcludedfrombeingcategorizedas
zombie firms.
2.2 Identification of Zombie Firms 11
52
Figure 2.3. Number of Zombie Firms by Measure of Financial Support (Excluding One-Shot
Zombie Firms), 1995-2008
Source: Nakamura (2017, 15)
9
fiscal year, 50.4 keep zombie status, and 2.0 percent delist.
10
Thus, on average, zombie listed
firms recover in slightly over two years (the peak average duration of zombie firms is 2.87 years in
1996 and declines thereafter), meaning that zombie firms among this corporate category typically
undergo zombie status for only a short time (Nakamura 2017, 15).
Additionally, roughly half of the zombie firms among listed firms receive evergreen
lending, whereas zombie firms that take either interest rate relief or both forms of financial
support respectively account for 31 and 20 percent of zombies (Figure 2.3.). Furthermore, for
zombie firms whose zombie status lasts for only one fiscal year, 59 percent receive only evergreen
lending and 24 percent receive only interest relief, while for zombies whose duration is more than
eight years, 76 percent obtain interest rate relief and only 9 percent receive evergreen lending
(Nakamura 2017, 15–17). In short, evergreen lending appears to be an essential aspect of the
9
This figure excludes zombie firms that Nakamura calls “one-shot zombies,” firms identified as a zombie only once
in the sample period from 1995 to 2008.
10
The “delist” category includes circumstances such as liquidation, court-guided rehabilitation, and corporate
buyout. The Corporate Financial Databank does not specify which outcome happens to firms in this category.
“delisted” category includes various situations, both negative and positive, such as
liquidation, court-guided rehabilitation, and buyout. From 1995 to 2008, there are
around 2300 zombie samples, 47.6 % of which recover in the next period, 50.4 %
retain zombie status, and 2.0 % are delisted. In other words, on average, zombie
firms recovered in a little over 2 years, probably a far shorter period than generally
thought. Indeed, as shown in Table 2.1, the average duration of zombie firms
reached its peak in 1996 at 2.87 years and persistently declined thereafter.
For the measure of financial support to zombie firms, shown in Fig. 2.5, the
category “evergreen lending only” accounts for nearly half of the total through
1995–2008,while“interestratereliefonly”accountsfor31 %and“both”for20 %.
0
50
100
150
200
250
300
350
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Both
Evergreen lending only
Interest relief only
Fig. 2.5 Number of zombie firms (excluding one-shot zombie firms) by measure of financial
support
Table 2.1 Average duration of zombie firms by industry (in years)
All industry Manufacturing (a) Non-manufacturing (b) Difference (b − a)
1995 2.47 2.38 2.69 0.31
1996 2.87 2.73 3.15 0.42
1997 2.61 2.47 2.87 0.40
1998 2.16 1.89 2.86 0.97
1999 2.33 2.21 2.69 0.48
2000 2.13 2.01 2.37 0.35
2001 1.87 1.75 2.26 0.50
2002 2.02 1.93 2.27 0.34
2003 1.92 1.98 1.83 −0.15
2004 1.93 1.98 1.88 −0.10
2005 1.79 1.70 1.94 0.24
2006 1.81 1.80 1.84 0.04
2007 1.71 1.80 1.57 −0.23
2008 1.54 1.52 1.63 0.11
2.3 Evolution and Recovery of Zombie Firms … 15
53
Figure 2.4. Number of Zombie Firms among Japanese Listed Firms by Industry, 1995-2008
Source: Nakamura (2017, 28)
zombie firm problem as alluded by earlier research (Sekine, Kobayashi, and Saita 2003; Fukuda
and Nakamura 2011; Nakamura and Fukuda 2013), but it should be viewed primarily as a
temporary assistance measure for distressed businesses needing to overcome momentary cash
shortages, rather than as perpetual life support for hopelessly weak firms.
Zombie firms among listed firms also manifest in distinct ways in different industries.
Manufacturing firms constitute the numerical majority of zombie listed firms (Figure 2.4.) and
generally have a higher zombie ratio than nonmanufacturing firms (Figure 2.5.). However,
zombie firms and nonperforming loan problems are most highly concentrated in the
construction, real estate, wholesale and retail sectors. Manufacturing firms also likely suffer from
short-term liquidity crunches because of the unstable business environment in the sample period,
which makes them suited to evergreening as a stopgap to maintain operations, and recover from
zombie status because of the positive external environment between 1999 and 2006 due to
expanding demand from growing overseas economies like China. On the other hand, distressed
effective than innovation for the recovery of zombiefirms in the second sub-period.
Although the improved macroeconomic environment had a positive impact, zombie
firms in the second sub-period did not have sufficient time or resources to expand
their own businesses through innovation. In such circumstances, they were trapped
in cost competition without sales growth.
In addition, for shareholder composition, the coefficient of Financial institu-
tions’ ownership takes a significantly positive sign in the second sub-period,
inconsistent with the widely accepted view that bank ownership delayed the
recovery of zombie firms. Unlike the first sub-period, bank ownership declined
substantially due to the regulatory reform mentioned in Sect. 2.4.2. Moreover, after
the reform of corporate governance rules as well as accounting rules and bank
supervision policies, banks could not help being sensitive to the impairment of their
assets. In these circumstances, if banks kept holding the stocks of a troubled firm,
they must have a strong incentive to discipline the troubled firm for the recovery.
By contrast, there is no evidence that Foreign ownership promoted the recovery,
even after the Koizumi–Takenaka reform.
2.8 Estimation of the Sub-periods by Industry
The issues of zombiefirms and the non-performing loan problem were most serious
in the construction, real estate, and wholesale/retail industries. Indeed, most
well-knowntroubledfirmsthatendangeredbanks’healthassociatedwithlarge-scale
debt reorganization belonged to these industries. However, with respect to the
numberofzombiefirms,manufacturingfirmsaccountedforthemajoritythroughout
thesampleperiod,asshownin Fig. 2.6.Atthesametime,wefound threeimportant
0
50
100
150
200
250
300
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Non-manufacturing
Manufacturing
Fig. 2.6 Number of zombie firms (excluding one-shot zombie firms) by industry
28 2 Evolution and Recovery of Zombie Firms: Japan’s Experience
54
Figure 2.5. Zombie Firm Ratio by Industry among Japanese Listed Firms (Excluding One-Shot
Zombie Firms), 1995-2008
Source: Nakamura (2017, 29)
firms in nonmanufacturing industries typically keep zombie status for longer, especially before
2001, because their low profitability is less influenced by foreign demand. This makes
nonmanufacturing firms more likely to be supported by interest rate relief, since extended
evergreening without improved performance can exacerbate firms’ debt problems. The average
duration of zombie status for manufacturing firms is 1.89 years, while the average duration for
zombie nonmanufacturing firms is 2.86 years (Nakamura 2017, 25, 28–29).
Nakamura’s study also presents findings about zombie nonmanufacturing firms’
corporate governance and operational restructuring characteristics. In the escalation period until
2001, bank ownership and cross-shareholding have significantly positive relationships with
zombie formation, suggesting that these two aspects of Japan’s corporate system once thought to
be beneficial for stability and long-term growth actually shielded ineffective managers from
market pressure after the bubble’s implosion (Nakamura 2017, 19). However, from 2001 onward,
there is a reverse relationship between bank ownership and zombie nonmanufacturing firms,
differencesbetweenmanufacturingandnon-manufacturingintermsofthedynamics
of zombie firms and their choice offinancial support measures. Firstly, for the time
series variation in the alleged zombie ratios shown in Fig. 2.7, the ratio of manu-
facturing has peaks in 1998 and 2001 butfluctuates widely along with the change in
the macroeconomic environment, while that of non-manufacturing reached its peak
in 1998 and declined slowly but persistently thereafter. Secondly, as shown in
Table 2.1, troubled firms in non-manufacturing industries tended to stay in the
zombie status longer than those of manufacturing industries during the first
sub-period, namely before the Koizumi–Takenaka reform. For example, in 1998
when the Japanese banking crisis reached its climax, the discrepancy between
manufacturing andnon-manufacturingfor the average duration ofzombiefirmsalso
peaked (0.97 years), where the average duration for manufacturing zombie firms
was 1.89 years compared with 2.86 years for non-manufacturing zombie firms.
Interestingly, this discrepancy almost disappeared in the second sub-period largely
because ofthereductionin duration in thenon-manufacturingsector.Thirdly,partly
as a consequence of thefirst and second differences, zombiefirms in manufacturing
industries relied more heavily upon evergreen lending and less upon interest relief
than those in non-manufacturing industries. In a broad sense, manufacturing firms
face a more volatile business environment than non-manufacturing firms and thus
tendtosufferfromcashshortages.Hence,evergreenlendingisasuitableremedyfor
such volatile firms because they can quickly recover as soon as the adverse envi-
ronment ends. By contrast, troubled firms in non-manufacturing industries tend to
face low profitability and thus take more time to recover. Hence, interest relief is
suitable for such firms, while evergreen lending without improvement in interest
coverage could even worsen the debt problem.
0%
5%
10%
15%
20%
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Manufacturing Non-manufacturing
Fig. 2.7 Zombie ratio (excluding one-shot zombie firms) by industry
2.8 Estimation of the Sub-periods by Industry 29
55
with bank ownership having a significantly positive association with zombie recovery. This
finding implies that banks became sensitive to the potential damage of their assets in the latter
period, and had strong incentives to steer distressed firms toward improvement.
11
In sum, Nakamura’s study indicates that Japan’s experience with zombie firms after the
burst of the bubble economy, at least among listed firms, was less overwhelming than commonly
thought. Not only were zombies among this firm category a small absolute presence in the 1990s
in contrast to the original CHK estimate, but their existence as a percentage of all listed firms
also greatly declined from 2001, meaning that listed zombies were not widespread by the start of
the second lost decade. Moreover, a sizeable portion of zombie listed firms were sustained by
temporary evergreen lending that plugged cash flow holes created by macroeconomic instability
and exited from zombie status after one fiscal year. In short, the scale of Japan’s zombie firm
problem among listed firms was relatively minor, and to a certain degree zombies were an
unavoidable outcome of the broader economic context. In many cases, they quickly improved
and stopped being zombies.
While this more moderate picture calls into question the economic damage wreaked by
zombie firms in Japan’s lost decades, especially after the early 2000s, it does not negate that some
financial support to listed firms also occurred with the purpose of hiding banks’ nonperforming
loans during the banking crisis (Nakamura 2017, 17). Although Nakamura and Uesugi (2008, 45)
assert that the health of Japan’s financial institutions did not have an outsized influence on the
overall generation of zombie firms, Peek and Rosengreen (2005) and Fukuda, Kasuya and
Nakajima (2006) show that troubled banks with reported capital ratios close to their required
ratios—which banks often manipulated to understate their nonperforming loans prior to
11
Bank ownership fell considerably after the introduction of regulation in 1998 that established a maximum of 5
percent for each bank’s share in terms of voting rights (Nakamura 2017, 19).
56
regulatory reform in the late 1990s (Peek and Rosengren 2005, 1165)—and banks with
deteriorating nonperforming loan ratios did in fact increase lending to weak firms. In other
words, the lending behavior that created zombies among listed firms may have largely been an
outcome of exogenous economic circumstances, but it was also influenced to a degree by the
banks’ poor capital conditions.
The fact that the health of Japanese banks contributed to the rise of zombie firms begs the
question: How were banks able to continue holding so much debt when its size and quality made the banks more
likely to practice forbearance toward weak borrowers and thereby turn them into zombies? Though the asset
price bubble’s collapse in 1990-91 quickly led to distressed businesses and banks, Japan’s banks
could have been required to realize their losses and sell their assets, and thereby spark
Schumpeterian creative destruction and clear a path for economic recovery (Peek and Rosengren
2005, 1145). As Amyx (2004) notes, governments are ordinarily keen to prevent banks’
accumulation of “bad debt,” and quick to enact policies to address such debt when it occurs,
because uncertainty that banks may have to write off substantial loan losses and lose capital can
incite depositors to withdraw their holdings and lead to bank runs that are economically and
socially destabilizing. Therefore, Japan’s delay in confronting its banking sector’s health—and its
resulting effect on banks’ forbearance behavior which created zombie firms—constitutes an
empirical puzzle.
While tolerance toward banks’ low capitalization levels was one factor in the formation of
zombie firms among listed firms in Japan during the 1990s, the decision by troubled firms to
request forbearance and thus assume zombie status might also be considered a questionable
choice from the firms’ own perspective of preserving corporate value, especially for deeply
impaired firms with dubious odds of recovery. In principle, before distressed businesses move to
take financial relief from banks, they may also try to tackle their debt through restructuring to
57
stave off possible risky decisions in the future that could exacerbate their business conditions.
Legal arrangements that facilitate debt resolution via bankruptcy, such as Chapter 11 in the
United States Bankruptcy code, have been shown to encourage indebted firms to promptly file
for bankruptcy and thereby avoid the potential danger of further financial deterioration
(Berkovitch and Israel 1998). Therefore, the incentive on the part of Japanese listed firms to
undergo their own zombification, particularly heavily indebted firms, is a second puzzle related
to listed firms in Japan.
Aside from the murkiness that enshrouded the troubling debt levels at Japanese banks for
years after the banking crisis and the strangely unenticing nature of Japan’s bankruptcy law for
distressed firms after the crash of the bubble, one other potential explanation for the increase in
zombie listed firms might have been an assurance policy that produced a moral hazard in banks’
lending decisions toward distressed firms, as outlined by the third orienting hypothesis. However,
such an explanation would be unlikely for listed firms for the reason that listed firms have
shareholders who can be expected to have little tolerance for the sort of chronically weak
corporate performance associated with the zombie firms in the third hypothesis. Moreover, the
high visibility of large listed firms should make any costly assurance policy unpopular with the
public and politically difficult to sustain over time.
The following analysis thus attempts to explain the zombie firm phenomenon among
listed firms by exploring the orienting hypotheses of transparency and bankruptcy law.
Transparency
Economics research has established that banks with weak balance sheets have a perverse
incentive to lend to poorly performing listed firms and thus give birth to zombies. Though many
58
of Japan’s zombie listed firms were produced by the adverse economic circumstances in the
fallout of the bubble’s collapse, the fact that the condition of Japanese banks also played a certain
role in zombie formation begs the question of why the sector was allowed to fester in an
unhealthy state and its nonperforming and sub-performing loans were left unaddressed for so
long after the crisis struck. The delay in dealing with troubled banks was a major reason why
Japanese banks persisted in credit misallocation for years after the crisis, and contributed to
Japan’s protracted experience with zombies among listed firms.
Blame for the slow response to the crisis has traditionally fallen at the door of the Ministry
of Finance (MOF), the chief government regulator and supervisor of Japan’s banking sector until
1998 (Mabuchi 1997; Hartcher 1998; Amyx 2004; Vogel 2006; Hoshi and Kashyap 2010; Lipscy
and Takinami 2013). Explanations for MOF’s indecisiveness vary slightly—Amyx (2004) argues
that MOF wanted to manage damaging information about weak banks that would reveal
ministerial incompetence, while Vogel (2006) states that MOF’s regulatory approach had become
path dependent in the postwar period and struggled to adjust to the sudden debt surge in the
early 1990s—but the general consensus is that MOF had primary responsibility for managing the
crisis, yet took no strong action for years. Instead, the ministry concealed the magnitude of
Japan’s nonperforming loan problem and encouraged banks to sweep such loans “under the rug”
(Vogel 1996, 194; Amyx 2004, 150–51). As the foregoing discussion suggests, the persistence of
this substantial debt at Japanese banks was an important reason, if not the only reason, why
zombie listed firms appeared in Japan during the 1990s.
Although factors like damage control and ingrained practices may have contributed to
MOF’s unhelpful behavior following the burst of the bubble in 1990-1991, they alone do not
provide clues for why the ministry was able to remain ascendant in Japan’s response to the crisis
for so long. As Amyx (2004, 29) emphasizes, governments around the world generally prioritize
59
the early detection of bank solvency problems to avoid possible bank runs, and national
legislatures tightly monitor government bank regulators for precisely this reason. Thus, a logical
place to look for how and why MOF remained influential in shielding banks with low
capitalization—and thus steered the banks toward zombie firm creation—is the relationship
between the ministry and elected representatives, namely, members of Japan’s national
parliament (an institution commonly known as the Diet). The alignment of power between these
two sets of actors could reveal how Japanese national politicians were able—or unable—to
execute their roles as the ultimate overseers of Japan’s banking sector, and direct MOF to fix the
sector’s troubles once the crisis hit.
In the political science literature on institutions, the relationship between the legislative
and executive branches of government is typically characterized by delegation, meaning that
politicians with time constraints and limited technical expertise grant bureaucrats a certain
amount of influence to formulate and implement policy according to the politicians’ objectives
(Krause 2010, 524). While the degree of policymaking discretion bestowed on bureaucrats is
sometimes deliberately determined by politicians’ intentions for achieving certain policy goals, it
can also be a function of political and institutional factors (Huber and Shipan 2002, 9). For
example, in parliamentary systems, majority coalition and minority coalition governments often
exercise tight control over bureaucrats, because politicians from opposing parties worry about the
bureaucracy’s ability to enact their favored policies. As Huber and Shipan (2002, 185) argue, “All
else equal, efforts to place policy details in laws should increase whenever there is conflict among
the parties that together constitute the government’s majority.” Moreover, in such settings where
incumbents may be unseated by electoral competition, politicians should be inclined to establish
binding rules that protect their preferred positions against change or destruction. In the words of
Moe (1995, 124), politicians “know that whatever structures they put in place today may be
60
subject to the authoritative direction of others tomorrow […]. If today’s authoritative decisions
are to have staying power and continue generating benefits for their creators into the future, they
must somehow be insulated from tomorrow’s exercise of authority.”
By contrast, politicians in single-party majority governments may put fewer constraints on
bureaucrats through detailed legislation that limits the latter’s policymaking discretion. This may
be especially true in situations where politicians face little risk of falling out of power for a long
time, since, in such conditions, bureaucrats and politicians from the dominant party will be
cognizant that the politicians can punish the bureaucrats in the future, which leads both sides to
establish good working relations in which bargaining and flexibility are advantages rather than
risks (Muramatsu and Scheiner 2009). In these situations, the imposition of binding structures
would constrain politicians as much as bureaucrats, so politicians may bestow greater leeway on
the executive to make and implement policy, without feeling pressure to develop their own
independent legislative capacity and monitoring ability towards the bureaucrats’ performance.
Weak political oversight of bureaucrats’ policymaking may not be especially problematic
if the policy areas under the bureaucrats’ watch are operating normally. However, in the event of
policy trouble, politicians without established mechanisms for gathering information may be
hindered from having a clear picture of the situation, while bureaucrats may be unwilling to
disclose facts that reveal their mismanagement and damage their reputations. In such moments
without transparency, politicians may be at a disadvantage in trying to fix the issue, because they
do not comprehend its full nature. Consequently, bureaucratic preferences may reign supreme in
the government’s response, even if bureaucrats were originally responsible for the problem.
Evidence suggests that Japan had this sort of hypothetical relationship between its
legislative and executive government branches in much of its postwar history, and that a legacy
of weak political oversight of MOF hobbled politicians and enabled ministry bureaucrats to
61
pursue their preferred regulatory approach following the asset bubble collapse in the early 1990s.
The ministry’s approach allowed distressed banks to conceal their nonperforming loans, thereby
enabling banks to act on their incentive to buy time for their own recovery by sustaining weak
firms, many of which subsequently became zombies. The key structural reason was the long-term
dominance of the Liberal Democratic Party (LDP), the political party that held a majority in the
important lower house of the Diet for 38 straight years between 1955 and 1993.
Historically, Japan’s bureaucracy commanded wide jurisdiction to promote the county’s
modernization drive to catch up with western countries from the late 19
th
century (Muramatsu
1997, 20). Although bureaucrats were heavily involved in state policymaking during World War
II, their influence on policy in the postwar period was preserved by Allied occupation authorities
whose purges concentrated mainly on wartime conservative politicians (Saunavaara 2009, 11).
Meanwhile, the LDP emerged as Japan’s single dominant political party from 1955 due to a
combination of factors that helped the party consistently ward off challengers at the national
level, including a multimember nontransferable vote electoral system, clientelism, fiscal
centralization, and a rural bias in the electoral system (Krauss and Pekkanen 2010, 6; Scheiner
2006, 3–5). Accordingly, postwar Japan policymaking became characterized by cooperation
between bureaucrats and the LDP, with LDP politicians feeling secure about their ability to win
elections and stay in power, and thus confident about having a continuing relationship with
bureaucrats (Campbell and Scheiner 2008, 94).
12
12
Campbell and Scheiner (2008) argue that postwar bureaucrat-politician relations were cooperative, which is a
nuanced position situated between other readings of Japanese politics that assert one side’s power over the other.
More extreme viewpoints include Johnson (1982, 154), who famously depicted the bureaucracy’s dominance with
the dictum “the politicians reign, and the bureaucrats rule,” and Ramseyer and Rosenbluth (1993), who saw a
principal-agent relationship in which politicians were the principals and bureaucrats were the agents.
62
As expected by the orienting hypothesis, a corollary of the stable long-term relationship
between bureaucrats and successive LDP governments was that Japanese politicians generally did
not develop strong legislative capacity on their own. Because bureaucrats typically contributed
policymaking assistance, politicians felt little need to acquire substantial amounts of staff and
resources to independently draft legislation.
13
This state of affairs left politicians dependent on
bureaucratic expertise, and caused them to hand over substantial authority to government
ministries (Muramatsu and Scheiner 2009).
Evidence indicates that this pattern of delegation held in the realm of banking regulation.
As Vogel (1996, 172) describes, throughout the postwar period, “LDP Diet members were
normally happy to leave finance policy up to [MOF]. Most Diet members did not understand the
details of financial regulation and lacked the staff to support them in this area. Furthermore, they
had relatively little interest in finance politically, for although banks and securities houses were
important donors, they counted for very few votes.” Politicians’ incentives to participate in
banking regulation were also diluted by the dispersion of bank branches throughout the country,
and the fact that regional banks’ territory typically covered multiple electoral districts. Politicians
were generally only concerned with MOF activities insofar as they touched on fiscal policy,
which carried the possibility of specific distributive benefits for politicians’ home districts (Amyx
2004, 56–60).
Because Japanese politicians lacked the support staff, technical knowledge, and incentives
to engage in financial policymaking, MOF enjoyed substantial discretion to make and execute
banking regulation in postwar Japan. Since MOF desired above all a stable financial system to
promote industrial growth, it prioritized the goal of minimization of bank failure, which it
13
For example, Japanese Diet members each had two publically-funded secretaries until 1993, when it became
possible to hire a third secretary.
63
considered both economically damaging and harmful to the credibility of the ministry and
Japan’s financial system. Toward this end, MOF officials tightly monitored local banks for signs
of weakness, advised struggling banks, dispatched retiring ministry employees to assume
management of these banks, and, in extreme cases, directed mergers between problem banks and
healthier banks. These regulatory actions relied on close relationships with banks, and required
information about the banks’ management and financial health—information that the banks
were willing to provide because they trusted MOF, and because the ministry oversaw the permits
and licenses which the banks themselves needed to operate in the industry. Banks also depended
on the ministry because of its tendency to flexibly administer the laws according to which the
banks operated (Vogel 1996, 170).
Over time, this style of policy guidance and regulation produced close informal links
between MOF and Japanese banks. Meanwhile, though the ministry’s shepherding of the
banking sector needed information to operate smoothly, few formal informational requirements
were established prior to the 1980s because capital scarcity and the tight insulation of Japan’s
financial market mitigated against risky lending behavior (Amyx 2004, 31). The ministry did at
one point try to reform its regulatory practices in the early 1980s, after the increasing issuance of
government bonds in the late 1970s and the growing internationalization of financial markets
induced liberalization in Japan’s financial system. Under liberalization, the trend of large listed
firms increasingly moving their funding from banks to capital markets put downward pressure on
banks’ profits, and pushed banks to increase real estate lending and commercial lending to
smaller businesses to make up the difference (Fukao 2007, 275). In this context, in 1981, MOF
tried to expand banks’ formal disclosure requirements. However, the attempt was rejected by the
LDP, leaving the ministry to continue its reliance on informal relationships with banks for
information (Amyx 2004, 32).
64
As informal banking regulation remained the norm in the 1980s, bank increasingly
engaged in real estate lending, even as they often disregarded the cash holdings of borrowers.
The reason was that real estate loans were usually secured by collateral in the form of land,
which had historically enjoyed an upward price trend in the postwar era, and had consistently
higher price inflation than the government bond interest rate. These perceived advantages of real
estate collateral gave banks a sense of safety that they would never encounter loan losses, and led
them to increasingly lend to real estate and financing companies to make up for their declining
loan business with manufacturing firms in the context of the decade’s financial liberalization
(Fukao 2007, 275). At the same time, this lending trend also contributed to an asset bubble in
which the banks became deeply involved and eventually vulnerable.
By the late 1980s, both MOF and the Bank of Japan (BOJ) were concerned about the
high concentration of bank loans in Japan’s real estate, construction, wholesale, and retail sectors.
In April 1990, the ministry attempted to slow this development by enacting lending restrictions
on real estate lending, which coincided with a series of BOJ moves to raise the official discount
rate that began one year earlier. However, successive rate rises between May 1989 and August
1990 substantially diminished stock and real estate prices, which continued to plummet through
the first half of 1992. In August 1992, the Nikkei stock index in the TSE closed below 15,000
points, a drop of more than 60 percent from its peak in the bubble era. Furthermore, the total
amount of banks’ nonperforming loans reached $26 billion, and the severity of this amount for
banks was compounded by the stock market collapse, which made it impossible for the banks to
write off nonperforming loans by selling stock and realizing capital gains (Amyx 2004, 149–50).
The massive amount of debt in Japan’s banking system increased the possibility of bank
failures in the early 1990s. As mentioned above, such precarious circumstances should ordinarily
trigger intervention by politicians to restore banking sector health to avert bank runs. However,
65
in line with the hypothesized expectation that prolonged single party dominance leads to a
significant level of bureaucratic discretion and weak political monitoring, Japanese politicians
were unable to take decisive action because they were generally ill-informed about the sector’s
troubles. Instead, MOF remained in charge of government’s response to the crisis, and applied
countermeasures that both obfuscated the ministry’s own accountability for the situation, and
encouraged distressed banks act on their incentive to save themselves by camouflaging their debt
through financial relief to weak customers. This latter effort was one reason why zombies
emerged among listed firms in Japan’s first lost decade.
Several specific MOF actions contributed to banks’ creation of zombie firms once the
crisis broke. For one, in 1991, the ministry began working through its informal networks to
permit banks to falsify their accounts and adjust their accounting procedures. The next year,
MOF also enabled banks to postpone the announcement of stock portfolio losses for several
months, and to transfer bad loans to dummy corporations. Ministry officials reportedly
considered these forms of debt concealment as acceptable short-term measures, because they
believed that a positive economic climate in the imminent future would revive the value of the
banks’ investments. Importantly, however, MOF never ordered the banks to liquidate their debt
(Amyx 2004, 150–51). This meant that while MOF recognized the seriousness of the banks’
situation for their continued ability to operate, it wanted the banks to maintain their lending
relationships with delinquent businesses on the assumption that the latter would eventually
recover and resuscitate the banks’ health, rather than have the banks acknowledge and write off
their debt, lose equity, and possibly collapse. The ministry’s approach, combined with the reality
that weakened banks had difficulty finding healthy new borrowers and drawing capitalization
from other lenders in Japan’s adverse economic environment, heightened the banks’ inclination
to practice forbearance and thereby generate zombie firms.
66
Somewhat counter to the orienting hypothesis, there is indication that Japanese
politicians were not entirely unaware of the worsened conditions at the country’s banks. For
example, Prime Minister Miyazawa Kiichi hinted at the possibility of using public funds to
address the nonperforming loan problem as early as August 1992 (Amyx 2004, 159). A former
MOF director also strongly disputed that politicians were totally ignorant of the crisis at the time
(Interview 22). However, the LDP appeared generally preoccupied by its declining electoral
prospects in the face of several recent corruption scandals including the Sagawa Kyubin scandal
involving a prominent party leader and some 130 LDP Diet members, and focused on the issue
of political reform. As a result, politicians largely failed to appreciate the systemic nature of the
nonperforming loan problem in its early years, and still basically trusted MOF even after
suspicious accounting practices at Nippon Credit Bank were reported in the media in 1993
(Amyx 2004, 158–60). As Grimes (2001, 160) suggests, politicians might have privately
questioned the ministry’s policy, but they chose not to publically defy it, given their own political
troubles and their limited independent means to verify the ministry’s appraisal of the situation.
When the LDP ultimately lost the general election in July 1993 and gave up government
control for the first time in 38 years, informal relations between MOF and Japanese finance
institutions became even more important in the response to the banking crisis. The LDP’s
unprecedented exit and the introduction of political uncertainty caused banks to converge their
lobbying efforts on MOF, deepening the ministry’s insight into the woes of the sector and further
keeping politicians from information that might have otherwise been used to press MOF to
resolve the nonperforming loan problem. Muramatsu and Scheiner (2009) show that interest
groups were far likelier to communicate with bureaucrats than politicians in 1994—a year when
there was a rapid succession from an eight-party non-LDP government, to a seven-party non-
67
LDP government, and then to a Socialist-LDP coalition government—than in 1979 at the height
of stable LDP dominance.
Though the LDP ultimately returned to power as the junior member of a coalition
government between June 1994 and January 1996, its brief absence from government damaged
its traditionally collegial relationship with MOF, which had previously reoriented itself to serve
the two short-lived non-LDP governments. The LDP now adopted a more critical stance toward
the ministry, first disregarding its preferred solution to the dissolution of several highly-indebted
housing loan corporations in 1995, then requiring that Diet debate on the corporations’
resolution address the additional subject of MOF reform (Amyx 2004, 168). The party further
shifted toward ministerial reform in October 1996, when it declared that it would take away
MOF’s authority over the inspection and supervision of private finance institutions to gain
support ahead of an impending general election, and in March 1997, when it and its junior
coalition partners submitted bills to the Diet that would formally divest MOF of these regulatory
responsibilities and transfer them to a new independent agency attached to the prime minister’s
office called the Financial Supervisory Agency (FSA) (Amyx 2004, 175).
14
Collectively, these
actions by the LDP following its reappearance in government are in line with the hypothetical
expectation that the party’s more electorally vulnerable standing would make it want to enact
constraints on bureaucratic policymaking.
These steps by the LDP represented initial attempts to increase the party’s involvement in
banking regulation. Yet, for several reasons, they did not immediately lift the veil on the
conditions in Japan’s banking sector. For one, though the FSA was supposed to produce
actionable information on the nonperforming loan problem by inspecting and supervising
14
In 2000, the name of the Financial Supervisory Agency changed to the Financial Services Agency.
68
Japanese banks, MOF exercised a heavy hand in steering the agency’s regulatory agenda away
from rigorous assessment and reporting of bank failure. Also, the founding of the agency
deepened MOF officials’ reluctance to disclose information to politicians. Furthermore, because
MOF did not stop the collapse of several small banks beginning in 1995, banks began to doubt
the ministry’s capacity to rescue all troubled financial institutions, which lessened banks’
inclination to share information with officials and thereby limited what the latter theoretically
could have said to politicians (Amyx 2004, 177–78). These factors prolonged the opacity
surrounding the debt lurking in the banking system until November 1997, when a series of
failures by the major financial institutions Sanyo Securities, Hokkaido Takushoku Bank,
Yamaichi Securities, and Tokuyo City Bank publically exposed MOF’s powerlessness to protect
against bank failure, and foreboded massive problems across Japan’s entire financial sector.
The fallout from these major institutions’ collapse led LDP politicians to respond with
bank recapitalization efforts beginning in late 1997. In principle, these efforts could have been a
significant move toward addressing the debt that incentivized banks to create zombie firms.
However, the first wave of capital injections was underutilized, with only ¥1.8 trillion out of a
possible ¥13 trillion going to 21 large banks. Moreover, the injections did not force banks to
write off their bad loans (Fukao 2000, 4). A second round of recapitalization then occurred with
the Financial Revitalization Act and the Financial Function Early Strengthening Act in October
1998, with the former targeting the resolution of insolvent deposit finance institutions, and the
latter addressing solvent financial institutions that needed capital to maintain investor and
depositor confidence. These two laws were better conceived than the preceding attempt to
support weak banks, and succeeded in bringing a measure of stability to Japan’s financial system
because of the sizable public funding involved—¥60 trillion ($500 billion)—and their
accompaniment by legal and institutional adjustments to system governance. These latter
69
changes included binding requirements on nonperforming loan disclosure, adherence to the
international standard definition of nonperforming loans, and stronger autonomy for the FSA to
regulate the banking sector without MOF interference (Amyx 2004, 209–10).
Following the implementation of these acts in late 1998, the ratio of nonperforming loans
at major Japanese banks fell over the next two years, as did the ratio zombie listed firms (Figure
2.1.). The outcome was consistent with the theoretical expectations that as the banking sector’s
health became transparent, politicians would intervene to assist the banks through efforts such as
recapitalization, and that stabilized banks would subsequently have less incentive to support
zombie firms. It was also aligned with the theory that electoral vulnerability would lead
politicians to establish more robust oversight mechanisms toward the sector, given that LDP
politicians could no longer count on having a steady relationship with MOF and would need
better policy information to act decisively in their more uncertain and possibly shorter periods in
government.
While the nonperforming loan ratio gradually fell for several years after 1998 in
accordance with these changes, the sum of the nonperforming and sub-performing loan ratios
rose slightly over the same period, largely because of poor external business conditions and
deteriorating asset prices. This combined ratio and the nonperforming loan ratio then spiked in
2001, a year when Japan’s stock market endured repeated declines and shrunk the unrealized
profits in banks’ portfolios (Amyx 2004, 209, 238). Because the drop in unrealized profits affected
banks’ ability to write off bad loans, the year saw a surge in zombie firms that reached nearly the
same level as the height of the zombie problem in 1998.
In 2002, the stock market began to rebound, which facilitated banks’ debt disposal and
corresponded with a decline in the zombie firm ratio. That same year, acting from a new cabinet
position created by the Koizumi administration to support the prime minister’s leadership,
70
Financial Services Minister Takenaka Heizo directed the FSA to take a firm stance on banks’ bad
loans, and concentrated on inspections of corporate borrowers with outstanding loans of ¥10
billion or more (Amyx 2004, 250)—highly leveraged firms that were more likely to have
repayment difficulties and demand financial relief. These actions, as well as the improved
economic climate, caused major banks’ nonperforming loan ratios to fall from 8.4 percent in
March 2002 to 2.9 percent in March 2005 (Vogel 2006, 29), lessening the amount of debt that
had previously incentivized banks to engage in forbearance lending and make distressed
borrowers into zombies.
At the same time, the activation of several accounting reforms, including a change to
market-value reporting for cross-shareholding in 2001, put profitability pressures on banks and
large companies that had once bought cross-shareholdings in the bubble years and now had to
report these stocks at the prices which they were currently worth, which was much less after stock
market decline beginning in 2000. These new rules pushed banks and firms to sell off cross-
shareholdings before 2001 to avoid potential negative impacts on institutional performance
(Schaede 2013, 100), which reduced banks’ incentive to sustain weak firms after the rules came
into effect. This change, in addition to the assertive regulation by Takenaka’s politically-
empowered FSA that increased the transparency around the banking sector and a recovering
macroeconomic environment, substantially decreased the zombie firm ratio among listed firms in
the first two years after 2001, especially in Japan’s manufacturing sector (Figure 2.4.). By 2003,
the ratio fell to a lower bound of roughly 5 percent, a modest level that continued until the global
financial crisis.
71
Bankruptcy Law
Lack of transparent banking regulation until reforms in the late 1990s was a key condition
that enabled undercapitalized banks to act on their incentive disguise debt through financial
relief to troubled firms, which turned the latter into zombies. However, from distressed firms’
standpoint, zombie status may not have been an optimal outcome, particularly if the firms’ future
business prospects were doubtful. The reason is that firms that persist in carrying large debt may
be forced to take risky decisions in the future to recover financial health that, if unsuccessful,
further damage or even destroy the firms’ solvency. It may be cheaper for impaired firms to
undergo restructuring at an early date to negotiate their debt (Povel 1999), rather than prolong
indebtedness with financial relief and confront choices down the line that possibly cause a loss of
core assets or even closure.
A traditional way that firms to pursue restructuring is through legal bankruptcy.
Restructuring through bankruptcy can be an appealing option for distressed firms looking to deal
with overinvestment under certain conditions, such as those in the famous Chapter 11 of the US
Bankruptcy Code which allows insolvent firms to keep control of their business operations as
debtor-in-possession, a status that affords privileges like the power to cancel or reject contracts
and automatic stays against creditor collection and litigation attempts. Moreover, for the
managers of such firms, the benefit of restructuring through bankruptcy may also extend to the
managers’ personal ability to keep their jobs. Therefore, a basic hypothesis about zombie firms is
that they should be less likely to occur where bankruptcy legal frameworks are conducive to
restructuring. If restructuring is a viable alternative, distressed firms should have a greater
incentive to forgo requests for financial relief associated with zombie status, meaning that there
should be fewer zombie firms.
72
Consistent with these theoretical expectations, the prevailing bankruptcy law in Japan in
the 1990s made it difficult for indebted firms to enter bankruptcy courts, let alone pursue private
liquidation, which coincides with the high incidence of zombie firms during those years. Under
the existing legal framework, three significant barriers confronted listed firms considering
reorganization. First, managers had to cede control of their firms during reorganization
proceedings, an impediment that immediately made managers doubtful of the desirability of
reorganization and delay bankruptcy as long as possible (Xu 2008, 190). Second, firms had to
pay the estimated costs of the proceedings up front, which placed demands on the firms’ already-
limited cash flow and put them at risk of igniting creditor panic if their creditors were to find out.
Third, there was no automatic stay of the exercise of unsecured claims, meaning that firms risked
losing necessary assets to stay in business if they went to court (Packer and Ryser 1992, 17–18).
Although the law did not force incumbent managers to resign when proceedings started (Hirose
2009, 211), these other factors cumulatively meant that there were few incentives for indebted
firms to ever voluntarily file for bankruptcy. Thus, during the banking crisis in the 1990s, it was
logical that distressed firms sought financial relief from banks and thereby took zombie status.
Complications within the reorganization framework also existed on the side of banks as
the firms’ creditors. Though impediments to entering reorganization court were primarily on the
debtors’ side, when cases did go to court, their outcomes tended to be strongly pro-debtor. The
existing corporate reorganization law prevented secured creditors from foreclosing on debtors,
meaning that banks could do little to reclaim assets once proceedings were underway. Moreover,
since courts wanted to see successful reorganization, creditors were typically forced to make large
concessions (Yanagawa, Hirose, and Akiyoshi 2005). Consequently, it was rational from banks’
perspective to extend credit assistance to weak businesses, rather than force them into bankruptcy
73
court where the banks would likely incur equity losses. Hirose (2009, 209) argues that this was
one reason why banks were willing to turn distressed firms into zombies during the 1990s.
Due to the cumbersome nature of bankruptcy law, Japan had an exceptionally low rate of
formal insolvency from 1952 until the early 1980s, averaging only 2,254 filings annually
including consumer bankruptcies. While the number rose in the 1980s and grew dramatically in
decade of the banking crisis from 1992 until 2001 when it reached 150,000 filings per year, most
of the increase was from personal liquidations. The total number of corporate reorganizations
remained relatively stable in Japan between 1952 and 2000, and never passed 1,000 in any year
(Anderson and Ito 2007, 595). The scarcity of bankruptcy proceedings also meant that there were
few lawyers who handled insolvency, with the result that when civil disputes occurred, organized
criminals (yakuza) often became involved as intermediaries. Informal resolution of financial
distress, particularly among smaller businesses, developed into a significant source of income for
underworld groups in the postwar period, to the extent that some gangsters even became
specialists in bankruptcy management (tosan seiri) (Hill 2003, 122–24; Kaplan and Dubro 2012,
213; Campana and Okamura 2016). The fundamental reason for this trend was the inhospitable
character of the bankruptcy regime, which, ironically but often with tragic consequence, made
engaging criminals a sufferable alternative for many troubled firms. The only other option for
indebted firms was to seek financial relief from their creditors, which made zombie status a
comparably acceptable outcome.
Awareness of this situation existed to some degree among Japanese government officials
in the early 1990s when the number of impaired firms surged after the implosion of the bubble
economy. One MOF official whose duties concerned legal aspects of bank-firm relationships at
the time said, “In cases which I’ve worked where banks didn’t lend to distressed firms, gangsters
came to the court…People within MOF did try to make it so that gangsters wouldn’t come when
74
firms were on the verge of bankruptcy” (Interview 22). On the other hand, the same official
confided that his own boss did not understand this point, and that there were relatively few
officials who realized the connection between bankruptcy rules and banks’ forbearance behavior,
possibly because the ministry did not have enough staff. Furthermore, Japan’s bankruptcy
framework had remained largely unchanged since the 1920s, albeit with minor adjustments in
1936 and 1952, so even thinking about bankruptcy reform was radical at the time (Interview 22).
The government moved to rectify the bankruptcy law in October 1996 with the
convention of an insolvency law deliberative council (shingikai). Based on discussions at the
council, a new bankruptcy law called the Civil Rehabilitation Law (Minji saisei ho) was proposed at
the Diet, passed in December 1999, and activated in April 2000. Specifically, this law was an
update to the Composition Law (Wagi ho), which had come into existence nearly 80 years earlier
in 1922.
The Civil Rehabilitation Law went a considerable way toward realigning the incentive
structure under which debtors and creditors approached restructuring. For one, the law
introduced the debtor-in-possession principle, which enabled distressed firms to stay in control of
their business after filing for bankruptcy, and gave managers confidence that they would keep
their jobs during the proceedings. The law also granted an order of stay on foreclosure and an
extinction of security rights to help keep firms in business during the proceedings, and relaxed the
approval requirements for firms’ proposed reorganization plans from near unanimity to a simple
majority, thereby expediting the decisions by unsecured creditors and bankruptcy judges (Hirose
2009, 211–12; Vogel 2006, 86).
The appeal of the Civil Rehabilitation Law for indebted firms was evident in its
immediate uptake. In the first nine months since the law’s activation in 2000, 729 firms applied
to use the law and 556 received approval. The next three years saw even more firms take
75
advantage: 916 applications and 709 approvals in 2001, 838 applications and 719 approvals in
2002, and 706 applications and 590 approvals in 2003 (Tokyo Shoko Research 2017). Though
not all the firms entering bankruptcy proceedings through the Civil Rehabilitation Law were
listed firms, most of Japan’s largest insolvencies utilized this law, including famous cases like the
department store Sogo and the supermarket chain Mycal (Anderson and Ito 2007, 594). Using a
sample of listed firms from the Tokyo Shoko Research database between 1995 and 2003, Hirose
and Akiyoshi (2010) empirically show that the introduction of the Civil Rehabilitation Law
quickened the decision of distressed firms in this corporate category to pursue restructuring.
While it is a counterfactual assumption that these listed firms would have become zombies if not
for this change in Japan’s legal environment, the considerable shift in incentives toward
restructuring brought about by this law corresponds with the decline in zombie listed firms
observed from 2001, meaning that it is plausible that bankruptcy reform played a role in the
reduction of zombies among Japan’s listed firms.
Conclusion
This chapter examined the case of zombie firms among listed firms in Japan in the lost
decades. Using descriptive statistics from a recent study covering the period 1995-2008, it began
from the observation that zombies increased after the bubble economy’s implosion in the early
1990s, and declined after 2001. In short, zombie listed firms were a problem primarily in Japan’s
first lost decade, not afterward.
After the reviewing the three orienting hypotheses and excluding the hypothesis of policy
drift on grounds of inapplicability to listed firms, the chapter explored the hypotheses of
transparency and bankruptcy law in relation to the statistical rise and fall of zombies among this
76
corporate category. It first found evidence that was largely consistent with the transparency
hypothesis. In line with the hypothesis’s expectations, lack of transparency delayed politicians’
interventions to address the health of Japan’s banking sector, and encouraged undercapitalized
banks to practice forbearance toward weak firms and thereby sustain zombie firms. A
fundamental reason for this opacity was the structural relationship between the singularly
dominant LDP and MOF, which had historically empowered MOF to manage the banking
sector and provided politicians with little motivation to develop independent monitoring
capacities over the ministry’s regulatory activities. Once the asset bubble burst in the early 1990s,
LDP politicians struggled to gain information about the scale of the problems at Japanese banks,
which gave MOF time to pursue its preferred regulatory approach that supported banks in
disguising their debt by offering financial relief to impaired borrowers, in the hopes that the
borrowing firms’ eventual repayment would rescue the banks’ balance sheets. In this context,
indebted banks had only weak incentive to support the Schumpeterian creative destruction of
large borrowers, and found it logical to allow distressed firms to become zombies, especially if the
zombie status was believed to be temporary.
The transparency hypothesis did not predict that the LDP would lose power early into
the crisis, but it did generally anticipate the actions that party politicians took after reentering
government. Together with coalition partners, newly vulnerable politicians from successive LDP
governments began to chip away at MOF’s control over regulation in order to increase
transparency and gain more information about the conditions at Japan’s banks. Though MOF
tried to protect its traditional turf, massive failures at several major banks in November 1997
ultimately revealed the severity of the situation, and prompted much stronger political
intervention to improve banks’ health, including recapitalization efforts and rule changes aimed
at clarifying the true extent of banks’ nonperforming loans and getting them off banks’ balance
77
sheets once and for all. While the poor economic environment, exacerbated by drops in the stock
market, hindered banks’ ability to write off debt, cumulative actions by LDP governments
starting in the late 1990s initiated the recovery of the banking sector in the early 2000s, and
reduced banks’ incentive to generate zombie firms. Consequently, the ratio of zombie listed firms
rapidly declined and remained at a low level in the mid-2000s.
After finding support for the transparency hypothesis, the chapter next uncovered
evidence aligned with the second orienting hypothesis on the influence of bankruptcy laws on the
prevalence of zombie firms. In the first decade after the burst of the bubble economy, Japan’s
extant bankruptcy regime was unconducive to restructuring by listed firms. Not only were
managers forced to hand over control of their firms in reorganization proceedings, but they also
had to pay for the proceedings in advance, and had no guarantee that they would not lose
necessary assets if they went to court. The unappealing nature of these rules meant that distressed
firms were highly incentivized to avoid bankruptcy entirely, and made demanding financial relief
from banks and taking zombie status a palatable choice, especially because the alternative
involved engaging gangsters in risky informal arrangements to settle firms’ debt.
The situation changed with the introduction of the Civil Rehabilitation Act in April 2000.
Under the new bankruptcy framework, firms enjoyed the principle of debtor-in-possession, which
protected their control of their business after they filed for bankruptcy. Firms also received stays
on foreclosure during court proceedings, and benefitted from a simplified approval process for
their reorganization proposals. Together, these improvements inspired many distressed listed
firms to take advantage of the law soon after its establishment. Though it is a counterfactual
argument that the firms which subsequently entered reorganization proceedings after 2000
would have necessarily become zombies had they not moved to restructure under the new law,
78
the substantial number of restructurings that occurred in the early 2000s closely coincides with
the decrease in zombie listed firms observed in the same period.
Having unearthed evidence supportive of two of the dissertation’s orienting hypotheses,
the chapter offers two preliminary explanations for why zombie firms can rise and fall after
economic crises. These explanations, derived from the case of Japanese listed firms, carry several
implications for our understanding of the political economy of zombie firms more generally. On
a positive note, the fact that Japan was able to resolve its zombie problem among listed firms,
even though resolution took considerable time, means that transparency and bankruptcy law can
be seen as possible lessons from Japan’s experience with zombies for other countries hoping to
avoid zombie firms after their own crises. Moreover, the rareness of dominant single party
systems like Japan’s suggests that few other countries have equivalent distributions of regulatory
authority and monitoring ability between their bureaucracies and legislatures which could
threaten transparency in the similar way as in Japan during the 1990s. In short, the
uncommonness of the structural power aspect of the transparency mechanism may be a good
thing for other countries concerned about possibly repeating a prolonged struggle with zombies
like Japan did during the first lost decade until the early 2000s.
Though the case of listed firms reveals how and why Japan was slow to stem the rise of
certain zombie firms as well as how and why it eventually accomplished this task, listed firms are
a small fraction of the overall firms in Japan’s economy, like they are in any economy. To gain a
fuller picture of Japan’s zombie situation in the lost decades, it is also important to comprehend
the status of unlisted firms, most of which fall under the broad corporate category of SME.
Recent economics research has also suggested that while Japan ultimately dealt with zombies
among listed firms around the beginning of the new millennium, the zombie ratio for SME was
stubbornly high throughout the 2000s and even into the early 2010s, implying that zombies
79
among this other firm category were produced by a mechanism that did not affect listed firms
and was not impacted by transparency and bankruptcy law changes.
The next chapter turns to this important but often overlooked firm category and a
different potential theoretical political economy explanation for zombie firms, drift.
80
CHAPTER 3.
Zombie Firms among Japanese SME, 1994-2014
This chapter examines the case of zombie firms among small and medium-sized
enterprises (SME) in Japan during the two lost decades since the banking crisis of the 1990s.
Though SME—a corporate category comprised of manufacturing firms with no more than 300
employees and ¥300 million in capital, and wholesale, service and retail firms with no more than
100 employees and ¥100 million in capital—may not attract as much attention as listed firms
with higher individual employment and capitalization figures and more famous products and
services, they are commonly considered the “backbone” of the Japanese economy for their
importance in terms of overall firm numbers and contribution to national employment. In recent
years, SME have represented more than 99 percent of Japanese companies, 70 percent of
Japanese employment, and more than 50 percent of added value in the manufacturing industry
(Small and Medium Enterprise Agency 2008, 24–25). Therefore, the zombie presence in this
corporate category carries broad significance for the overall Japanese economy as well as the
wider debate about zombie firms.
Contrary to the previous chapter’s finding that zombie listed firms emerged during the
banking crisis and declined in the early 2000s, this chapter shows that the zombie ratio among
SME remained high throughout the 2000s. This outcome represents a notable divergence from
the situation among listed firms, and implies that to the extent that zombie firms explain Japan’s
economic performance during the second lost decade, greater responsibility lies with this class of
enterprise that is smaller in size yet considerably more numerous within the Japanese economy.
It also signifies that a different mechanism possibly undergirds zombie firms among SME which
81
has endured much longer than the factors that created zombies among Japan’s listed firms, and
that some form of protectionism toward these firms has persisted in defiance of major theoretical
argument that electoral system reform in the mid-1990s brought neoliberal economic policies to
Japan in the early 21
st
century.
The chapter proceeds in three parts. First, it presents descriptive statistics on zombie firms
among Japanese SME. These statistics depict the zombie ratio among SME across the lost
decades—the outcome variable of interest—as well as certain corporate attributes of zombie
SME, including the zombie presence among SME of different sizes, the zombie presence across
different industries and industrial sectors, and the average time duration that zombie SME keep
zombie status. These statistics are mainly drawn from an analysis by Goto and Wilbur
(Forthcoming) of a large sample of SME from the Ministry of Economy, Trade and Industry
(METI) Basic Survey of Japanese Business Structure and Activities (BSBSA), making them
broadly representative of SME with 50 or more employees and whose paid-up capital and
investment are higher than ¥30 million.
Next, the chapter discusses the role of Japan’s credit guarantee system for SME in
relationship to the relatively high zombie ratio apparent in these statistics. It first describes how
assurance policies that systematically increase the availability of credit may lead to higher
incidence of zombie firms. It then explains how Japan’s credit guarantee system expanded the
credit available to SME throughout the 1990s and 2000s—even during the generally positive
economic climate in the mid-2000s that contributed to the recovery of zombie listed firms—with
emphasis on how the system’s substantial size and complete and nearly-complete coverage ratios
facilitated bank lending to firms with dubious creditworthiness and high potential to demand
financial relief.
82
Third, the chapter explains the political underpinnings of Japan’s credit guarantee
system, clarifying how and why the system’s high-level guarantees lasted throughout the 2000s
despite their questionable effect on giving life support to fundamentally weak firms, many of
which became zombies. In accounting for how the system endured largely unchanged during this
period, the chapter marshals evidence that the system experienced policy drift through a
combination of factors including limited public access to policymaking, uncertainty about the
possible consequences of policy adjustment, and collective action barriers for the pro-reform
coalition. These factors trumped opposition by economists concerned about the system’s effect on
assisting growth, and public sentiment which generally favored neoliberal structural reforms and
showed little interest in SME policy compared to other areas for government action.
Statistics on Zombie Firms among Japan’s SME
Until recently, research on zombie firms in Japan neglected their occurrence among
SME, despite the category’s inclusion of the vast majority of Japanese firms (Caballero, Hoshi,
and Kashyap 2008; Fukuda and Nakamura 2011). The first known analysis of zombie SME is by
Imai (2016), who, using the FN definition of zombie firms, identifies the zombie ratio among a
sample of unlisted firms from the Tokyo Shoko Research database. Imai finds that in contrast to
earlier studies suggesting that Japanese banks rarely evergreened their loans to SME and thereby
produced few zombie firms (Fukuda, Kasuya, and Nakajima 2006; Sakai, Uesugi, and Watanabe
2010), the zombie firm ratio for SME annually ranged between 5 and 14 percent and averaged 8
percent throughout the period 1999-2008 (Imai 2016, 94–95). This ratio was notably higher than
the ratio for listed firms, which, in the original FN estimates, rose to a high of nearly 10 percent
83
Figure 3.1. Zombie Firm Ratio among Japanese SME by Equity Capital, FN Definition, 1999-
2008
Source: Imai (2016, 96)
of firms in 2001 and then remained near 5 percent between 2003 and 2007 until the start of the
global financial crisis (Fukuda and Nakamura 2011, 1128; Nakamura and Fukuda 2013, 6).
Imai further shows that different SME subcategories had different zombie ratios. Very
small firms with less than ¥10 million (roughly $100,000) in equity capital had a peak zombie
ratio of more than 25 percent in 1999, and a baseline ratio of at least 12 percent throughout the
period 1999-2008. During the same period, the zombie ratio for the largest size category of SME
with more than ¥1 billion (roughly $10 million) in equity capital is at least 5 percent lower in
every year except 2004 (Imai 2016, 95–96) (Figure 3.1.). In short, the zombie ratio among
Japanese SME is significantly higher than the ratio for listed firms for much of the 2000s, and the
smallest SME have a particularly high tendency to undergo zombification (Imai 2016, 97–98).
One weakness in Imai’s study is its relatively small sample size of 2,357 firms, which limits
the scale of its inference about zombie firms among SME. To amend this deficiency and explore
96 K. Imai / J. Japanese Int. Economies 39 (2016) 91–107
0
0.05
0.1
0.15
0.2
0.25
0.3
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
YEAR
Less than ¥ 10million Between ¥ 10million and ¥ 100million
Between ¥ 100million and ¥ 1billion More than ¥ 1billion
Fig. 1. FN zombie ratios by equity capital.
0
0.05
0.1
0.15
0.2
0.25
2002 2003 2004 2005 2006 2007 2008
YEAR
Less than ¥ 10million Between ¥ 10million and ¥ 100million
Between ¥ 100million and ¥ 1billion More than ¥ 1billion
Fig. 2. Modified FN zombie ratios by equity capital.
capitalized at less than 10 million yen. Between 7% and 20% of the SMEs capitalized at less than
10 million yen were zombie firms by the modified FN zombie criterion. Using the modified FN zom-
bie criterion, the percentage of SMEs capitalized at between 10 million and 100 million yen that were
zombies is approximately 4%. Additionally, the percentage of SMEs capitalized at between 100 million
1 billion yen that were zombies is in the range of 4–7%. These percentages are relatively low and have
similar values. In addition, employing the modified FN criterion, the percentage of firms capitalized
at more than 1 billion yen that were zombies is in the range of 4–13%. The percentage of zombies for
these large firms was relatively high.
Interestingly, the results above indicate that evergreening occurred even in SMEs. Therefore, it sug-
gests that the Regional Banks, Second Association of Regional Banks, Sinkin Banks and Credit Unions,
84
Figure 3.2. Zombie Firm Ratio among Japanese SME by Debt Type, FN Definition, 2009-2014
Source: Goto and Wilbur (Forthcoming)
additional issues surrounding zombie SME like the zombie firms’ average time spent in zombie
status, Goto and Wilbur (Forthcoming) use a sample from the METI BSBSA of more than
30,000 firms with between 50 and 499 employees and at least ¥30 million in equity capital. Their
estimate of the zombie ratio among SME with FN zombie definition with this enlarged sample is
limited to the period 2010-2014, since the BSBSA only contains data on short-term and long-
term bank borrowing and corporate bond issuance from the year 2009, and the FN definition
requires the calculation of differences in bank borrowing amount from the previous year.
However, even in this short period, they observe that the zombie ratio among SME was more
than 6 percent of sample firms, a non-negligible amount (Figure 3.2.).
More suggestively, Goto and Wilbur also estimate the zombie ratio with current and fixed
liabilities (what they call “debt-based” zombies) as a proxy for bank borrowing to generate a
longer time series covering the period 1994-2014. Using this proxy variable for zombie firms,
they find that the debt-based zombie ratio is at least 10 percent of firms throughout both lost
decades, a level roughly twice as high as the level among large firms estimated by FN between
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
20.0
2009 2010 2011 2012 2013 2014
by borrowings from banks by current and fixed liabilities
85
Figure 3.3. Debt-Based Zombie Firm Ratio among Japanese SME, 1994-2014
Source: Goto and Wilbur (Forthcoming)
2003 and 2007 (Nakamura and Fukuda 2013, 6), though they are careful to stress that their
proxy measure is not the same as the FN definition of zombie firms (Figure 3.3.).
Like Imai, Goto and Wilbur also disaggregate the zombie ratio among SME of different
sizes. They find that the zombie ratio for SME with equity capital between ¥30 million and ¥100
million was roughly 8 percent in the period 2010-2014, a consistently higher level than those of
two larger-size SME categories whose ratios stayed between 3 and 5 percent (Figure 3.4.). This
outcome extends Imai’s finding that smaller-size SME have higher zombie ratios than their
larger counterparts, though Goto and Wilbur are unable to replicate Imai’s finding on very small
SME with less than ¥10 million in equity capital due to the limitations of the BSBSA data which
only contains firms with more than ¥30 million in paid-up capital.
Other noteworthy results in Goto and Wilbur include the finding that for SME identified
as zombies in one year, their probability of keeping zombie status in the next year decreases,
meaning that zombie SME tend not to keep zombie status indefinitely, but rather recover
74.0
76.0
78.0
80.0
82.0
84.0
86.0
88.0
90.0
92.0
0.0
5.0
10.0
15.0
20.0
25.0
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
FN Imai CHK (right scale)
86
Figure 3.4. Zombie Firm Ratio among Japanese SME by Firm Size, 2010-2014
Source: Goto and Wilbur (Forthcoming)
viability or exit the marketplace in a short time. This finding is somewhat similar to the finding
about zombie listed firms by Nakamura, who shows that large zombies tend to retain their
zombie status for slightly more than two years on average (Nakamura 2017, 15). Goto and
Wilbur further find that manufacturing firms display a higher zombie ratio than non-
manufacturing firms, and that there is a positive correlation between zombie formation and firm
age.
Zombie Firms among Japan’s SME and the Credit Guarantee System
These statistics signify that zombie SME existed at a notably higher rate than zombie
listed firms in the 2000s. This outcome suggests that a different mechanism possibly supported
the creation of zombie SME, given that the factors which produced zombies among Japan’s
listed firms were largely tackled by bank supervision and accounting reforms starting in the late
1990s, and that the zombie ratio among listed firms decreased to a low level after 2001 as
addressed in the previous chapter. Moreover, Japan’s economy expanded in the mid-2000s,
posting average annual real GDP growth of 2 percent between 2003 and 2007 that was twice the
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
2009 2010 2011 2012 2013 2014
FN criterion
Small Middle Large
87
1 percent average between 1992 and 2002 (Lincoln 2011, 357). Aggregate economic recovery
lessened the “unnatural selection” rationale for indebted banks to evergreen loans to distressed
firms to hide the banks’ debt, since growth propelled many firms to financial health and
decreased the level of nonperforming loans in Japanese banks from 8.4 percent of total lending in
March 2002 to 4.7 percent in September 2004 (OECD 2005, 14).
Additionally, the continually high zombie ratio among SME in the 2000s puzzlingly
defies the expectations of prominent theory holding that reform of Japan’s electoral institutions in
the mid-1990s shifted Japan’s political economy away from particularism toward neoliberalism.
According to this theory, change in electoral rules in Japan’s lower house in 1994 from multi-
member districts with a single nontransferable vote to a combination of single member districts
and proportional blocs “shook the hidden ballast undergirding the ‘Japanese variety of
capitalism’ by lessening political incentives to cultivate personal sources of campaign capital,”
thereby auguring economic revitalization through policies that would be less beholden to narrow
interests, and instead responsive to urban voters concerned about curbing wasteful spending and
lowering budget deficits (Rosenbluth and Thies 2010, 125, 134). Nonetheless, substantial lending
to zombie SME occurred well into the 21
st
century, calling into question this theory’s idea that
“capital has become more mobile in Japan’s new political economy” since electoral reform
(Rosenbluth and Thies 2010, 126).
How did zombie firms continue to exist at a high level among Japan’s SME in the early
21
st
century despite financial reforms, a generally positive macroeconomic climate, and recent
electoral change? One possibility is that zombie firms were generated by policies that
systematically enhanced the availability of loans to firms whose financial conditions were
fundamentally tenuous and therefore prone to repayment difficulty. To recall, banks have an
incentive to extend financial leniency toward distressed firms when they believe that the firms can
88
ultimately return to profitability and viability (Peek 2009). This motivation may be heightened (or
“perverted”) in serious recessions when banks hold large amounts of debt and view refinancing to
potentially defaulting firms as a way to disguise nonperforming loans and avoid fire sales of
collateral to cover debt write-offs (Peek and Rosengren 2005; Jaskowski 2015). But there is no
theoretical reason to believe that it only exists during crises. Rather, the key factor remains banks’
assessment of default risk. If banks have grounds to believe that they can ultimately recover their
loans, the banks may have an incentive to provide financial relief that turns borrowers into
zombie firms, since the banks should both recover their loans and earn interest payments.
Consequently, one potential political cause of zombie firms might be any policy
arrangement that systematically induces banks to provide financing to firms with weak financial
health, which should increase the number of instances where banks choose to support distress-
prone borrowers as zombies. To be clear, such a credit enhancing arrangement would not have
creating zombie firms as its main intension, since banks profit more when loans are repaid on-
time and at non-discounted interest rates, and should therefore prefer to avoid the risks of
zombification associated with weakly-performing borrowers. However, it would heighten the
probability that banks are confronted with scenarios where they may decide to assist firms in
arrears and turn them into zombies, because it would expand the banks’ potential customer base
to include businesses whose financial conditions might otherwise exclude them as borrowers or at
the very least rank them lower among the banks’ desired clientele.
Japan’s Credit Guarantee System as a Credit Enhancement Arrangement
Japan has possessed precisely such a credit enhancement arrangement in the form of its
credit guarantee system (Shin’yo hosho seido) for SME, a system through which the Japanese
89
government guarantees loans to SME issued by private banks. This system dates to 1937, when
the Tokyo metropolitan government first established the Tokyo Credit Guarantee Corporation
based on a model that existed in Germany. Following the inclusion of the system as a key
measure in the SME Financial Measures Summary by the cabinet of Prime Minister Ashida
Hitoshi’s Democratic and Socialist coalition government in August 1948 (Bank of Japan
Financial Research Institute 1989, 602–3), other local governments founded their own credit
guarantee corporations, and the SME Credit Insurance Law and Credit Guarantee Corporation
Law were respectively enacted in December 1950 and August 1953 to codify the system’s
emerging financial structure.
15
Since 1961, Japan’s 47 prefectures and the five cities of Osaka,
Nagoya, Yokohama, Kawasaki, and Gifu have each had their own credit guarantee corporation,
which collectively constitute Japan’s credit guarantee system (SME Policy Council Finance
Working Group 2015, 8).
As its name implies, the credit guarantee system enhances the credit available to SME by
offering government guarantees on private bank financing, which mitigates banks’ default risk
and encourages them to lend to SME borrowers. Figure 3.5. illustrates how the system fulfills this
function. First, a small business that wants to obtain a guaranteed loan applies for a credit
guarantee either at the counter of a credit guarantee corporation, or through a financial
institution that applies to the corporation on the firm’s behalf (1.). The corporation then conducts
a credit investigation of the firm, and if the guarantee is approved (2.), the bank issues a loan with
this guarantee to the firm (3.). After the loan is issued, the firm begins repaying the bank, and
15
In 1951, the SME Credit Insurance Law was partially revised so that credit guarantees provided by credit
guarantee corporations would be insured. This revision led to the creation of what is sometimes referred to as the
Credit Supplementation System (Shin’yo hokan seido) that combines credit guarantees with credit insurance (National
Federation of Credit Guarantee Corporations 2011, 2). To avoid confusion, the dissertation’s use of the term “credit
guarantee system” includes the system’s insurance aspect, and the term “credit supplementation system” is not used.
90
Figure 3.5. Schematic Map of Japan’s Credit Guarantee System
Source: Author, based on Ito (2011, 4)
pays a separate guarantee fee to the corporation (4.). If the bank does not receive the firm’s loan
repayment by a certain time, the corporation assumes the principal and interest obligations of the
bank, a process called subrogation (5.). After the corporation makes its subrogation payment to
the bank, it collects outstanding funds from the firm while looking at its business conditions (6.)
(Ito 2011, 3–4).
While the credit guarantee system engages SME insofar as it helps them acquire private
financing at the small additional cost of the guarantee fee, it also insures individual credit
guarantee corporations when they make subrogation payments to banks. The corporations’
primary insurer is the Japan Finance Corporation (JFC), a government finance institution (GFI)
Individual Credit Guarantee Corporations
Credit Guarantee Side
Credit Insurance Side
Small and Medium-sized Enterprises Financial Institutions
Regional Governments
National Federation of Credit
Guarantee Corporations
Japan Finance Corporation
Central Government
4. Repayment
3. Loan
1. Guarantee
Application
2. Guarantee
Agreement
5. Subrogation
Payment
6. Recovery 4. Guarantee
Fee Payment
1. Guarantee
Application
Loss Compensation Insurance
Payment
Insurance
Premium
Payment
Recovery
Payment
Subsidies & Supervision
Financing & Supervision
Financial Endowment
Subsidies & Supervision
Contributions, Loans,
Report Requests &
Inspections
91
that also provides direct loans to SME.
16
Credit guarantee corporations currently pay between
0.45 and 1.9 percent in insurance premiums to the JFC according to the type of subrogation
insurance (Yamori 2015, 8). They also pay the JFC any funds that the corporations collect from
defaulting firms after the subrogation payment, though the collection rate tends to be very low
(Ito 2011, 5).
17
Beside the JFC, three other actors operate on the insurance side of the system where the
cost of the system is primarily located. In terms of financial support for the system, two of these
actors have lesser importance: the National Federation of Credit Guarantee Corporations, which
is the peak organization for the 52 individual credit guarantee corporations; and regional
governments, including prefectural governments and municipal governments for the five cities
with credit guarantee corporations. These two actors primarily assist with individual credit
guarantee corporations’ loss compensation. The third actor who carries much more weight is the
central government, which provides compensation and contributions to the JFC and gives small
direct subsidies to individual corporations.
A comparison of these actors’ spending reveals their relative economic significance for
maintaining the system. In fiscal year 2011, regional governments paid a combined total of
nearly ¥75.5 billion in loss compensation to credit guarantee corporations. In the same year, the
JFC paid ¥660.1 billion in insurance payments to the corporations and received only ¥146.5
16
The credit guarantee system’s original insurer was the SME Credit Insurance Corporation founded in 1958 as a
public corporation. In July 1999, this corporation merged with the Japan Small Business Corporation and the
Textile Industry Restructuring Agency to form a new organization called the Japan Small and Medium Enterprise
Corporation (JASMEC). Five years later, in July 2004, JASMEC transferred its credit insurance operations to the
Japan Finance Corporation for Small and Medium Enterprise (JASME). In October 2008, JASME, the National
Life Finance Corporation (NLFC), the Agriculture, Forestry and Fisheries Finance Corporation, and the
International Financing Operations of Japan Bank for International Cooperation (JBIC) integrated and became the
Japan Finance Corporation (JFC), which currently insures the credit guarantee system (National Federation of
Credit Guarantee Corporations 2011, 2–3).
17
A credit guarantee corporation official corroborated this point (Interview 64).
92
billion in insurance premiums from the corporations, leaving the JFC with an insurance deficit of
¥397.9 billion (roughly $4 billion). To enable the JFC to absorb this sizable insurance loss and
continue its main backstop function for the credit guarantee system, the central government
apportioned ¥1.04 trillion to the JFC (Yamori 2015, 14–15). Though these figures vary slightly
by year, it is clear from this example that the central government bears the largest cost for
enabling the credit guarantee system to serve SME.
The Credit Guarantee System’s Role in SME Finance during the Lost Decades
Japan’s credit guarantee system was established in the early postwar period to enhance
financial access for all SME. But part of the system’s function has also been to boost credit
availability to targeted classes of firms through specific guarantee programs designed to meet
current policy objectives. Since 1955, the system has had over 50 specific programs in addition to
its regular guarantee program. For example, several specific programs have aimed at facilitating
business creation and expansion, such as programs in new technology development, overseas
investment financing, and new business development during the 1980s. Other targeted programs
have tried to increase the availability of credit to SME affected by natural disasters, such as the
1994 Kobe earthquake. Among the system’s specific programs, the largest have typically served
firms affected by economic change, such as its management stabilization programs to support
SME after the oil crisis in the 1970s and in the period of yen appreciation following the 1985
Plaza Accord (Nitani and Riding 2005, 59–62).
The biggest ever specific guarantee program was enacted as a response to Japan’s
banking crisis in the 1990s. In August 1998, the cabinet of newly designated Prime Minister
Obuchi Keizo decided on an outline of countermeasures for Japan’s tight credit market, and
93
included a ¥30 trillion increase in the credit guarantee system’s budget to expand the number of
loans issued through the system. Two months after this announcement, at the same time as the
Obuchi government passed the Financial Revitalization Act and the Financial Function Early
Strengthening Act to recapitalize major banks, it pushed through a specific guarantee program
called the “Special Guarantee for the Financial Stabilization of SME” (Chusho kigyo kin’yu anteika
tokubetsu shin’yo hosho) (henceforth “Special Guarantee Program”), which offered 100 percent
credit guarantees to SME whose lending institutions could show that the firms’ business
conditions had deteriorated since the previous year. Between October 1998 and termination in
March 2001, the Special Guarantee Program was used in 1.7 million loans totaling ¥28.9 trillion
($270 billion) in guaranteed loans, making it the biggest ever expansion of Japan’s credit
guarantee system (Uesugi and Sakai 2005, 6).
18
In the year 2000 alone, 455,959 loans were
approved through the program with a total loan value of ¥7.4 trillion, which were, respectively,
30 percent of the loans and 41.1 percent of the total loan value issued through the entire system
(Nitani and Riding 2005, 61).
19
While the Special Guarantee Program encouraged staggering numbers of SME to take
advantage of the credit guarantee system, a shift simultaneously underway in the Japanese
government’s relationship with SME financing also heightened the system’s profile at the turn of
the 21
st
century. Since the early postwar period, the Japanese government had been directly
involved in lending to SME through several GFI including the National Life Finance
18
Part of the reason for the Special Guarantee Program’s huge uptake was the ease with which SME could succeed
in their guarantee applications through the program. Credit guarantee corporations had a negative list of conditions
to reject applicants, including default and window dressing of balance sheets. However, these conditions were
notoriously hard to satisfy (Interview 32).
19
The METI Small and Medium Enterprise Agency (SMEA) claimed that the Special Guarantee Program’s massive
scale and utilization saved some 9,600 small firms from bankruptcy compared to the more than 28,000 firms that
went bankrupt in the program’s two-year period (SME Policy Council Finance Working Group 2015, 9). On the
other hand, outside observers countered that the program merely delayed rather than stopped a wave of
bankruptcies that occurred after the program’s end (OECD 2000, 46–47).
94
Corporation (NLFC) (1949- ), the Japan Finance Corporation for Small and Medium Enterprise
(JASME) (1953- ) and the Shoko Chukin Bank (1936- ). Loans by these three GFI, especially the
first two, served as a form of redistribution to SME dealing with economic turbulence during the
early postwar years, and were a preferred policy tool of the government because they did not
require additional budgetary outlays and thus enabled the government to concentrate its limited
fiscal resources on reindustrialization (Park 2011, 74). Successive LDP governments gradually
increased the amount of policy finance available to these GFI through the Fiscal Investment and
Loan Program (FILP) in the 1950s and 1960s, so that they could continue to provide for the
social security of SME and appeal to them as a political constituency while maintaining budget
restraint (Park 2011, 97–111). However, after the LDP’s poor showing in the December 1972
general election, the government significantly expanded GFI loans, increasing them sevenfold to
¥28.6 billion in 1979 and ultimately bringing them to ¥30.9 trillion in 1992 (Calder 1988, 346;
Shimizu 2013, 161).
As Japan’s economic situation deteriorated during the banking crisis in the mid-1990s, a
combination of mounting fiscal pressures and enhanced Cabinet Office powers enabled Prime
Minister Hashimoto Ryutaro initiate the long-held goal of FILP reform (Toyoda 2011, 162),
which implied a decline in the government’s ability to provide SME loans through the GFI. At
the same time, because the credit guarantee system relied on private banks, it became a crucial
platform for the government to continue its support of small firms. Figure 3.6. shows the
contrasting trends of public loans and guaranteed private loans as portions of overall SME
lending in the period 1997-2014. While the amounts of these two loan types were roughly equal
in 1997, from the time of the introduction of the Special Guarantee Program in 1998, the credit
guarantee system supported a considerably larger amount of loans to SME than the GFI. The
share of GFI loans within Japan’s overall SME finance stayed between 8.7 and 10.3 percent
95
Figure 3.6. Outstanding Loans and Outstanding Guaranteed Loans for Japanese SME and
Microenterprises, 1997-2014
Source: SME Policy Council Finance Working Group (2015, 11)
Unit: ¥1 trillion
between 1998 and 2014, but the absolute amount fell from ¥29 trillion in 1998 to ¥21.9 trillion
in 2014, a nearly 25 percent decline. By contrast, the credit guarantee system’s weight among
SME finance stayed between 11.4 and 14.8 percent during these years. In fiscal year 2014,
911,000 SME borrowed from the NLFC, 47,000 SME borrowed from the JASME, and 76,000
SME borrowed from the Shoko Chukin Bank. However, 1.4 million SME used credit
guarantees, meaning that the system served approximately one-third of Japan’s 3.85 million
SME (SME Policy Council Finance Working Group 2015, 9).
With the implementation of the Special Guarantee Program in 1998 and the
government’s concurrent shift from direct finance through GFI to indirect finance through
private banks, Japan’s credit guarantee system became the largest credit guarantee system in the
world (Christensen et al. 1999). The amount of loans guaranteed through the system declined
0
50
100
150
200
250
300
350
400
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
Government Finance Institutions Private Finance Institutions
Guaranteed Loans from PFI
96
Table 3.1. International Comparison of Credit Guarantee Systems, 2014
Country
Guarantee Acceptances
Outstanding Guarantee
Obligation (Flow)
Cases Amount Guaranteed (Stock)
America 63,460 2,603.4 7,959.1
Germany 6,472 130.4 701.2
France 86,207 428.2 1,662.9
UK 2,718 50.6 133.5
Korea 659,390 503.6 681.8
Japan 714,340 8,939.4 27,701.7
Source: Small and Medium Enterprise Agency (2016b, 3)
20
Financial Unit: ¥1 billion
after the Special Guarantee Program ended in 2001, but rose again following the global financial
crisis, when the Japanese government instituted another large-scale specific loan program called
the “Emergency Guarantee System” (Kinkyu hosho seido) that provided 100 percent guarantees
between October 2008 and March 2011. In 2014, outstanding guarantee obligations in Japan’s
credit guarantee system reached nearly ¥28 trillion (Table 3.1.). This figure was nearly four times
higher than the next largest credit guarantee system in America despite the two economies’ vast
size difference, and bigger than other major foreign credit guarantee systems combined.
Aside from unparalleled size, two other points are notable about Japan’s credit guarantee
system in the 21
st
century. The first point is the system’s high level of guarantee coverage.
Whereas credit guarantee systems in other major countries typically offered loan guarantees in a
range between 40 and 75 percent of loan values—a level at which banks still assume substantial
risk in case of loan default—Japan’s system provided guarantees between 80 and 100 percent of
loan values, with a large majority of the system users receiving 100 percent guarantees (Small and
Medium Enterprise Agency 2016b, 5) (Table 3.2.). As will be elaborated in the following section,
20
Figures from America are from 2015.
97
Table 3.2. International Comparison of Credit Guarantee Systems in Normal and Crisis Times
Normal Time Crisis Time (After the GFC)
Target Firms Coverage
Percentage
Range of 100
Percent Coverage
Measures to
Raise the
Guarantee
Ratio
Implementation
Period
America SME whom financial
institutions have difficulty
financing without
guarantees, etc.
Startup period, expansion
period, etc.
75, 85% -- Raised from 75,
85% in normal
time to 90%
uniformly
2009.2 2011.3
Germany All SME in principle
Startup period, business
succession period, etc.
60 80% -- Raised to a
maximum of
90%
2008.12 2010.12
France All SME in principle
Startup period, innovation
period, expansion period,
business transfer &
acquisition period
40 60% -- Raised to a
maximum of
90%
2008.12
2010.12
UK SME who have difficulty
borrowing due to collateral
shortage, etc.
Startup period, expansion
period, stable period, etc.
75% -- -- 2008.12 2010.12
Korea All SME in principle
Startup period, expansion
period, stable period,
regeneration period, etc.
50 100% Young
entrepreneur
startups, small
sum funds, policy
objectives
(promotion
projects, etc.)
Raised to 90 100%
--
Japan All SME in principle
Startup period, expansion
period, stable period,
regeneration period, etc.
80, 100% Microbusinesses,
startup-stage
firms, safety net
guarantee, etc.
Target firms
under Safety
Net Assurance
No. 5 expanded
sequentially to
cover all
industries
2008.10 2012.10
Designation of
all industries
from 2010.2
Source: Small and Medium Enterprise Agency (2016b, 5)
this high level of coverage freed banks to loosen their lending standards and provide loans to
firms with weaker creditworthiness, making Japan’s credit guarantee system more likely to sustain
zombies among its SME users.
98
The second point relates to the timing when Japan’s credit guarantee system provided its
high-level guarantees. As Table 3.2. shows, many countries temporarily increase their systems’
coverage ratios in response to economic downturns, which reflects how governments can adjust
credit guarantee systems to increase the availability of financing which otherwise might not exist,
a property of credit guarantee systems commonly known as “additionality” or “incrementality”
(Riding, Madill, and Haines 2007). By contrast, Japan’s provision of complete and nearly-
complete coverage occurred not only in crisis times, but even in generally positive
macroeconomic conditions. For example, Japan’s system consistently gave 100 percent
guarantees to microenterprises with 5 or less employees, a firm category constituting 85 percent
of all Japanese firms and 87 percent of the system’s 3.9 million users (Small and Medium
Enterprise Agency 2016a, 18; SME Policy Council Finance Working Group 2015, 9). The
continuation of high-level coverage to many users even in relatively good times implies that
Japan’s system was of vital importance to large numbers of Japanese SME that were
fundamentally challenged in obtaining credit.
Japan’s Credit Guarantee System as a Facilitator of Zombie Firms
Japan’s credit guarantee system, like other countries’ credit guarantee systems, has
primarily functioned to enhance users’ access to credit. Yet, by existing on a massive scale and
providing perfect or near-perfect guarantee ratios to millions of SME, it has also been responsible
for facilitating lending relationships between banks and weak firms, some of which have
undergone zombie status. A key reason why Japan’s system contributes to zombie formation is
that its guarantee ratios entirely or almost entirely eliminate banks’ default risk, which enables
banks to lend to SME with much less concern about the possibility of nonpayment. Removal of
99
risk allows banks to overcome the information asymmetries which typically make SME less
desirable customers, increasing banks’ willingness to finance to this category of borrower.
The certainty that Japan’s credit guarantee system affords banks in avoiding default
risk—which increases the banks’ willingness to lend to all firms, including weak firms more likely
to default and need refinancing that turns them into zombies—has widely been discussed by
Japanese economists as a moral hazard, since the insurance side of the system reduces banks’
potential for incurring financial loss and their corresponding motivation to distinguish between
viable and nonviable borrowers when screening loan applications (Ito 2011; Ono, Uesugi, and
Yasuda 2013; Saito and Tsuruta 2014). The system’s 100 percent guarantee, widely used by
entire categories of SME like microenterprises but also by many firms qualifying for specific loan
programs such as the Safety Net Number 5 program for SME in declining industries, is the
clearest example of how the system enables adverse selection, since it fully covers banks in the
case of repayment failure—thereby removing banks’ need to minimize loan defaults by
monitoring borrowers or advocating improved business planning—and gives borrowers no
incentive to restructure (Yamori 2011, 24). But even the system’s 80 percent guarantee, which
was first instituted in October 2007 as the so-called “responsibility-sharing system” (sekinin kyoyu
seido), has been insufficient at eliminating moral hazard, though it lessened it to a degree (Saito
and Tsuruta 2014).
While moral hazard is important in explaining the credit guarantee system’s effect on
increasing bank lending, it is only capable of creating large numbers of zombie firms to the extent
that many of the system’s users are prone to loan default and possibly needing banks to show
forbearance. Available evidence suggests that the financial health of many SME receiving credit
guarantees is indeed tenuous. Based on data from the credit guarantee system’s Credit Risk
Database (CRD), Shikano estimates that the average system user in fiscal year 2003 had 6
100
employees, ¥125 million ($1.25 million) in sales, ¥84 million ($840,000) in total assets, and ¥10
million ($100,000) in capital stock (Shikano 2008, 25). The CRD data further show that the
average user’s operating profit and net income were ¥1 million ($10,000) and ¥400,000 ($4,000).
These statistics reveal the very small size of most system users as well as their typically thin
operational margins, a pattern that occurs because much of the firms’ operational expenses like
human resources are fixed. Consequently, these firms are highly dependent on sales to survive,
and sales fluctuations can easily threaten their solvency (Shikano 2008, 52).
Beside the fragile business conditions of many credit guarantee system users, their
financial structure also makes them more likely to become zombie firms. The CRD data indicate
that most system users have low equity capital ratios—on average, approximately 11 percent—
which leaves them highly dependent on long-term borrowing and reinforces the significance of
their relationships with banks. Also, since more than half of system users with less than 20
employees have operational loss carryovers, it is apparent that many users are only able to
survive because of the financial life support they receive from bank loans (Shikano 2008, 85).
These findings are less true of large SME users with more than 100 employees, implying that the
bigger a SME grows, the less likely it is to have tenuous business conditions and rely on
borrowing, and thus possibly need financial relief associated with zombie status.
The descriptive statistics cited in this chapter depict a corresponding picture of the
zombie firm ratio among Japanese SME during the lost decades. The closest evidence appears in
Imai (2016), who shows that the smallest size category of SME in his sample—firms with less
than ¥10 million ($100,000) in capital stock—had an extremely high zombie ratio that exceeded
25 percent in 1999, the peak year of Japan’s credit guarantee system in terms of absolute amount
of guaranteed loans due to the Special Guarantee Program (Figure 3.6.). Imai also shows that this
SME category’s zombie ratio subsequently stayed above 12 percent through 2008, an outcome
101
that reflects the generally frail business conditions of this category of users which depend
significantly on bank loans for financing.
Drift in Japan’s Credit Guarantee System
Given the negative criticism surrounding zombie firms, the strong evidence implicating
Japan’s credit guarantee system in the creation of zombie SME begs the question: How could the
system’s effect on zombie formation among SME persist largely unchanged throughout the lost decades? One
hypothesis is that the system’s high level of protection, namely its large scale and generous
guarantee ratios, underwent drift. Conceptually, drift occurs when decision-makers do not
update formal rules when shifting developments change the social effects of the rules in ways that
are recognized by at least some political actors. In the case of Japan’s credit guarantee system,
drift could explain why the system’s substantial size and high guarantee ratios persisted through
the lost decades, despite a moderately positive economic climate and widespread concern that
the system posed a moral hazard in giving financial life support to firms with fundamentally poor
business prospects.
For the hypothesis of drift to be supported, evidence would need to show that its causal
mechanisms existed in the politics surrounding credit guarantee system and contributed to a
status quo bias in the system. Possible mechanisms for drift include the interaction between
decision-making institutions such the separation of powers that gives rise to veto players (Tsebelis
1995), limited public access to policymaking which may indirectly bolster the influence of well-
organized interests in shaping legislative proposals (Hacker, Pierson, and Thelen 2015), policy
feedback effects that create vested interests who subsequently rally to block change (Pierson 1996;
Levinson and Sachs 2015; Starr 2015), uncertainty about the possible consequences of policy
102
adjustment (Shepsle 1986), and collective action barriers to building pro-reform coalitions
(Hardin 1989).
Evidence suggests that many of these mechanisms were indeed present in the politics of
the credit guarantee system, hampering reform of the system and reinforcing the system’s
contribution to zombie SME creation during the lost decades. The only of these mechanisms that
was absent was the separation of powers. In Japan’s parliamentary system, the executive is
responsible to the legislature, namely the lower house of the Diet, and is not separately elected.
Moreover, the upper house has few powers to check the lower house’s decisions. Though the
upper house can vote down bills passed by the lower house and force the lower house to
overcome a two-thirds vote to pass the vetoed legislation, it has no power to block the lower
house on budgetary matters. Thus, even though there were multiple times in the period 1994-
2014 when Japan’s Diet was “twisted” (nejire) and different political parties controlled the upper
and lower houses—1997-1998, 1998-1999, 2007-2009, 2010-2012, and 2012-2013—there was
only modest potential for the separation of powers to stymie reform of the credit guarantee
system.
Multiple barriers other than the separation of powers obstructed potential change to the
system. For one, limited public access to the policymaking behind the system afforded a few well-
organized interests with stakes in system’s high guarantee ratios a substantial degree of influence
over the system’s structure. Government officials’ uncertainty about the potential effect of system
modification, as well as collective action challenges for those who wanted to pare down the
system, also played roles in keeping the credit guarantee system largely unaltered. This
combination of factors enabled the system to continue serving large numbers of financially fragile
firms and contribute to zombie SME creation throughout the 2000s, well after Japan’s banking
crisis had ended.
103
Concern about the ill effects of the credit guarantee system emerged almost immediately
after the Special Guarantee Program’s implementation in October 1998. In January 1999, the
widely-read Nikkei business newspaper reported that some program users filed for bankruptcy less
than one month after receiving guaranteed loans, or took loans with no purpose just because they
were available. In February 2000, the Nikkei Financial newspaper also reported that program users
made stock investments with guaranteed loans intended for company operations (Uesugi and
Sakai 2005, 6). Then, in November 2000, the Tokyo District Attorney’s Special Investigation
Department announced that several secretaries belonging to sitting Diet members from the
Komeito, Democratic Party and New Conservative Party had illegally taken commissions on
guaranteed loans for ineligible borrowers, and that at least 450 Diet members and local
assemblymen had intervened as brokers in more than 10,000 cases of guaranteed loans issued by
the Tokyo Credit Guarantee Corporation. At the time, this abuse of the credit guarantee system
was considered indicative of a “social problem” (shakai mondai) (Yamazaki Tai 2017).
Japanese economists were also quick to critique the credit guarantee system after the
initiation of the Special Guarantee Program. Matsuura and Takezawa (2001) argued that the
program’s guarantees did not in fact increase the supply of loans to SME, while Takezawa,
Matsuura and Hori (2004) contended that, after 1998, the widespread use of the credit guarantee
system and especially the Special Guarantee Program led to increasing defaults which in turn
reduced the loan supply for SME because of higher credit risk. Doubts about the system’s impact
on SME with healthy business conditions, as well as acknowledgement that the system required
substantial government outlays to cover the annual deficits of individual credit guarantee
corporations—which ranged between ¥200 billion and ¥600 billion in the period 1999-2004)—
spurred the SME Agency (SMEA), the government agency in charge of Japan’s SME policy and
104
the credit guarantee system, to convene a deliberative council (shingikai) about the future of the
system in late 2004 (Uesugi and Sakai 2005, 7).
21
On the surface, the deliberative council may have seemed like an opportunity for public
concern to effect change in the credit guarantee system’s structure and reduce its protectionism.
However, despite the semblance of openness to outside opinion, the most vocal attendees at the
council’s seven meetings between December 2004 and June 2005 were the system’s primary
stakeholders. Of the 80 opinions about the system which were formally recorded by the SMEA at
the conclusion of the council, 30 came from credit guarantee corporations, 20 came from
associations representing SME, 11 came from financial institutions, and 10 came from regional
governments. Only 9 opinions came from individual SME, scholars and citizens (Small and
Medium Enterprise Agency 2005, 1). Thus, it was mainly insiders’ ideas about the credit
guarantee system that the SMEA heard at the council as it weighed how to reform the system.
Weak public representation was not peculiar to this particular deliberative council, but consistent
with the common operation of deliberative councils throughout postwar Japan as venues for
interest groups to directly shape policymaking (Schwartz 1998).
Evidence suggests that SMEA officials were at least partly amenable to downsizing the
credit guarantee system’s largess when they initiated the council. One official then working in the
agency’s finance department remarked, “METI knew that we needed to change the 100 percent
guarantee system to the partial guarantee system maybe for 20 years—maybe from the 1980s,
METI wanted to change the market-skewing guarantee system,” indicating that key political
actors behind the system had considered possible alternatives to the system, and desired that a
21
A close observer of the credit guarantee system reported that SMEA officials began worrying about the system
after the inception of the Special Guarantee Program, wondering, “Was it was really okay?” (Hontou ni daijoubu ka?)
(Interview 32).
105
different arrangement be set in motion (Interview 59). However, the same official emphasized
that agency officials were simultaneously reluctant to reform the system because of possible
effects on SME, saying, “METI has a responsibility to maintain the economy, to grow the
economy. But they don’t want to have a huge, drastic impact to one of the specific sectors. So we
had that kind of system. We didn’t, we couldn’t want to change drastically. Maybe you can say
that it is kind of an ambivalent feeling, but it’s true. We would like to change the system
gradually, but we don’t want to have a huge impact on the SME” (Interview 59).
Aside from uncertainty about reform’s possible impact on SME, SMEA officials’ views on
adjusting the credit guarantee system were also tempered by their interest in amassing
bureaucratic power for themselves. In particular, within the agency’s finance department,
influence and esteem were partly defined by the ability to secure large annual budgets from the
Ministry of Finance (MOF) to pay for the system. Therefore, department officials typically strove
to obtain as much funding as possible from MOF in the short term to enhance the department’s
standing vis-à-vis other departments in the SMEA and METI—even coining a term for
“plundering” funds from MOF (bundoriai)—despite that their long-term goal of downscaling the
system was aligned with MOF (Interview 59).
Given SMEA officials’ ambivalence about system adjustment and internal power
motivations, key stakeholders in the credit guarantee system faced compromised bureaucratic
opposition when they voiced their preferences about the system at the deliberative council. Two
of the main players in the system, SME and financial institutions, were particularly cautious
about system reform and expressed conservatism about the degree of change they could accept.
On the council’s main issue of whether to reduce the system’s 100 percent guarantees,
associations representing SME criticized the council’s proposed introduction of a “responsibility-
sharing system” (sekinin kyouyu seido) that would lower certain guarantee ratios to levels where
106
banks would face loan loss risk, with one association saying, “While public sentiment that
financial institutions bear no risk is not incomprehensible, the influence [of a responsibility-
sharing system] on banks’ lending behavior will be clear, and we cannot support it” (Small and
Medium Enterprise Agency 2005, 14). Financial institutions also resisted the proposed change,
with one commenting that Japan should not introduce such a system just because other foreign
countries had done so, and another asking for more time—“at least two years”—for banks to
prepare for the details of such a system should it be realized (Small and Medium Enterprise
Agency 2005, 15).
These stakeholders’ unwillingness to allow radical change reflected their vested interests
in the credit guarantee system as it was then designed. As mentioned above, for many SME with
fundamentally precarious business conditions, especially microenterprises, the system represented
one of the few existing ways to obtain financing. Reduction in the system’s guarantee ratio thus
threatened to extinguish a key credit source, since heightened possibility for loan loss might
discourage banks from lending to weak customers. Because SME associations present at the
council like the National Federation of Small Business Associations (NFSBA) and the Central
Federation of Societies of Commerce and Industry (CFSCI) were pyramid organizations covering
tens of thousands of local associations and chambers at the prefectural and sub-prefectural levels
and collectively representing millions of SME, their opinions against credit guarantee system
reform at the council could not easily be ignored by SMEA officials (Interviews 35, 36).
Financial institutions were also beneficiaries of the credit guarantee system’s high
coverage ratios and thus disinclined toward change. Regional banks and smaller financial
institutions like credit cooperatives and credit unions were particularly strong backers of the
system, since their operations in the Japanese countryside meant that their loan portfolios
107
consisted largely of loans to SME, and loans to SME carried higher interest rates.
22
Furthermore, regional markets faced significant demographic decline and stagnant economic
growth, which intensified competition over the dwindling pool of remaining SME borrowers and
created profit pressures that made loan guarantees an attractive device for maintaining if not
improving lenders’ bottom lines (Shimizu 2008, 128–31). Surveys conducted by the FSA showed
that 70 percent of SME using credit guarantees were recommended to do so by their banks
(Interview 28), suggesting that banks appreciated the system’s elimination of risk and thus had an
interest in preserving the status quo.
While these two key stakeholders communicated their preferences about system reform to
the SMEA through opinions formally issued at the deliberative council, it is also reasonable to
believe that they advocated to agency officials in ways that were less observable to other council
attendees and the wider Japanese public. For example, when asked about how the CFSCI gives
policy recommendations to the SMEA, a representative from the association said that, besides
making paper reports, “on detailed matters, it makes a phone call or sends an email. Basically, it
talks to them directly” (Interview 36). A NFSBA representative also mentioned having “close
information exchange” (missetsu na jouhou koukan) with METI, implying that that the association’s
leadership was familiar with ministry officials to a degree that transcended contact at the
deliberative council meetings (Interview 35). In other words, while SMEA opinion records show
that central players in the credit guarantee system had conservative attitudes about adjusting the
22
Data from after the deliberative council suggest the critical importance of SME to regional banks’ business
operations. In the fiscal year ending in March 2012, only three out of 116 Japanese banks had less than 50 percent of
their loan portfolios filled by loans to SME—Mizuho Corporate Bank (37.3 percent), Mitsubishi UFJ (46.8 percent),
and Sumitomo Trust Bank (47.4 percent)—and these were all large banks. By contrast, for regional banks such as
Suruga Bank (95.2 percent), Taisho Bank (93.1 percent), Kinki Osaka Bank (92.7 percent), Shizuoka Chuo Bank
(92.2 percent), Minami Nihon Bank (92.2 percent), Kansai Urban Bank (92.0 percent) and Awa Bank (90.1 percent),
SME constituted more than 90 percent of their loan portfolios (Tokyo Shoko Research 2012).
108
system, they do not necessarily convey the full extent to which these powerful players lobbied
SMEA officials to maintain the status quo.
Records from the deliberative council also omit a key group of actors who likely
influenced officials’ thinking about amending the system behind the scenes—politicians. While
politicians’ election campaign manifestos did not touch on the credit guarantee system and few if
any politicians took strong public stances on SME policy (Interviews 36, 35), politicians were still
uniform in their underlying support for SME, irrespective of political party (Interviews 32, 59).
This was particularly true of politicians at the prefectural and municipal levels, where
multimember electoral districts with narrower electoral margins heightened the significance of
SME as a reliable voting bloc that could be mobilized on politicians’ behalf (Shimizu 2013, 155).
Though proof of politicians’ direct involvement in the deliberative council is hard to
come by, other evidence suggests how politicians influenced the credit guarantee system to
benefit SME. For example, around the time of the deliberative council in 2005, one of the main
tasks inside the SMEA finance department was reportedly an activity called “window
introduction” (madoguchi shokai) whereby department personnel would take phone calls from
politicians who wanted the agency to help certain SME in the politicians’ home districts.
Ostensibly, the politicians’ calls were simply to request the location of nearest GFI, but in truth
the calls were intended to make agency officials then contact the GFI and intimate that the SME
could use favorable treatment in their loan applications. Politicians’ appeals for window
introductions were so frequent that the division assigned a dedicated chief clerk to handle them
(Interview 59). It is not unreasonable to assume that politicians making these frequent calls also
impressed upon agency officials the importance of the credit guarantee system to their local
SME, thus affecting the officials’ thinking about how far system revision could go.
109
A more telling indicator of politicians’ general support for the credit guarantee system was
the formation of the system’s budgets. Since MOF opposed the system’s 100 percent guarantee
under the belief that it undermined banks’ monitoring of borrowers, it strongly pushed back
against SMEA officials when they made their annual budget request of the ministry to
compensate the system’s deficits, which caused the system’s primary budgets to be generally
small. However, the system was also allocated a supplementary budget each year, and the process
for making this latter budget was notably lenient. Because the supplementary budget was made
according to a shorter schedule and often timed to immediately follow the installation of a new
cabinet to maximize political appeal, MOF found it hard to deny large outlays for the credit
guarantee system through this other budget. Moreover, because MOF knew that it would
encounter politicians’ anger if it disallowed substantial funding for the system, it reportedly asked
SMEA officials to propose large budget “balls” (tama o motte koi) for the supplementary budget
which it would then summarily authorize. Agency officials were fully aware of these constraints
on MOF around the time of the supplementary budget, and were confident that they could
exploit them to obtain as much funding as necessary cover for the credit guarantee system. If
MOF did not comply, the officials would simply ask politicians to pressure the ministry until it
conceded (Interview 59). Table 3.3. shows that the annual supplementary budget for the credit
guarantee system was regularly larger than the primary budget until 2012, suggesting the
constancy of politicians’ support for the system as a means to help SME.
Against this combination of ambivalent SMEA officials and status quo-favoring interest
groups and politicians, the voices who unequivocally opposed the system’s high-level guarantees
were the minority at the deliberative council. One individual opinion did point toward the cost of
the system from the public’s perspective, saying, “The issue of conserving tax money is an
110
Table 3.3. Central Government Budget Allocations to the Credit Guarantee System, 1999-2015
Source: SME Policy Council Finance Working Group (2015, 27)
Unit: ¥100 million
extremely important subject given the current situation in which there is overbanking toward
SME and reducing the national debt is an urgent duty,” while another individual opinion drew
attention to the opacity of the council’s own assessment of the system, saying, “It is a problem of
the system that there is no transparency around who is evaluating the system or how it is being
evaluated. If it is ultimately supposed to be the peoples’ decision, one would think that opinion
Year
Primary Budget
Additional
Budget
Supplementary
Budget 1
Supplementary
Budget 2
Supplementary
Budget 3
Supplementary
Budget 4
Total Budget
Allocation
1999 225 4205 4430
2000 221 6066 6287
2001 301 1625 1926
2002 332 4195 4527
2003 434 592 1026
2004 434 3279 3713
2005 434 522 956
2006 419 580 999
2007 378 2207 2585
2008 454 4000 3891 8345
2009 579 11236 9783 21598
2010 698 330 5335 6363
2011 852 3209 3703 4011 11775
2012 1066 951 500 2517
2013 678 452 1130
2014 678 597 1275
2015 681 681
111
Figure 3.7. Public Opinion on Issues Needing Government Action
Source: Government of Japan Cabinet Office (Multiple years)
Unit: Percentage of Respondents Saying Issue Needs Government Action
would overwhelmingly favor substantial downsizing of the system” (Small and Medium
Enterprise Agency 2005, 19, 18). However, such critical views were the exception at what was
largely a forum for system insiders to reaffirm their interest in minimizing potential change.
Ironically, there were some grounds for believing that ordinary citizens might have
spoken up against the scale of the credit guarantee system had they attended the deliberative
council. Annual public opinion surveys conducted by the Cabinet Office indicate that the
Japanese populace put little priority on government action on SME policy throughout the 2000s,
implying that the public probably had little interest in keeping the system’s substantial scale and
guarantee ratios (Figure 3.7.). Public opinion polls by the Asahi Shimbun newspaper in the early
years of the Koizumi administration also found that popular support for structural reforms
remained above 50 percent (Kume 2009, 245), further suggesting that citizens might have
favored revision of the credit guarantee system as they did Koizumi’s other neoliberal reforms.
0
10
20
30
40
50
60
70
80
1996
1999
2002
2005
2008
2011
2014
2017
Commodity Prices Economic Policy
Medical, Welfare, and Pension Issues Taxes
SME Policy
112
On the other hand, it is questionable whether the Japanese public really knew enough
about the credit guarantee system to want reform. One possible explanation for weak public
demand for government action on SME policy—if SME policy includes the credit guarantee
system—may have been that the system’s insurance structure diffused potential local opposition.
Because losses by individual credit guarantee corporations were insured by the central
government through the JFC, their cost was spread out across taxpayers nationwide, rather than
concentrated among the citizens in the jurisdictions where the losses occurred (Shimizu 2008,
76). A related possible explanation is that the system’s overall structure was simply too complex
for many people to understand, with the government’s role masked by the subrogation process
that, on the surface, made it seem like the system consisted mainly of banks and individual credit
guarantee corporations, rather than the central government with some support from local
governments (Interviews 34, 59).
Given that Japanese citizens showed little interest in protectionism toward SME yet were
not vocal about the credit guarantee system, the handful of academic economists who criticized
the system’s high guarantee ratios in the early 2000s had few allies in pushing for substantial
modification of the system at the deliberative council. It thus fell largely to SMEA officials to
decide on the kind of reform that would address the troubling questions raised by the system and
especially the Special Guarantee Program—the same officials who did not want to impose drastic
change upon SME, upset the stakeholders who were strongly supportive of the status quo, or
offend politicians from every political party. Balancing these conflicting imperatives that leaned
mostly towards stasis in the system, agency officials ultimately announced that a responsibility-
sharing system that would lower the system’s primary guarantee ratio from 100 to 80 percent
would come online in October 2007—a date that was two years after the deliberative council, in
alignment with the request from a financial institution representative at the council.
113
While the responsibility-sharing system portended to sharpen banks’ incentive to monitor
guarantee users and screen out SME prone to default on loan repayment—and thus possibly
reduce the number of zombie SME in the system—100 percent guarantees were still granted to
several large categories of system users after the reform, thereby mitigating its effects on
removing moral hazard. These categories included microenterprises (most system users), SME
that were affected by natural disasters, SME that were startups, and SME that qualified for the
system’s various specific safety-net programs. One of these safety-net guarantees in particular,
Safety Net 5 for firms in declining industries, continued to have conspicuously high subscription
in future years, with more than 2 million SME users between 2006 and 2015 (SME Policy
Council Finance Working Group 2016, 12). In short, though responsibility-sharing signified a
shift away from the prior system’s total elimination of risk in banks’ lending to SME, numerous
exempted categories of users perpetuated the system’s agency problem for banks. Later studies
showed that the 80 percent guarantee in the responsibility-sharing system did not entirely remove
moral hazard either, meaning that the reform’s impact was modest at best (Saito and Tsuruta
2014).
Less than a year after the responsibility-sharing system was unveiled, Japan was hit by the
global financial crisis. As one would expect, in response to the crisis, Japanese politicians rallied
to the cause of SME with various assistance measures, including enhancements to the credit
guarantee system. On October 31, 2008, an “Emergency Guarantee System” (Kinkyu hosho seido)
was established to provide 100 percent guarantees to SME in designated industries whose
average sales had declined by 3 percent from the previous year. Under this new system and the
existing Safety Net Number 5 program, new loan guarantees surged in fiscal year 2008, climbing
to approximately ¥20 trillion and putting the total balance of loan guarantees at the end of the
year at ¥34 trillion (Yamori et al. 2013, 74). Statistics from the National Federation of Credit
114
Guarantee Corporations, the peak organization for Japan’s individual credit guarantee
corporations, show that more than half of the SME using the Emergency Guarantee System were
users of the credit guarantee system prior to taking advantage of the new system, demonstrating
that many SME were already weak and dependent on credit guarantees before the crisis struck.
Moreover, many of these firms would have likely qualified for the Emergency Guarantee System
even had the crisis never happened, since their financial precariousness meant they were highly
prone to sales swings exceeding the minimalistic qualifying threshold of 3 percent (Yamori 2011,
22–23).
Though the Emergency Guarantee System ended in March 2011, the popularity of the
remaining Safety Net Number 5 program was such that loans subject to 100 percent guarantees
through the program were a larger portion of the overall stock of outstanding guaranteed loans
than the loans issued through the responsibility-sharing system until 2013 (SME Policy Council
Finance Working Group 2016, 27). Given the widespread use of 100 percent guarantees after the
global financial crisis and the fact that even the responsibility-sharing system’s 80 percent
guarantees produced some degree of moral hazard, it is reasonable to conclude that the credit
guarantee system continued to play a large role in zombie SME creation well into the 2010s.
Conclusion
This chapter examined zombie firms among Japanese SME during the period 1994-2014.
Starting from the intriguing statistical finding that the zombie ratio among SME remained
consistently high throughout this period, the chapter speculated that a distinct causal mechanism
underpinned zombie SME which was absent from listed firms where, as described in Chapter 2,
the zombie ratio declined in the early 2000s because of transparency-enhancing regulatory
115
changes and a new bankruptcy law. Specifically, the chapter investigated the possibility that
extraordinary assurances given to private banks via a particular policy institution—the credit
guarantee system—contributed to the formation of zombie SME.
While the enlargement of Japan’s credit guarantee system was initially conceived as a
crisis countermeasure for the credit crunch in the late 1990s, the continuation of the system’s
generous scale and guarantee ratios throughout the 2000s, which served many fundamentally
weak SME with high potential for loan default, was puzzling from the perspective of Japan’s
generally positive macroeconomic environment, as well as the perspective of theory suggesting
that electoral reform in the 1990s should have reduced protectionism by the government. As a
possible explanation for why the system remained largely unchanged and persisted in protecting
inefficient SME likely to become zombies even after Japan generally recovered from the banking
crisis and had more majoritarian rules in the lower house of the Diet, the chapter explored the
hypothesis that the system underwent drift.
The chapter found evidence largely consistent with this hypothesis. In particular, limited
public participation in a deliberative council where system reform was discussed in the mid-2000s
enabled key system stakeholders, especially SME associations and regional financial institutions,
to influence how much adjustment was possible. Though government officials in charge of the
system had long wanted to reduce the system’s guarantee ratios, these stakeholders’ strong
opposition, as well as the officials’ own uncertainty about the possible effects of reform and
constant informal pressure from politicians, impaired revision to the ratios that would have
caused banks to accept significant risk when lending to SME and thereby incentivized the banks
to screen out weak guarantee applicants. Academic economists were generally critical of the
system at the time of the deliberative council, but their outspokenness was unmatched by
116
Japanese taxpayers, who, despite placing low priority on SME policy, appeared either indifferent
or unaware toward the system’s substantial costs and were not well represented at the council.
Facing little organized opposition, officials subsequently implemented a responsibility-
sharing system whose effect on changing banks’ lending behavior toward weak SME was modest.
Furthermore, the credit guarantee system was expanded again during the global financial crisis,
and mostly functioned to serve habitual systems users that were chronically weak SME instead of
healthy SME that temporarily suffered from the crisis. In this way, the highly protectionist
character of the system was maintained well into the second decade of the 21
st
century,
prolonging Japan’s encounter with zombie firms.
While listed firms have traditionally received the most attention in the discussion of
Japan’s zombie firms, this chapter showed that SME have been a more important firm category
for containing substantial numbers of zombies during the lost decades, and that extraordinary
credit assurance, upheld over time by the political mechanism of policy drift, may be a crucial
process behind the creation of zombie firms, in addition to the mechanisms of transparency and
bankruptcy law examined in the previous chapter. In the next chapter, these three mechanisms
drawn from exploration of Japanese listed firms and SME will be investigated in a non-Japanese
case following the global financial crisis, to see how well they explain the rise of zombie firms in a
nationally and temporally distinct context where economics research suggests zombies have
existed to a substantial degree similar to Japan. For this task, the dissertation turns to China,
where concern about zombie firms has exploded in recent years and there has been widespread
interest in resolving firms with this troubling status.
117
CHAPTER 4.
Zombie Firms among Chinese Firms, 1998-2015
This chapter investigates the case of zombie firms in China in the period 1998-2015.
While Japan’s lost decades initially popularized the problem of zombie firms and inspired
research on their economic effects, arguably no recent case of zombies has attracted as much
attention as China. Zombie firms first received official recognition in China in October 2015,
when National Development and Reform Commission vice chairman Liu He singled out the
elimination of zombies as a priority for economic reform (Zhongguo Xinwen Wang 2015). The
following month, at the fifth plenary session of the 18
th
Chinese Communist Party (CCP) central
committee, Premier Li Keqiang argued that the party needed to “accelerate the restructuring or
market exit of zombie firms.” Soon thereafter, President Xi Jinping, at a financial work leading
group meeting, said that China should “promote the effective elimination of excess capacity, and
promote industrial restructuring and reorganization,” implicating zombies as a key reform target
(Ding 2015). Since these authoritative statements by Chinese political leaders, media coverage of
China’s zombie firms, especially zombie state-owned enterprises (SOE), has exploded
domestically and internationally (Li 2015; Du 2016; Zhongzheng Wang 2016; Zhu and Hua
2016; Schuman 2015; Wildau 2016b; Bloomberg 2017). Zombies have also rapidly emerged as a
scholarly topic in Chinese academia, particularly in the fields of economics and finance.
23
23
A search of major Chinese financial media for articles with “zombie firm” (jiangshi qiye) as a subject term (zhuti) in
the CNKI database on October 20, 2016, revealed rapidly growing journalistic coverage after the central
government turned to zombie firms as a policy issue in late 2015. For example, in the newspaper Diyi Caijing Ribao
(China Business News), the number of articles about zombie firms increased quickly in 2016: 2012 (1), 2013 (1), 2014
(2), 2015 (5), and 2016 (22). Similar jumps in 2016 were seen in other leading newspapers: in Jingji Cancao Bao
(Economic Information Daily), the yearly total was 2012 (1), 2013 (0), 2014 (2), 2015 (11), and 2016 (41); in
Guangming Ribao (Guangming Daily), the yearly total was 2015 (2) and 2016 (14); in Renmin Ribao (People’s Daily), the
118
This chapter shows that zombie firms have indeed been a considerable presence in China
over the past two decades and in recent times since the global financial crisis. However, it stresses
that the general trajectory of the zombie ratio in China has been in a downward direction,
notwithstanding the government’s large stimulus after 2008 that increased zombie firms in
certain industries and particularly among SOE. Moreover, China’s zombie firm story should be
understood as encompassing substantial variation, with particular corporate categories and
geographic regions having much higher concentrations of zombies than others.
The chapter is structured in three parts. First, it presents descriptive statistics on zombie
firms in China. Spanning the period 1998-2015, these statistics illustrate the zombie firm ratio—
the outcome variable of interest—as well as certain attributes of these firms, including their
presence across different industrial sectors, their geographical dispersion, and their ownership
type. These statistics are sourced from a recent Chinese report by Nie et al. (2016b) analyzing
two large samples of firms from the Listed Company Database (Shangshi gongsi shujuku) and the
Chinese Industrial Enterprise Database (Zhongguo gongye qiye shujuku).
Second, the chapter examines potential explanations for China’s zombie firms suggested
by the dissertation’s three orienting hypotheses and case studies on listed firms and SME in
Japan. It finds little evidence that two of the hypotheses for Japan’s zombie firms, transparency
and policy drift, have explanatory power in the case of China. However, for the orienting
hypothesis of bankruptcy law, the chapter identifies an unconducive bankruptcy framework as a
likely obstacle for early restructuring by distressed Chinese firms. Challenges in China’s
yearly total was 2015 (5) and 2016 (39). A search of Chinese academic articles with “zombie firm” as a key word
(guanjianzi) in the CNKI Scholar database on July 31, 2017, showed only two scholarly articles about zombie firms in
2015. However, the number climbed to 45 articles in 2016, and reached 34 articles by the end of July 2017.
119
bankruptcy framework, particularly for SOE, are likely one zombie-inducing condition that
China in the early 21
st
century has shared with Japan during its first lost decade in the 1990s.
Third, due to the limited explanatory ability of the dissertation’s original orienting
hypotheses for China’s zombie firms, the chapter explores an additional orienting hypothesis
related to soft budget constraint, a political economy condition traditionally associated with
socialist systems in transition. The chapter finds general support for this hypothesis with regard to
China’s zombie firms, particularly among SOE where the zombie ratio remained higher than the
ratio among private firms.
Statistics on Zombie Firms among Chinese Firms
Research on zombie firms in China has largely come on the heels of Chinese leaders’
stated goal in late 2015 to adjust “supply-side” policy and implement structural reforms. This
goal was captured in the slogan “three cuts, one lowering, and one improvement” (san qu yi xiang
yi bu): cut excess industrial capacity, cut the stock of real estate, cut risky financial leveraging,
lower corporate costs, and improve effective supply. Zombie firms were explicitly associated with
this goal, making academic studies on them immediately topical. Research was also motivated by
the reality that the government itself was unclear about the actual extent of the problem and how
to solve it, despite its own publicized intention to address zombie firms (Interview 53).
Among the recent wave of analyses on zombie firms in China, a report by Nie et al.
(2016b) is noteworthy for its longitudinal large-n data and its general conceptual alignment with
the FN definition of zombie firms. Encompassing the period 1998-2015, the report utilizes two
firm-level databases, the Listed Company Database and the Chinese Industrial Enterprise
Database, to estimate the prevalence of zombies among listed firms and unlisted manufacturing
120
firms. The former database primarily contains large firms with higher average profit margins and
total assets, and has a substantial percentage of SOE and collective enterprises (approximately 35
percent in 2013). The latter database covers relatively smaller firms with lower profitability, lower
total assets, and lower propensity to be state-owned (approximately 2.75 percent in 2013). The
databases’ complementary qualities give the report’s findings a considerable degree of
representativeness, particularly since the Chinese Industrial Enterprise Database covers 792,267
firms and 3,636,479 total observations during the period 1998-2013 (excluding the year 2010),
enabling the report to represent 90 percent of the total sales of China’s industrial firms (Nie et al.
2016b, 20).
Beyond having a long chronological span and a large number of observations, the report
is notable for the similarity of its conceptualization of zombie firms with studies of zombies in
Japan using the FN definition. The report presents three different zombie firm definitions—the
CHK definition, the FN definition, and the “official” definition announced by Premier Li in
December 2015—before proposing a fourth definition called the “Renmin University National
Academic of Development and Strategy” definition.
24
This latter definition is almost identical to
the FN definition, with the exception that it requires a distressed firm to have two years of FN
zombie status before it is considered a zombie firm in the second year. The report contends that
this revised definition is an improvement over the original FN definition, which risks mislabeling
fundamentally healthy firms that suffer from short-term shocks as zombies (Nie et al. 2016b, 13–
17).
24
Li defined zombie firms as firms that operate at a loss for three consecutive years and fail to make structural
adjustments such as asset reorganization, property right transfers, or closure through bankruptcy. However, as noted
in Chapter 1 (Footnote 1), the extent to which Li’s definition is official is questionable.
121
Figure 4.1. Chinese Zombie Firms by Year, Listed Firms Database, 2000-2015
Source: Nie et al. (2016b, 21)
With these data and conceptual strengths, the report provides findings about zombie
firms in China that are at once highly representative, and generally comparable to studies of
zombie firms elsewhere.
25
On the main question of the presence of zombie firms in China’s
economy, the report finds that, among listed firms, the number of zombies steadily increased
from 2001 until peaking in 2013. After 2013, the number of listed zombies slightly declined
(Figure 4.1.). However, for the 12-year period between 2003 and 2015, the zombie ratio
generally remained around 13 percent, meaning that the growth in the number of zombies was
proportional to the expanding number of listed firms in the period 2000-2015, rather than the
result of change in listed firms’ business conditions (Nie et al. 2016b, 22).
25
Nevertheless, it bears reiteration that Nie et al.’s use of a modified FN definition produces estimates that are not
directly comparable to the earlier chapters’ statistical analysis of the zombie firm situation in Japan.
122
Figure 4.2. Chinese Zombie Firms by Year, Industrial Enterprise Database, 2000-2013
Source: Nie et al. (2016b, 22)
By contrast, among unlisted industrial firms, the number of zombie firms fell from a high
of over 40,000 firms (roughly 27 percent of firms) in 2000 to approximately 23,000 firms (less
than 10 percent of firms) in 2004, before holding at an average of 7.51 percent of firms during
the period 2005-2013 (Figure 4.2.). The sharp drops in the number of zombie firms and the
zombie ratio in the early 2000s signify that China’s worst encounter with zombies happened at
the start of the 21
st
century (Nie et al. 2016b, 23–24). In other words, while the recent spotlight
on zombie firms in China is not without reason, the historical record suggests that the peak of the
China’s zombie problem was more than a decade ago.
Besides offering a big picture view of the stability of zombie listed firms and the decline of
zombie unlisted firms since 2000, the report analyzes zombies at the industry and provincial
levels. In 2013, among listed firms—the firm category in which zombies have remained more of
a problem and SOE are relatively more common—the industries with the highest zombie
123
Table 4.1. Chinese Zombie Firms by Industry, Listed Firms Database, 2013
Industry name
Number of
firms
Number of
zombie firms
Zombie firm
ratio
Banking 16 0 0.00%
Media 97 4 4.12%
Non-bank finance 43 2 4.65%
Computer 153 8 5.23%
Leisure service 34 2 5.88%
Electronics 158 11 6.96%
Textile and apparel 78 6 7.69%
Communication 65 5 7.69%
Agriculture, forestry, animal husbandry and fisheries 85 7 8.24%
Food and drink 78 7 8.97%
Household appliances 60 6 10.00%
Electrical equipment 160 17 10.63%
Equipment 270 29 10.74%
Chemical industry 259 28 10.81%
Building materials 72 8 11.11%
Medical creatures 217 26 11.98%
Light industry manufacturing 98 12 12.24%
National defense 34 5 14.71%
Automobile 124 19 15.32%
Nonferrous metals 106 18 16.98%
Mining 57 10 17.54%
Transportation 91 16 17.58%
Utilities 122 25 20.49%
Integrated 41 9 21.95%
Commercial trade 90 26 28.89%
Building and decoration 85 27 31.76%
Real estate 137 61 44.53%
Steel 35 18 51.43%
Total 2865 412 14.38%
Total (Industrial Firms) 2184 294 13.46%
Source: Nie et al. (2016b, 26–27)
concentrations were steel (51.43 percent), real estate (44.53 percent), building and decoration
(31.76 percent), and commercial trade (28.89 percent). The lowest zombie firm concentrations
were in banking (0.00 percent), media (4.12 percent), non-bank finance (4.65 percent), and
computer (5.23 percent) (Table 4.1.) (Nie et al. 2016b, 24–26).
124
Figure 4.3. Chinese Zombie Firms by Province, 2000-2013
Source: Nie et al. (2016b, 28)
At the provincial level, wealthy coastal provinces such as Shandong, Jiangsu, Zhejiang,
and Guangdong have had the highest numbers of zombie firms, whereas poor interior provinces
like Qinghai, Xizang, Gansu, and Guizhou have had the lowest numbers. In terms of the zombie
ratio, however, the situation has been the opposite: more developed eastern and southern
provinces have had lower zombie firm ratios, while less developed southwestern, northwestern,
and northeastern provinces have had higher ratios despite their lower numbers of zombie firms
(Figure 4.3.). Moreover, central and southwest provinces had declining zombie firm ratios over
125
Figure 4.4. Zombie Firms among Chinese Firms by Ownership Type, 2000-2013
Source: Nie et al. (2016b, 29)
the two periods 2000-2004 and 2005-2013, but western provinces had a higher ratio in the more
recent period. In the more recent period, the provinces with the highest zombie ratios were the
western provinces of Ningxia (17.06 percent), Shanxi (15.31 percent), and Gansu (15.09 percent),
while the provinces with the lowest ratios were either rural provinces like Xizang (3.61 percent),
or relatively more developed provinces like Henan (4.23 percent), Shandong (4.29 percent),
Hunan (4.44 percent), and Fujian (4.68 percent) (Nie et al. 2016b, 27).
Beyond analyzing zombies at the industry and provincial levels, the report investigates
several other zombie firm characteristics. It finds that zombie status was more prevalent among
firms in its sample that are state-owned (Figure 4.4.), large-sized, and older (Nie et al. 2016b, 29–
31). Furthermore, the 4 trillion-yuan ($586 billion) stimulus issued by the Chinese government
126
Figure 4.5. Profitability and Debt Ratios among Chinese SOE and Collective Enterprises, 1998-
2013
Source: Nie et al. (2016b, 47)
after the global financial crisis was associated with increases in the zombie ratio among specific
industries such as coal, steel, and cement with high SOE presence. Also, in the period 2011-2013
when the zombie ratio rallied in these industries, the producer price index fell, implying that an
inverse relationship occurred between the concentration of zombie firms and the average selling
price received by goods producers (Nie et al. 2016b, 33–36). In other words, the Chinese
government’s response to the crisis not only produced higher ratios of zombie firms, but resulted
in a congestion effect as indicated by simultaneous industry-specific price decreases due to
overproduction by zombie firms.
Finally, the report examines the debt ratios (liabilities to total assets) and profit margins of
SOE and collective enterprises, and private enterprises. It finds that the debt ratio for SOE and
collective enterprises increased after 2008, while profit margins declined (Figure 4.5.). By
contrast, for private firms, the debt ratio decreased and profit margins were generally stable
127
Figure 4.6. Profitability and Debt Ratios among Chinese Private Firms, 1998-2013
Source: Nie et al. (2016b, 47)
(Figure 4.6.). These outcomes show that, after the global financial crisis, SOE and collective
enterprises found it easy to obtain loans despite declining profitability, whereas private
enterprises had difficulty getting loans despite steady profitability (Nie et al. 2016b, 46).
Japanization Redux: Japanese Explanations for Chinese Zombies
Given the increase in zombie firms in China after the global financial crisis, it is
understandable that many observers have drawn comparisons between contemporary China and
Japan during the lost decades (Orlik 2011; Guilford 2014; Lewis, Mitchell, and Yang 2017).
Chinese government officials have also been keenly interested in knowing lessons from Japan,
due to its experience with an asset price bubble and collapsed real estate market, an aging
society, and relational banking and collateral-based loans that have equivalents in China’s
banking system (Interview 53). For these reasons, examining the dissertation’s orienting
128
hypotheses for zombie firms in the context of China is not only a valuable intellectual exercise
from standpoint of assessing the hypotheses’ external validity in a non-Japanese case. It also
meaningful in speaking to an ongoing policy concern about political conditions that may be
causing China to reprise one of the notorious features of Japanization—the sustainment of
zombie firms—and harm its own economic growth.
The following analysis investigates each of the orienting hypotheses to uncover the extent
to which they explain the rise of zombies in China.
Orienting Hypothesis 1. Transparency
At first blush, China and Japan appear generally similar with regard to their dominant
one-party systems. As examined in Chapter 2, the LDP’s stable long-term rule was a key
structural reason why Japanese politicians failed to develop independent oversight of the
domestic banking sector, an outcome that had serious consequences for politicians’ ability to gain
regulatory information about the sector’s troubles and ultimately tackle the debt at Japanese
banks once the bubble imploded in the early 1990s. Therefore, on the surface, the CCP’s
unrivaled control of government might seem to satisfy the condition proposed by the hypothesis
that single party dominance leads to significant policymaking delegation to bureaucrats and puts
politicians at risk of lacking clarity about regulatory implementation.
However, the case study of listed firms in Japan revealed that protracted single party
dominance only matters for zombie formation to the extent that banks themselves have
capitalization levels near the required minimum, and are thus incentivized to engage in
“unnatural selection” and offer financial life support to weak borrowers to disguise the banks’
own vulnerability. Absent high levels of “bad” debt in the first place, banks have less reason from
129
a business perspective to offer credit assistance toward distressed firms and thereby turn them
into zombies. The alignment of structural power between politicians and bureaucrats may give
bureaucrats an upper hand in covering for banks’ forbearance behavior, but it alone is
insufficient to cause banks to give birth to zombie firms.
For China’s zombie firms to be explained by the transparency hypothesis, evidence would
need to show both that the Chinese politicians entrusted substantial discretion in financial
policymaking to bureaucrats as a result of the CCP’s stable control of government, and that
Chinese banks have had the incentive to support zombie firms because of low capitalization. In
the case of China, arguably neither of these conditions has been present. On the one hand,
though China has an official legislature in the form of the National People’s Congress (NPC)
nominally invested with lawmaking powers, the authoritarian nature of CCP rule means that
NPC delegates do not autonomously determine policies and delegate policymaking capability to
bureaucrats in the executive branch. Instead, delegates act as the quintessential rubber stamp
(Truex 2014). While NPC delegates may internally deliberate policies, actual policy decisions are
made by higher-level CCP bodies or in the State Council (Tanner 1999), and implemented by
bureaucrats who, despite having far greater autonomy and ability to act entrepreneurially than in
the pre-reform era of Mao Zedong, must still ultimately heed the CCP’s “fragmented
authoritarianism” (Lieberthal 1992). Thus, Chinese legislators have never been in a position to
delegate the kind of discretion which Japanese bureaucrats in MOF enjoyed in banking
regulation, and Chinese bureaucrats have never possessed an equivalent degree of policymaking
authority vis-à-vis the CCP to develop an autonomous regulatory approach toward local banks.
Second, official data on Chinese banks suggests that debt levels have not been as serious
as they were in Japan’s first lost decade, when low capitalization and the related concerns of loan
loss, equity loss and potential bank failure influenced Japanese banks’ decision calculus toward
130
Figure 4.7. Bank Nonperforming Loans to Total Gross Loans, 2010-2015
Source: International Monetary Fund Global Financial Stability Report (2016a)
Unit: Percentage
lending to distressed firms in jeopardy of taking zombie status. According to figures announced
by the China Banking Regulatory Commission, commercial bank nonperforming loans in China
rose to an 11-year high of 1.4 trillion yuan, or 1.75 percent of total bank lending, in May 2016
(Figure 4.7.). Even though so-called “special mention” loans at risk of falling into the
nonperforming loan category exceeded 4 percent of the total loan volume for commercial banks
at the end of March 2016 (Reuters Staff 2016), the cumulative total of these figures was far lower
than any year in the period 1998-2008 in Japan, including in the mid-2000s after worst of
Japan’s banking crisis had already passed (Figure 2.1.).
Though debt levels in Chinese banks have nominally been small compared to the levels in
Japanese banks the 1990s, there has nevertheless been worry about a potential debt crisis due to
doubts about the credibility of official Chinese data. For example, based on analysis of individual
corporate balance sheets, the IMF’s April 2016 Global Financial Stability Report estimated that
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
2010 2011 2012 2013 2014 2015
131
15.5 percent of total commercial bank loans to China’s corporate sector ($1.3 trillion) could be
considered nonperforming loans (International Monetary Fund 2016a, 16). A May 2016 report
by the brokerage CLSA put the number even higher, claiming that 15 to 19 percent of loans
were nonperforming (Reuters Staff 2016).
While these nonofficial estimates are ominous, it is noteworthy that some Chinese banks
have also been aggressively disposing of nonperforming loans. For instance, in the first half of
2015, Citic Bank and the Bank of Communications wrote off 75 and 30 percent, respectively, of
their year-end 2014 nonperforming loan balances. Also, the average provision coverage ratio at
Chinese banks has been approximately 150 percent of reported nonperforming loans (Lardy
2016), a situation that stands in contrast to Japan during the first lost decade when banks
underestimated provisioning by 30 to 50 percent (Fujii and Kawai 2010, 6). Moreover, Chinese
banks have decreased their exposure to SOE from 60 to 30 percent of total loans since the 1990s,
and now increasingly lend to private firms and households that provide higher returns on assets
and have relatively strong balance sheets (Lardy 2016).
These facts do not necessarily mitigate the implications of possible inaccuracies in official
data on Chinese banks’ lending behavior. But based on what is publically known about the
health of Chinese banks, it is difficult to conclude that Chinese banks’ solvency levels have clearly
incentivized them to engage in the same sort of forbearance lending that Japanese banks did
from the early 1990s until the start of the 21st century when their health was highly precarious.
In short, there is no unambiguous evidence that this second necessary condition for the
transparency hypothesis has existed in China and contributed to the creation of zombie firms.
132
Orienting Hypothesis 2. Bankruptcy
In addition to the transparency hypothesis, the dissertation examined a second orienting
hypothesis for zombie firms among listed firms in Japan’s first lost decade—bankruptcy law. To
recall, this hypothesis predicts that the incentive structure inherent in a country’s bankruptcy
framework should affect impaired firms’ willingness to pursue restructuring at an early date as an
alternative to possibly requesting credit assistance and taking zombie status. In the case of
Japanese listed firms, the dissertation found that the existing bankruptcy regime in the 1990s was
highly unappealing for both distressed firms and the firms’ managers, which possibly explains
why few troubled firms took legal routes to restructuring during the first lost decade, and thus
why firms accepted financial relief and zombie status as a less-worse outcome, even though
protracted indebtedness put them at risk of having to make costlier business decisions down the
line. This situation was significantly alleviated with the introduction of the Civil Rehabilitation
Law in April 2000, which facilitated firms’ entry into bankruptcy proceedings due to certain
favorable conditions like the debtor-in-possession principle, and coincided with the decrease in
zombie listed firms from the early 2000s.
Evidence suggests that the character of bankruptcy law in China has had a similarly
dissuasive effect on firms’ choices regarding restructuring as it did Japan until the Civil
Rehabilitation Law. Bankruptcy law in China dates to 1986, when noted privatization advocate
Cao Siyuan drafted the Enterprise Bankrutpcy Law (EBL) and won support for its passage in the
NPC (Lawrence 2000, 45). Aimed specifically at SOE, the EBL was initially practiced on a trial
basis because CCP leadership believed that SOE first needed more autonomy in operations and
management, which was eventally granted through the State-Owned Industrial Enterprises Law
in 1988. However, even after 1988, full implementation of the EBL was delayed, and was
133
Table 4.2. Number of Bankruptcy Cases Disposed of by Chinese Courts, 2003 12
Year Cases Disposed
2003 6795
2004 5777
2005 3436
2006 4755
2007 4200
2008 3749
2009 3573
2010 3567
2011 2531
2012 2100
Source: Steele et al. (2018, 8–9)
subsequently suspended after the defeat of the democracy moment in 1989 (Tanner 1999, 165–
66).
The EBL was eventually revived and fully activated toward all Chinese firms in 2007,
with a one-year grace period for SOE. However, far from enticing distressed firms to pursue
restructuring, the EBL’s enactment was followed by a decline in insolvency filings from an
already low number (Table 4.2.). Several reasons explain why bankrutpcy has been rare under
the law.
26
First, local governments and courts have had an incentive to prevent large numbers of
bankruptcies to maintain social stability. Second, courts often avoid accepting insolvency cases
26
Other works, as well as interviews with bankruptcy lawyers and economists in China, corroborate many of these
points. On the first point, Wang (2015, 264) notes that in virtually all SOE bankruptcy cases, courts appoint so-called
“liquidation groups” (Qingsuanzu) that include local government officials from the SOE’s home jurisdiction. Since the
officials are charged with maintaining social stability and economic performance in their home districts, they are
incentivized to intervene the case on the SOE’s behalf, which generally prolongs the time it takes to settle the case.
The slow processing rate has a negative effect on judges’ willingness to hear bankruptcy cases, since judges are partly
evaluated by the number of cases they try. Numerous interviews also mention the importance of large firms,
including SOE, for local governments as sources of tax revenue, employment, and economic development
(Interviews 47, 50, 51, and 54). On the second point, Jiang (2014, 569) elaborates how courts’ discretion in deciding
to accept reorganization cases is a disincentive for firms to file for bankruptcy, since there is no clear standard for
financial health that firms must satisfy in order to file as in the United States, which allows courts to be subjective in
rejecting cases that could in fact be initiated. On the third point, one of China’s earliest judges specializing in
bankruptcy opined that there were no more than 100 lawyers and judges specializing in bankruptcy in all of China,
and more likely between 10 and 50 (Interview 50).
134
with potentially controversial problems, such as when a debtor does not have enough money to
pay all claimaints. Third, insolvency proceedings frequently take significant time and there is a
shortage of experienced judges and specialist courts, which make it simpler for creditors to
enforce their security directly or start execution proceedings. Fourth, firms themslves often lack a
clear idea about the difference between reorganization and liquidation, and are concerned about
liabilities if they pursue the former (Steele et al. 2018, 9–10).
Disincentives to restructuring and the shortage of lawyers specializing in bankrupcty law
are elements that China under the EBL has generally shared with Japan in the 1990s before the
Civil Rehabilitation Law. However, it bears emphasis that practice of the EBL has also seen
advancements in China in recent years, and that the effectiveness of the law differs to a
considerable degree by region. For example, according to a bankruptcy lawyer with ten years of
prior work experience as a bankruptcy judge, in the city of Shenzhen near Hong Kong, the
trends of marketization and fast growth have led to professionalization among legal personnel
involved in bankruptcies, and there is decreasing local government intervention in bankruptcy
proceedings with the rare exception of cases that are sensitive and have substantial social impact.
In most cases of private firms and even SOE in Shenzhen, there has been complete judicial
independence (Interview 50). A lawyer based in Hangzhou mentioned a similarly well-developed
bankruptcy court system in Zhejiang province, which is an area with a large and robust private
sector (Interview 48). However, both lawyers noted that these locales were exceptional in their
well-developed bankruptcy administration, and that less developed regions in China’s interior
have less mature private sectors and less experience with insolvencies, and thus less-developed
institutions.
27
Though the lawyers’ observations about regional variation in China’s bankruptcy
27
For these rural provinces with less developed private sectors, the comparatively large importance of the remaining
state sector continues to make it hard for local governments to allow SOE to enter bankruptcy (Interview 54).
135
regime is anecdotal evidence, it is aligned with Figure 4.3.’s picture of the geographic distribution
of zombie firms wherein more developed provinces have lower zombie firm ratios, and less
developed provinces have higher zombie ratios.
In sum, circumstances in China under the EBL, though not exactly the same as in Japan
prior to the Civil Rehabilitation Law, are generally consistent with the hypothesis that
bankruptcy law influences the level of zombie firms. Because of government officials’ strong
interest in protecting local SOE which are typically large-sized and have many employees,
officials try to prevent SOE from entering bankruptcy proceedings in the first place, and are
motivated to intervene in such cases when they occur, which slows down the proceedings and
makes judges less inclined to accept future bankruptcy cases with other SOE. While the situation
is gradually changing for private firms in China’s wealthier regions, the relatively small number
of legal professionals means that it may still be more convenient for creditors to pursue other
avenues of redress than bankruptcy. Moreover, there is still confusion among private firms about
what reorganization entails, and diminished incentive to file for bankruptcy because of fear of
arbitrary rejection by courts with excessive discretion over accepting cases. In short, considerable
regional variation and noteworthy advancement notwithstanding, China’s current bankruptcy
framework has not been conducive to restructuring, which may explain why many distressed
firms, especially SOE, have instead become zombies.
Orienting Hypothesis 3. Drift
The dissertation’s third and final orienting hypothesis concerns the drift of policy
arrangements that provide extraordinary assurance to banks about loan recovery, thereby
reducing or even eliminating the banks’ attention to default risk in their lending behavior, and
136
encouraging them to lend to firms which might ordinarily be unappealing customers on business
grounds. In the case of Japan during the lost decades, such an arrangement manifested in the
form of the credit guarantee system, which provided high-level guarantees to private banks to
facilitate lending toward SME. While these guarantees were largely intended as a crisis response
measure during Japan’s credit crunch in the late 1990s, they persisted long afterward because of
a status quo-favoring combination of limited public involvement in the policymaking behind the
system, strong support from system insiders, and bureaucrats who held conflicting incentives
regarding the system’s future. The perpetration of the system’s high guarantee ratios, as well as
the system’s enormous size, contributed to a notable zombie SME presence throughout the lost
decades.
Like Japan, China offers credit guarantees to SME. Like Japan, some of China’s
guarantees have also been 100 percent guarantees (Mu 2002, 22). However, the structure and
financial scale of China’s credit guarantee infrastructure are vastly different, and do not represent
an equivalent to Japan’s credit guarantee system that has been capable of widespread zombie
creation. On structure, China possesses a much larger number of credit guarantee institutions—
3,366 were registered at the end of 2006 (Reuters Staff 2009), compared to 52 in Japan. These
institutions fall into three types: government-sponsored credit guarantee corporations, member-
funded mutual guarantee funds, and private sector-invested commercial guarantee companies.
But these institutions do not exist in every part of the country (G. Zhang 2009, 72), and their
financial wherewithal to support SME. For example, in 2006, China’s credit guarantee
institutions had on average registered capital of less than 30 million yuan ($4.39 million) each. By
contrast, the national budget for Japan’s credit guarantee system was nearly $900 million in 2006
for a country one-tenth China’s size, and that budget was further augmented by prefectural and
municipal governments (Table 3.3.). Moreover, the guarantee ratio in China has generally
137
ranged between 90 percent and 50 percent (P. Zhang and Ye 2010, 101), meaning that it is
unlikely that China’s guarantee institutions have generated a similar degree of moral hazard in
banks’ lending toward SME. Therefore, it is unlikely that credit guarantees have contributed to
widespread zombie formation in China.
Since there is little reason to believe that credit guarantees have been a significant source
of zombie creation in China, it is difficult to conclude that drift in China’s credit guarantee
infrastructure has perpetuated a large-scale zombie problem similar to the problem among
Japanese SME.
Soft Budgetary Constraint as an Alternative Explanation
Because the explanatory power of the orienting hypotheses is largely restricted to the
hypothesis for bankruptcy law and not apparent in the other hypotheses for transparency and
policy drift, it follows that limited comparison can be made between Japan and China’s
encounters with zombie firms. While this finding might reassure some that China is likely to
avoid a lost decade of its own because it has lacked certain mechanisms associated with zombie
firm creation in Japan, the resiliency of zombies among certain firm types in China, especially
SOE, begs the question of what mechanisms are then at work in the Chinese case.
Soft Budgetary Constraint and Zombie Firms
One possible causal mechanism behind zombie firms in China is soft budgetary
constraint, a concept proposed by Kornai (1979) in relation to socialist systems in transition.
Lacking precise and exclusive indicators—credit, subsidies, taxes, administrative pricing or any
combination thereof are all included (Kornai 1992, 140–42)—soft budgetary constraint generally
refers to governments’ provision of financial support that prevents firms in socialist systems from
138
failure, no matter how poor their corporate performance. Consequently, soft budgetary
constraint is often seen as a source of inefficiency in socialist systems, since it removes firms’
incentive to operate in ways that minimize deficits (Kornai, Maskin, and Roland 2003, 1096).
The theoretical connection between soft budgetary constraint and zombie firms is thus
obvious. In conditions where the state acts as an unconditional creditor, firms can behave
without financial discipline because there is no chance of closure if expenses exceed income.
While this does not mean that all firms in socialist systems necessarily become zombies, soft
budgetary constraint should enable the state’s preferred firms to acquire zombie status without
fear for survival, because the firms are confident of receiving life support from the state. As a
result, this aspect of socialist systems should be associated with higher incidence of zombie firms,
since firms granted soft budgets have no clear motivation to avoid situations where they must ask
for forbearance from banks.
While socialist systems, soft budgetary constraint, and zombie firms are interlinked at the
theoretical level, it is unlikely that socialist systems ever exist in a pure form, at least for very long.
As De Grauwe (2017, 6–7) indicates, two reasons account for why pure socialist systems tend to
break down. First, socialist systems cannot keep up with the high informational demands of
meeting consumer preference for large populations, which leads to problems of large surpluses
and severe shortages. Second, they fail to activate effort and creativity on the part of their
populations, which causes technological stagnation.
Given that socialist systems must embrace some elements of capitalism in order to
function, it follows that soft budgets must also be partial. They cannot apply evenly to all
businesses and create limitless numbers of zombies, or the socialist systems themselves would
eventually fail. The question then arises: Which businesses in socialist systems are entitled to soft budgetary
constraint and thus have a greater likelihood of taking zombie status? In principle, since socialism as an idea
139
is centered on the common good, the citizens of socialist systems should collectively decide which
firms receive state support. In reality, socialism often takes the political form of dictatorship in
which a single political party nominally works on behalf of the population to achieve equality
within society. Thus, it is typically a dominant party and especially party leadership that decides
which firms receive soft budgets.
While political leaders in authoritarian socialist systems must exclude some firms from
soft budgetary constraints to maintain system survival over the long haul, like dictators elsewhere,
they must also be alert toward the possibility of being removed by regime insiders who may want
power for themselves. In theory, leaders often deal with this latter situation by offering some
combination of public and private goods to loyal supporters to keep them from defecting to the
opposition (Bueno de Mesquita et al. 2003; Gandhi and Przeworski 2006). They may also adopt
various aspects of “institutionalization,” an umbrella concept that generally refers to formal and
informal constraint “such as norms around leadership succession and promotion, functional
specialization of the bureaucracy, and (limited) popular political participation which lend
resiliency to the regime (Nathan 2003; Svolik 2012).
Thus, one hypothesis about firms that are more likely to face soft budgetary constraint
and thereby take zombie status are firms that confer public and private benefits to regime insiders
and whose existence is bolstered by institutionalize norms that support party rule.
Soft Budgetary Constraint and Zombie Firms in China
As the hypothesis about soft budgetary constraint predicts, China’s political leaders have
been selective about the firms that receive soft budgets. While the CCP has increasingly
supported private firms for their contribution to national growth in the “reform and opening”
period since the late 1970s, evidence suggests that party leaders have persisted in protecting
140
SOE, particularly since the mid-2000s, and that this firm category is granted soft budgets partly
because of its importance in compensating party elites. Furthermore, the imposition of
meritocratic standards for cadre evaluation help sustain SOE and their ability to take zombie
status, since local leaders evaluated on the achievement of economic development and social
stability targets are incentivized to support these firms’ survival and expansion rather than rein in
these firms and possibly reduce economic growth and employment.
From the start of the “reform and opening” period in the late 1970s, the CCP has
pursued an ambitious market reform agenda that has expanded the private sector and given rise
to ever larger numbers of privately-owned firms (Huang 2008). In 1978, the number of individual
businesses in urban areas was 150,000 and represented 0.2 percent of national employment.
Roughly two decades later, in 2000, China’s urban areas had 12.7 million private firms and 21.4
million individual businesses which together constituted 14.7 percent of national employment. In
2012, there were 75.6 million private firms and 56.4 million individual businesses which together
constituted 35.6 percent of national employment (Lardy 2014, 261).
At the same time as the CCP showed increasing tolerance and support for private firms in
China’s socialist system, it also tried to improve the productivity and profitability of the state
sector. On the heels of Deng Xiaoping’s “southern tour” in 1992 to the market-oriented province
of Guangdong, the party announced a major reform plan to establish market-supporting
institutions like fiscal federalism, separate SOE from the government, and privatize smaller SOE.
Insofar as these reforms concerned SOE, they were defined by the slogan “grasping the large and
releasing the small” (Zhua da fang xiao), and resulted in substantial reductions in the number of
SOE from 127,600 in 1996, to 61,300 in 1999 (Lardy 2014, 158–59), and then to 29,449 in 2002
(Yusuf, Nabeshima, and Perkins 2006, 78–80). Moreover, fiscal federalism introduced by a new
tax sharing system in 1994 presaged to incentivize provincial governments to advance economic
141
growth because of the system’s revenue-sharing rule with the central government, and cut off
subsidies from the center to provincial governments, which essentially meant that latter would be
presented with a hard budget for supporting their local SOE (Steinfeld 1998, 237).
While the reforms in the 1990s marked a turning point in China’s shift toward a market
economy and correlate with the decline in the number of zombie unlisted firms in the late 1990s
and early 2000s (Figure 4.2.), their impact on eliminating access to soft budgets by SOE was
limited in subsequent years. In 2003, new CCP leader Hu Jintao reversed the trend of separating
SOE from the government by establishing a new government body called the State-Owned
Assets Supervision and Administration Commission (SASAC) under the State Council, and
designated it to be the main shareholder of 196 central SOE. Local SASAC were also established
at the same time to keep controlling shares in provincial and municipal SOE across the country
(McGregor 2012, 17).
Though the SASAC’s nominal role of investor suggests that the SASAC held rights over
SOE, in truth both the central and local commissions had notably weak control powers, while
the CCP kept influence over many key aspects of SOE operations. For example, all SOE have a
dual personnel system composed of a normal corporate manager track and a party official track,
and there are frequent overlaps between the two tracks. Because the CCP appoints the party
officials—typically a secretary of the party committee, several deputy secretaries, a secretary from
the discipline inspection commission (an anticorruption office), and other members of the party
committee—the party has strong authority over SOE business decisions (Lin and Milhaupt 2013,
737).
The SOE Asset Law promulgated in 2008 and activated in 2009 also muddled the extent
of the SASAC’s authority. Though the law named the commission as the controlling shareholder
of all SOE and gave it the rights of an investor, it empowered the State Council and local
142
governments to “when necessary, authorize other departments or bodies to perform the
investor’s functions for state-invested enterprises on behalf of the corresponding people’s
government,” while clearly highlighting the obligation of SOE to “safeguard the basic economic
system of China . . . giving full play to the leading role of the state-owned economy in the
national economy” (McGregor 2012, 67). Furthermore, there is no top-down relationship
between the central SASAC and local SASAC. Instead, local SASAC are organizationally
situated under the provincial or municipal governments where they serve, so they cannot use the
authority of the central SASAC as leverage in dealing with local SOE, which undermines their
influence vis-à-vis local governments (Interview 54).
Because of the SASAC’s confused and weak governance role, the CCP is able to work
through the SASAC (or in spite of it) to supervise both personnel appointments and executive
compensation at SOE. As expected by the hypothesis on soft budgetary constraint, the rewards
confered through these decisions are seen as helpful to party insiders. In 2009, the average
compensation for CEOs at the central SOE supervised by the SASAC was 600,000 yuan
($88,000). Though nominally this may not seem like a large sum, CEOs at SOE are also entitled
to many bonus privileges such as large corporate expense accounts and free or heavily discounted
housing, education and medical care which make their positions highly remunerative (Lin and
Milhaupt 2013, 742–43).
In addition to these formal forms of compensation, SOE executives may also benefit from
their positions through various forms of illicit activity. For one, executives can generate
corruption income through awarding well-paying contracts to private businessmen who are
willing to pay bribes. They can also abuse corporate restructuring efforts by grossly undervaluing
the enterprise’s underlying assets, and then sell the assets to relatives or friends who make colossal
windfall gains, or sell them to third parties in exchange for bribes. Because of poor financial
143
controls at SOE, executives can also easily engage in embezzlement and misappropriation (Pei
2016, 43, 121, 160). These types of illegal behaviors can be enormously lucrative. One study
reports that the average amount of corruption income for SOE executives caught by authorities
was 30 million yuan, nearly five times the average figure for corrupt local officials (Pei 2016, 155).
While profitmaking potential for party elites is one reason that the CCP continues to
support SOE and tolerate their access to soft budgets, the party’s increasing emphasis on
meritocratic considerations in personnel promotion may also shape local officials’ decisions about
helping SOE even when the firms face financial trouble. Ang (2016, 110–25) shows that
throughout the 1980s and 1990s, performance evaluation of local officials in China was primarily
based on the meeting of quantifiable economic growth targets (e.g. increase in industrial output),
meaning that officials had a clear incentive to support increasing investment by SOE. Even
though the list of evaulation criteria grew during the 2000s to include various mandates like
environmental protection and energy conservation, local leaders continued to treat economic
and fiscal growth as the utmost priority because it was measurable and visible, and because it
provided personal benefits to the leaders such as the abilities to distribute patronage and collect
personal rents. These rationales correlate with evidence that SOE were prone to overinvestment
in the 2000s (Ying, Luo, and Wu 2013), and that financially independent local governments
worried about their own debt have been willing to allow local SOE to default on their loans,
which then become debt for the central government in the form of nonperforming loans held at
state-owned banks (Yeung 2009, 298).
Though local leaders’ focus on satisfying economic goals may partly explain China’s
economic prosperity, the heavy emphasis on growth targets also impacts leaders’ willingness to
allow distressed SOE to fail. As a Chinese economist at a government think tank pointed out, as
China’s remaining SOE have increased in size over time, their employment has increased as
144
well, which makes governments even more reluctantant to let them undergo bankruptcy.
“Because bankruptcy would result in huge layoffs, it would necessarily be a huge threat to social
stability (shehui wending). That is why whenever such a firm runs into operational difficulty, the
government thinks of various ways to save it. This pattern is the most fundamental reason why
certain SOE become zombie firms” (Interview 54). Maintaining social stabilty has also been one
of the few targets for local officials where failure can negate economic growth achiements and
potentially even lead to dismissal (Ang 2016, 111).
These two support dynamics for soft budgets toward SOE were apparent when the global
financial crisis hit in 2008 and threatened China’s economic growth. The large-scale stimulus
issued by the Hu administration as a countermeasure significantly facilitated bank lending to
SOE for years after the crisis, even though SOE had lower profitability than privately-owned
firms and thus ought to have had less access to credit based on market principles (Figures 4.5.
and 4.6.). By the time new CCP leader Xi Jinping pledged that market forces would play a
“decisive” role in China’s economy in 2013 (Subler and Yao 2013), numerous industries
dominated by SOE contained very high percentages of zombie firms, including mining (17.54
percent), transportation (17.58 percent), utilities (20.49 percent), and, above all, steel (51.43
percent) (Table 4.1.).
Conclusion
This chapter examined zombie firms in China in the period 1998-2015. While the
zombie firm ratio and total number of zombies in China declined through much of this period,
the zombie ratio grew within certain industrial sectors after 2008, due in part to a large-scale
government stimulus after the global financial crisis. Evidence suggests that the stimulus
145
particularly supported SOE by enhancing their ability to obtain bank loans, even though SOE
performed worse in terms of profit margins than private Chinese firms which conversely had
more difficulty obtaining financing.
Given the considerable presence of zombie firms in China, the chapter explored the
extent to which several possible causal mechanisms associated with zombie production in Japan
in the lost decades were present in China. It found that of the three hypothesized causal
mechanisms—transparency, bankruptcy law, and policy drift—evidence was most consistent
with the bankruptcy law mechanism, and not well aligned with the other two mechanisms.
Because of the general inability of the original hypotheses to explain China’s zombie
firms, the chapter laid out a fourth potential hypothesis related to the particular status of China
as a socialist system called the soft budgetary constraint hypothesis. Evidence for this alternative
hypothesis was consistent with the theoretical expectation that the CCP would selectively engage
in offering soft budgets to firms through which benefits could be conferred to party elites,
particularly if these firms’ existence was also supported by institutionalized norms that sustained
party rule. The chapter argued that SOE in particular were suitable to soft budgets due to their
ability to generously compensate their appointed party employees, and to the fact that local
government officials were incentivized to assist SOE because of their performance evaluation
criteria that prioritized economic outcomes as well as social stability. It is thus logical that
estimates of zombie firms in China would show that listed firms among which there was a high
percentage of SOE would have a relatively high zombie ratio from 2003, the turning point when
Hu administration reversed the trend of separating the state from SOE (Figure 4.1.).
Though the chapter’s statistics on China’s zombie ratio end in 2015 when there was still a
generally high level of zombies among listed firms, it bears noting that, since early 2016, the Xi
administration has made some attempt to remove indebted firms from China’s economy,
146
including in the state sector where resistance has traditionally been the strongest. For example, in
February 2016, high-level authorities pressed the state-owned Wuzhou Shipyard in Zhejiang
Province, a company that once had more than 3,000 employees, to file for bankruptcy after its
debts reached 911 million yuan against total assets of 534 million yuan in September 2015 (Juan
2016). In March 2016, securities regulators delisted the firm Zhuhai Boyuan Investment from the
Shanghai Stock Exchange after it issued a suspicious earnings report which likely hid debt, and
reportedly kept another 80 companies as candidates for future delisting (Otani 2016). In June
2016, a State Council task force expanded a pilot project to depoliticize central SOE
management by allowing corporate boards to select top managers. The project further proposed
to raise compensation for these managers in line with industry norms to better attract outside
managerial talent, while keeping political appointees’ pay at the standard government level
(Wildau 2016a).
On the surface, these recent top-down efforts to accelerate the reorganization of insolvent
SOE and raise corporate efficiency give some for optimism by implying awareness about the
problem of zombie firms. Alertness by China’s leaders stands in notable contrast to the Japanese
politicians discussed in the Chapters 2 and 3, whose inactivity and favoritism toward certain
constituents respectively led to high periods of zombies among Japanese listed firms and
particularly SME.
At the same time, because of the CCP’s widespread use of SOE as a political support
device for the party, deep changes to the state sector may prove difficult. Piecemeal steps to rein
in zombie firms notwithstanding, until recently the Xi administration was widely viewed as failing
in SOE reform, with lack of comprehensiveness and clarity in top-level policymaking being
common critiques (Donnan 2016; Buckley 2017). Not only have officials from foreign
governments raised concern about the effect of SOE overcapacity on low-priced imports like
147
steel and aluminum (Buckley and Perlez 2016; Donnan and Mitchell 2016b; Sanderson and Foy
2017); voices inside China have also criticized the pace of reform, including state-supported
research organizations and think tanks like the China Enterprise Reform and Development
Society and the Economic System and Management Institute of China’s National Development
and Reform Commission which are not usually outspoken about the direction of leaders’
decision-making (Caixin 2016; Buckley 2017). Asked about the Xi administration’s reform
efforts, the prominent Chinese economist Wu Jinglian said that implementation “hasn’t been
vigorous enough.” Fudan University economist Li Weisen similarly remarked, “We’ve maybe
issued over a hundred of these reform documents…But which ones have really been
implemented, including fiscal reforms?” (Buckley and Bradsher 2017).
China may have escaped Japan-style zombies for now, but the future solution to its own
zombie firm problem will largely depend on what the Xi administration does in his second term
as CCP secretary. While Xi went to unprecedented lengths to consolidate power under his
leadership in his first five-year term as party leader, it remains to be seen whether he use this
power going forward to address the soft budgetary constraint enjoyed by many debt-ridden and
inefficient SOE and bring their productivity closer to that of China’s increasingly vibrant private
firms.
148
CONCLUSION
Main Findings and Contributions
This dissertation began with the broad challenge of trying to understand the political
economy of zombie firms, unprofitable businesses that survive in the marketplace through
discounted interest rates and forbearance lending, and are generally thought to hinder recovery
by blocking creative destruction after economic crises. In attempting this challenge, the
dissertation asked two basic and interrelated questions: What political conditions make these problematic
firms possible, and what political conditions lead to their elimination?
Because of the relative newness of zombie firms as a subject of academic inquiry, the
dissertation aimed to develop first-cut theoretical explanations for their occurrence. It examined
several orienting hypotheses about these firms’ political underpinnings in two national case
studies on Japan after the banking crisis in the 1990s and China after the global financial crisis,
using a wide range of information including large-n statistics showing variation in the zombie
firm ratio over time, secondary source material, government and international organization
reports, and interviews with financial and legal industry professionals, government officials, and
academics in both cases. The main findings of these case studies can be summarized as follows.
1. Zombie firms increase when the prevailing regulatory structure allows banks
to disguise debt by extending financial assistance to troubled firms. In Japan
during the 1990s, weak legislative oversight of the country’s primary bank regulator, the
Ministry of Finance (MOF), enabled the ministry to pursue its preferred forbearance
149
regulatory approach toward indebted banks for several years following the outbreak of
the crisis. Lack of transparency about the banking sector’s health hindered politicians
from making policy interventions, and let banks mask their low capitalization by
expanding credit support to impaired firms, who were willing to deepen their own
indebtedness in the hopes of future recovery. The asymmetrical power between MOF
and Japanese politicians to deal with the crisis and stop the occurrence of zombie firms
was at least partly an outcome of the historical legacy of single party dominance by the
LDP, which discouraged politicians from establishing oversight mechanisms toward MOF
prior to the crisis because of the politicians’ sense of long-term electoral stability.
2. Not all zombie firms are birthed by banks wishing to conceal bad assets by
lending to hopelessly distressed businesses. Evidence from the case of listed firms in
Japan shows that among the zombie firms receiving evergreen lending, many held
zombie status for only one or two years. This implies that not all bank lending decisions
were influenced by capitalization levels, and that many decisions were based on the
perceived viability of borrowers who may have suffered from short-term demand shocks
and just needed temporary assistance to return to profitability. In other words, politics
explains the creation of some zombie firms better than others.
3. Strong banking regulation, particularly regulation that enhances transparency
about banks’ capitalization, reduces the prevalence of zombie firms after crises.
Starting in the late 1990s, the rise in Japanese politicians’ oversight power and demotion
of the MOF’s influence precipitated a clean-up of the high debt levels at Japanese banks.
Following these changes, a low zombie firm ratio was observed among Japan’s listed firms
after 2001. This finding also means that zombie firms—at least among listed firms—do
150
not explain the sluggish growth in Japan’s economy during the 2000s, a period commonly
known as Japan’s second lost decade.
4. Certain bankruptcy rules can fast-forward the decision by distressed firms to
pursue corporate restructuring, thereby helping the firms avoid zombie status
or exit from it more quickly. The initiation of bankruptcy legal reform in Japan in
2000, particularly the Civil Rehabilitation Law, coincided with a substantial decline in
zombie firms among listed firms from the following year. Japan’s previous bankruptcy
framework likely contributed to the incidence of zombie firms during the 1990s, since it
provided little incentive for managers at distressed firms to use legal avenues to shed
unproductive assets to recapture profitability.
5. Zombie firms increase when public finance policies significantly reduce
banks’ risk of loan default. As illustrated in the case study of SME in Japan, zombie
firms may emerge even when banks do not hold large amounts of bad assets. Instead,
through policies such as the credit guarantee system for small businesses—an
arrangement by which the government offers to guarantee loan repayment to private
banks lending to SME—banks may be incentivized to lend to firms that are ordinarily
unattractive customers because the banks feel assured of both loan recovery and interest
payments. These firms, which may be only marginally profitable and carry higher a priori
risk of repayment trouble, are more likely to request future financial assistance from their
lenders, thus increasing the firms’ likelihood of taking zombie status.
6. Public finance policies that increase the likelihood of zombie firms can have
their origins in government responses to economic crisis, but outlast the acute
151
crisis period and persist through times of relative economic growth when the
policies undergo drift. In the case of Japan’s credit guarantee system, strong support
from concentrated interests like SME and regional banks, and bureaucratic uncertainty
within the SMEA about the possible effects of reform, combined to block change in the
system’s high coverage ratios during the 2000s, even after the end of the domestic credit
crunch that initially justified the high ratios in late 1990s. The drift of the high coverage
ratios enabled the system to continue generating zombie firms during Japan’s second lost
decade, over opposition toward the coverage levels from government officials and
academics, and the Japanese public’s general preference for neoliberalism.
7. Zombie firms are likely to emerge in conditions where firms are subject to
soft budgets, a financial arrangement typically associated with SOE in socialist
systems. In the case of China, evidence indicates that more SOE than private firms held
zombie status both before and after the global financial crisis. However, the massive
government stimulus that followed the crisis appears to have disproportionately benefited
SOE, which could obtain loans with relative ease in the post-crisis period notwithstanding
declining profitability. By contrast, private enterprises faced comparative difficulty in
borrowing despite exhibiting stable profitability.
Taken together, these findings offer several contributions to our emerging knowledge of
the political economy of zombie firms. For one, the political mechanisms behind the generation
of zombie firms are manifold. Zombie firms may appear in post-crisis situations where loose
banking regulation allows weakly capitalized banks to act on their incentive to minimize
recognition of bad assets, a scenario whose antecedent cause may lie in the structural relationship
152
between the legislative and executive branches which enables regulators to avoid transparency
toward elected officials. Or zombies can be indirectly spawned by the characteristics of a
country’s bankruptcy regime, which may complicate attempts by creditors and debtors to pursue
restructuring and instead encourage impaired firms to seek additional credit support.
Additionally, they can persist on a wide scale even after acute crisis moments have ended, when
temporary assistance policies experience drift and incentivize banks to continue financing
marginally performing firms that might otherwise lack access to credit. Finally, they can
materialize in environments with soft budget constraints, a feature normally associated with
socialist systems where government-owned banks exercise financial leniency toward SOE.
Equifinality in the political creation of zombie firms is not theoretically impossible.
However, it calls into question the conditions that are universally necessary to trigger the creation
of these firms, and hints that not all zombies may be equal qua zombies, even though they satisfy
the common conceptual criteria of being unprofitable and in receipt of discounted interest rates
and evergreen lending. In short, economic crisis may not be a necessary condition for zombie
firms to emerge, since zombies may also happen in the presence of soft budgets or certain
support policies like Japan’s credit guarantee system, whose institutional origins date to the early
postwar era and likely contributed to some amount of zombie SME before being enhanced
during the banking crisis to serve a larger number of weak SME. Moreover, in the presence of
these factors, firms’ zombification may be an acceptable or even intentional outcome for its
contribution to non-economic objectives, even though it does not improve the firms’ objective
business conditions. Zombie firms in these settings may have a normative dimension that is
distinct from the zombies birthed by undercapitalized banks in the wake of economic crises, and
specific political rationales that override the economic criticism that the firms are inefficient and
exert problematic externalities on healthy firms by causing market congestion.
153
The complex significance of economic crisis to the explanation of zombie firms—a
necessary condition in some explanations, while more distantly present or even absent in
others—means that research on the political economy of zombie firms requires conceptual
boundaries for when zombies are a product of troubling policy configurations thrown into relief
by crisis, and for when they happen in ordinary times when their creation might be more closely
related to deliberate political intention. This dissertation attempted to focus on the former
scenario, but, in so doing, it revealed certain mechanisms that temporally transcend crisis
contexts. Thus, one of its chief contributions is to highlight the conceptual challenge of studying
zombie firms as primarily a phenomenon of post-crisis policy inertia.
Aside from examining multiple causal mechanisms for zombie firms, the dissertation also
supplies empirical evidence about the existence of these firms in its case studies of Japan and
China that may be useful to scholars of these countries’ political economy. For example, in
exploring zombies among both listed firms and SME in Japan’s lost decades, it shows not just
that zombie firms rose to high levels in both corporate categories in the 1990s, but that it was
really the small firm category where the zombie problem persisted in the 2000s. In this way, the
dissertation provides nuance about zombie firms in Japan which are often generalized as
occurring uniformly across the lost decades (Prestowitz 2015), in addition to clarifying that it was
a particular policy arrangement, the credit guarantee system, that played a role in the substantial
zombie presence among SME.
Furthermore, the dissertation shows that despite the recent concern about zombie firms
in China, the ratio of these firms generally declined in the early 21
st
century. While it is true that
zombies increased in certain specific sectors, especially those populated by SOE, after the global
financial crisis, the overall picture of these firms in the past two decades suggests that China
made a considerable transition toward embracing the process of creative destruction in its
154
economy. Additionally, the absence in China of two political mechanisms associated with the
creation of zombie firms in Japan demonstrates the limits of analogizing China’s recent economic
history to Japan’s lost decades.
Avenues for Future Research
This dissertation was conceived as an initial exploration of the political mechanisms
behind the emergence of zombie firms after economic crises. Due to the nascent stage of political
economy theory on zombies and data constraints, it used a case study research design to
intensively investigate a small number of orienting hypotheses in two national post-crisis cases.
With few reservations, it found empirical evidence in support of these hypotheses.
Given the dissertation’s identification of potential independent variables in the creation of
zombie firms, a logical next step in the research agenda would be to expand the amount of cases
under study, to assess and possibly extend the external validity of the hypotheses established in
the dissertation. Until very recently, this task would have been difficult because of a lack of cross-
national data with variation in the zombie firm ratio. However, newly appeared working papers
about zombie firms in OECD countries and Eurozone countries after the global financial crisis
indicate that such data might now be accessible (McGowan, Andrews, and Millot 2017; Acharya
et al. 2017). While these papers’ use of the CHK definition of zombie firms complicates direct
comparison of their zombie ratio estimates with those in the dissertation, the papers’ cross-
national econometric analysis generally suggests how a wider study with many cases might be
performed to evaluate the hypotheses presented in the dissertation.
Beyond theory testing, another future step could be to further refine the hypotheses
outlined in the dissertation. For example, the second orienting hypothesis on bankruptcy regimes
155
could be further specified to say what precise factors in insolvency laws affect managers’ decisions
to pursue restructuring. On this point, it is commonly known that Japan has strong legal
protections for lifetime workers that make it hard for managers to lay off workers with this type of
employment (Jiyeoun Song 2014). Perhaps specific conditions like employee protections (or other
conditions) could be the deeper reason why managers at distressed firms hesitate to pursue
restructuring that could lead to labor downsizing (or other managerial challenges), and more
inclined to request financial assistance that turns their firms into zombies.
In addition to these potential paths for improving political economy theory about zombie
firms, the findings in the dissertation’s case studies suggest several other projects about the
zombie issue in each case.
1. One of the dissertation’s main substantive findings is that zombie firms were not a large
presence among Japanese listed firms from the early 2000s onward. This finding upends
conventional thinking about zombie firms’ role across Japan’s lost decades, and begs a related
question: In the absence of obstruction by zombie firms, how and why did financially healthy listed firms
nevertheless fail to generate growth, and thereby contribute to Japan’s stagnation?
Nakamura (2017) proposes that the reforms in Japan’s accounting rules and bank
supervision policies at the start of the 2000s made managers at healthy firms worried about
possible financial distress and ratings downgrades, which in turn led them to emphasize debt
repayment over investment for business development. Consequently, many firms were reluctant
to take on leverage despite being cash-rich and did not absorb productive resources released from
troubled firms, preferring instead to guard return-on-equity ratios, pursue constant restructuring,
and minimize capital expenditures—activities that reduced risk and protected the status quo, but
did not increase firm productivity.
156
Nakamura’s argument, while incisive, lacks a political basis. If board members,
institutional investors, individual shareholders, and politicians had an interest in growth, how did
managers’ conservatism prevail in conflicts over corporate control, despite that their firms were
generally healthy and capable of investment and other growth-producing activities? Future work
on this puzzle might start with investigating the structural power of managers in governance
reforms, which Culpepper (2011) finds to be influential in impeding corporate takeovers in Japan
and several European countries during the 2000s. It is possible that Japan’s economic
performance in the second lost decade might be partly explained by the low political salience of
corporate governance issues, which allowed Japanese managers to pursue risk-averse corporate
strategies at the expense of other approaches that could have generated greater productivity and
profitability.
2. The dissertation’s study of zombie listed firms in Japan ends in 2008. However, that year
marked the outbreak of the global financial crisis, an event that brought economic turmoil from
which the country was slow to recover. Because the dissertation contends that banking regulatory
reforms starting in the late 1990s effected the decline in zombie listed firms in the early 2000s, it
would be useful to extend the study of listed firms to the post-2008 period to see whether the
zombie firm ratio remained at a low level or reemerged on a comparable scale to Japan’s
economy in the 1990s. The latter scenario would raise doubt about dissertation’s hypothesis that
strong and transparent banking regulation should minimize the presence of zombie firms. This
research would possibly expand on the notion that Japan responded differently to the two crises,
a notion that has been generally explored elsewhere in regard to contrasting crisis reactions
(Pempel and Tsunekawa 2015).
157
3. A common bugaboo in the debate about Japan’s sluggish economy is the country’s alleged lack
of entrepreneurship. The widely-held view is that Japan faces a scarcity of startup firms in
comparison to places like Silicon Valley in the United States because of cultural reasons. Japan is
said to be a “shame culture” or an “uncertainty-avoiding” country where people find it hard or
undesirable to take risks common to new business ventures (Pesek 2014, 118, 208–9; Makinen
2015). To an extent, statistics appear aligned with this account. The 2014 Global
Entrepreneurship Monitor report ranked Japan second-to-last in early stage entrepreneurial
activity, far behind geographic neighbors like China and Taiwan and only ahead of Suriname in
terms of perceived entrepreneurial opportunities and abilities (Global Entrepreneurship Monitor
2014). The 2015 Amway Global Entrepreneur Report ranked Japan dead last in entrepreneurial
spirit out of 44 surveyed countries, with only 13 percent of Japanese respondents saying that they
could imagine starting a business (Amway 2015, 8, 11).
The dissertation’s findings about zombie SME in Japan suggests a different possible
explanation for Japan’s low level of startups—zombie firms’ impediment on market entry by new
firms—which might be further evaluated in the debate on Japanese entrepreneurship. Evidence
from fieldwork interviews tentatively casts doubt on this explanation, however. For example, a
Japanese professor involved in the administration of the Global Entrepreneurship Monitor survey
used the example of a popular new barbershop chain to assert that older and less productive
shops did not pose significant barriers to market entry:
The problem is that people in the same industry don’t have the power to challenge to
make new things. Take the example of QB House, a barbershop that offers ten-minute
cuts for 1,000 yen. Now it has 500 or 600 stores in Japan. In some sense, it is a destroyer.
It’s said that Richard Katz likes this example. It’s innovative. It’s a disruptor. To some
extent, the established industry opposed QB House when it formed. But it wasn’t such
strong opposition. Now it has 500 or 600 shops, but there are tens of thousands of
barbershops, so it’s just a minority. It’s not a disruptor going to turn over the entire
158
industry. QB didn’t have the earnings power to upend the industry. Although it was very
innovative, but it’s just a small part of the industry (Interview 20).
Employees at two public university-based venture capital firms also asserted that Japan’s
lack of startups has been partly an outcome of prior government preference to directly invest in
new firms (Interview 14), and inadequate managerial and marketing skills (Interview 27), rather
than barriers imposed by an abundance of unproductive firms in the same economic sector.
4. This dissertation finds that certain features of Japan’s credit guarantee system contributed to
the high incidence of zombie firms among the country’s SME. Since the dissertation’s analysis of
zombie SME ends in 2014, however, it does not address how the credit guarantee system was
eventually reformed in July 2017. This reform, which was a goal of the Japan Revitalization
Strategy launched by the Abe government in June 2015, was gradually prepared through a series
of deliberative council meetings (shingikai) from late 2015 until December 2016. In February
2017, a bill encapsulating the ideas from these meetings was approved and forwarded by the Abe
cabinet to Japan’s Diet (Yamori 2017, 38), and in July 2017 the bill became law.
The 2017 reform of the credit guarantee system has three main components: 1) it sets a
fixed usage period for 100 percent guarantee coverage after major economic crises and reduces
the “safety net” guarantee coverage to 80 percent, two changes aimed at promoting greater
financial scrutiny by banks toward borrowing SME and thereby incentivizing the firms to quickly
restructure; 2) it increases the amount of loans eligible for 100 percent guarantee coverage for
startup firms (from 10 million yen to 20 million yen) and microenterprises (from 12.5 million yen
to 20 million yen); and 3) it encourages better risk-sharing between the system and private
financial institutions by ordering credit guarantee corporations to collect and disclose information
159
on financial institutions’ non-guaranteed, “proper” loans (Uesugi 2017). Collectively, these
changes portend to reduce of number of zombie firms among Japanese SME.
Future work explaining how this reform happened might start with two variables: 1) the
thinking of the SMEA Finance Department, which fieldwork interviews suggest was strongly
amenable to shifting the credit guarantee system’s emphasis from supporting existing firms to
assisting startups in December 2015, if not earlier; and 2) the role of budgetary constraint, a
factor that became more prominent after the system was last reformed in 2007. A SMEA
Finance Department bureaucrat interviewed in December 2015 said that the timing then was
good for revising the system, since there were concerns about a possible future shock in China
and emerging markets and the eventual rise of US interest rates (Interview 9). The bureaucrat’s
unambiguous pro-reform position stood in notable contrast to the hesitancy of a bureaucrat from
the same department in the mid-2000s, who reminisced, “METI has a responsibility to maintain
the economy, to grow the economy. But they don’t want to have a huge, drastic impact to one of
the specific sectors. So we had that kind of system. We didn’t, we couldn’t want to change
drastically. Maybe you can say that it is kind of an ambivalent feeling, but it’s true” (Interview
59).
5. While it is generally understood that local Chinese officials’ promotion incentives drive them
to act to prop up weak firms, it is less clear what effect this incentive structure has on the
aggregate creation of zombie firms. It is also unclear whether inter-provincial growth
competition, spurred in part by the 1994 tax system reform, has possibly contributed to the
incidence of zombies, and how the central government effort to promote mergers of small firms
into large firms after the global financial crisis may have ignited a sort of prisoners’ dilemma
between various government departments trying to protect their preferred firms by enlarging
160
them, which made the firms themselves lose competitiveness. These potential political causes of
zombie firms are ripe for further empirical study.
The Future of Zombie Firms
While a growing body of economics research and journalism has called attention to the
problem of zombie firms as an obstacle to post-crisis recovery, little if any work until now has
explored the political reasons why these troubled firms occur and obstruct recovery by impeding
creative destruction. This dissertation represents the first word on this important topic from a
political economy perspective. By examining this troubling corporate phenomenon that first
received widespread attention in Japan’s lost decades after the banking crisis in the 1990s and has
more recently been identified in numerous developed and developing countries since the global
financial crisis, it has attempted to advance our understanding of policy configurations that can
stifle post-crisis regeneration and growth. It hopes to have opened a debate about how and why
one of the most notorious elements of “Japanization” has occurred in Japan and beyond.
Appendix: List of Field Interviews
Institution
Number Date Interview Site
(proposed or
actual)
Admin.
Level
Prefecture,
Province
or State
Country Position Name Type
Scheduling
conflict
8/29/15 Palo Alto NA California USA professor Stanford University academia
1 10/13/15 Kasumigaseki NA Tokyo Japan researcher RIETI research
2 11/5/15 Otemachi NA Tokyo Japan economist Japan Finance
Corporation
government
finance
institution
3 11/12/15 Kasumigaseki national Tokyo Japan former official SMEA, METI government
4 11/30/15 Tachikawa NA Tokyo Japan director Startup Support
Center, Japan
Finance Corporation
government
finance
institution
5 12/3/15 Higashi
Yamato-shi
regional Tokyo Japan manager SME University,
SMRJ
government
6 12/4/15 Kokubunji municipal Tokyo Japan policy planning
official (1); city
council member
(2)
-- government
Scheduling
conflict
12/7/15 Yaesu NA Tokyo Japan president -- venture
capital firm
7 12/14/15 Kasumigaseki national Tokyo Japan official
Microbusiness
Promotion
Deparment, SMEA,
METI
government
8 12/18/15
Waseda NA Tokyo Japan professor
Waseda University academia
9 (E) 12/22/15 Kasumigaseki national Tokyo Japan official (1);
official (2)
Finance Department,
SMEA, METI
government
10 (E) 1/27/16 Waseda NA Tokyo Japan professor Waseda University academia
162
11 2/5/16
Toranomon national Tokyo Japan managing
director
--
consultancy
12 2/12/16
Kasumigaseki national Tokyo Japan director (1);
manager (2)
Office for
International
Cooperation, SMEA,
METI (1);
International Center,
Organization for
Small & Medium
Enterprises and
Regional Innovation
(SMRJ) (2)
government
13 2/25/16
Kasumigaseki national Tokyo Japan official (1);
official (2)
Innovation
Department, SMEA,
METI
government
Request
declined
2/28/18 Kanda national Tokyo Japan Regional Banks
Association of Japan
bank
association
14 3/3/16 Hongo NA Tokyo Japan president -- venture
capital firm
Scheduling
conflict
3/3/16 Nagatacho national Tokyo Japan member House of Councillors,
Democratic Party of
Japan
politician
Scheduling
conflict
3/3/16 Nagatacho national Tokyo Japan member
House of Councillors,
Liberal Democratic
Party
politician
Scheduling
conflict
3/3/16 Nagatacho national Tokyo Japan member House of
Representatives,
Liberal Democratic
Party
politician
15 (E) 3/11/16
Kasumigaseki NA Tokyo Japan senior
administrative
officer
Asia Development
Bank Institute
research
Scheduling
conflict
3/27/16 Nagatacho national Tokyo Japan special advisor Cabinet Office government
163
16 3/29/16
Kasumigaseki national Tokyo Japan official
University-Industry
Collaboration and
Regional R&D
Division, Ministry of
Education, Culture,
Sports, Science and
Technology
government
17 (E) 5/11/16 Mita NA Tokyo Japan professor Keio University academia
18 5/13/16 Roppongi NA Tokyo Japan professor National Graduate
Institute for Policy
Studies
academia
19 6/6/16 Waseda NA Tokyo Japan professor Waseda University academia
20 6/7/16 Toyota-makami NA Tokyo Japan professor Musashi University academia
21 6/8/16 Otemachi national Tokyo Japan researcher Development Bank of
Japan
government
finance
institution
22 6/16/16 Shinjuku national Tokyo Japan director
(retired)
Ministry of Finance government
Request
made
6/19/16 NA NA Tokyo Japan Japan Subculture
Research Center
research
23 6/20/16 Kudanminami national Tokyo Japan secretary National Conference
of the Association of
Small Business
Entrepreneurs
SME
association
Scheduling
conflict
6/24/16 Imabashi regional Osaka Japan special advisor Nippon Venture
Capital
venture
capital firm
Request
declined
6/27/16 Kyoto NA Kyoto Japan professor Ritsumeikan
University
academia
24 7/4/16 Sugimoto NA Osaka Japan professor Osaka City
University
academia
25 7/6/16 Kawachinagano NA Osaka Japan chairman -- small
business
26 7/6/16 Osaka NA Osaka Japan executive officer Resona Bank bank
27 7/8/16 Suita NA Osaka Japan executive Osaka University
Venture Capital
venture
capital firm
28 7/8/16 Suita NA Osaka Japan professor Osaka University academia
164
29 7/26/16 Shinjuku NA Tokyo Japan former loan
officer
Mizuho bank
30 7/27/16 Otemachi national Tokyo Japan executive Japan Finance
Corporation
government
finance
institution
31 (E) 7/28/16 Otemachi national Tokyo Japan executive Morgan Stanley UFJ bank
32 (E) 7/29/16 Kasumigaseki NA Tokyo Japan researcher Research Institute of
Economy, Trade and
Industry
research
33 8/2/16 Kasumigaseki NA Tokyo Japan professor Kobe University academia
34 8/5/16 Otemachi national Tokyo Japan economist (1);
economist (2)
Japan Finance
Corporation
government
finance
institution
35 8/22/16 Shinkawa national Tokyo Japan assistant
director
National Federation
of Small Business
Associations
(Chuokai)
SME
association
36 8/23/16 Yurakucho national Tokyo Japan section assistant Central Federation of
Societies of
Commerce and
Industry
SME
association
37 8/24/16 Hongo national Tokyo Japan director Ii Kaisha (Good
Company) NPO
SME
association
38 8/29/16 Shinkawa national Tokyo Japan vice director National Union for
the Promotion of
Retail Malls
SME
association
39 (C) 10/16/16 Hangzhou NA Zhejiang China bankruptcy
lawyer
Zhejiang People’s
Union Law Firm
legal service
40 (C) 10/16/16 Hangzhou NA Zhejiang China lawyer Capital Equity Legal
Group
legal service
41 (C) 10/17/16 Hangzhou NA Zhejiang China researcher Alibaba corporation
42 (C) 10/17/16 Wenzhou NA Zhejiang China lawyer Zhejiang People’s
Union Law Firm
legal service
43 (C) 10/18/16 Wenzhou district Zhejiang China judge Ouhai District Court legal
44 (C) 10/18/16 Wenzhou district Zhejiang China judge (retired) Ouhai District Court legal
165
45 (C) 10/19/16 Hangzhou NA Zhejiang China lawyer Shanghai Tongyi
Law Firm
legal service
46 (C) 10/21/16 Shanghai NA Shanghai China professor Shanghai Jiaotong
University
research
47 (C) 10/21/16 Hangzhou NA Zhejiang China lawyer -- legal service
48 (C) 10/21/16 Hangzhou NA Zhejiang China lawyer (1);
lawyer (2)
-- legal service
49 (C) 10/24/16 Guangzhou NA Guangdong China lawyers -- legal service
50 (C) 10/26/16 Shenzhen NA Guangdong China former judge;
lawyer
Shenzhen District
Court; --
legal service
51 (C) 10/26/16 Shenzhen NA Guangdong China lawyer -- legal service
52 (C) 10/28/16 Beijing NA Beijing China professor China University of
Political Science and
Law
academia
53 (C) 11/1/16 Beijing NA Beijing China professor Renmin University academia
54 (C) 11/1/16 Beijing NA Beijing China researcher Department of
Economics, Chinese
Academy of Social
Science
research
55 (C) 11/2/16 Beijing NA Beijing China banker (retired) China Construction
Bank
bank
56 (C) 11/2/16 Beijing NA Beijing China lawyer -- legal service
57 (C) 11/4/16 Beijing NA Beijing China researcher Department of
Industrial Economics,
Chinese Academy of
Social Science
research
58 11/16/16 Hiroshima City ward Tokyo Japan former city
council member
Nerima District,
Tokyo
politician
59 (E) 11/21/16 Mita national Tokyo Japan former deputy
director
Finance Department,
SMEA, METI
government
60 11/28/16 Nishi-Kanda national Tokyo Japan former member House of
Representatives,
Democratic Party of
Japan
politician
61 11/29/16 Matsuyama NA Ehime Japan professor Ehime University academia
62 11/29/16 Matsuyama prefectural Ehime Japan director -- shinkin bank
166
63 11/29/16 Matsuyama prefectural Ehime Japan deputy general
manager
-- regional
bank
64 11/30/16 Matsuyama prefectural Ehime Japan director -- credit
guarantee
corporation
65 12/7/16 Kyoto NA Kyoto Japan professor Doshisha University academia
66 12/8/16 Akasaka NA Tokyo Japan company
director
-- small
business
67 7/28/17 Kanda national Tokyo Japan executive National Federation
of Credit Guarantee
Corporations
SME finance
association
Note: All interviews conducted in Japanese unless otherwise noted by C (conducted in Chinese) or E (conducted in English). The
mark “--” represents identifying information that has been redacted to maintain interview subject anonymity.
References
Acemoglu, Daron, Simon Johnson, and James A. Robinson. 2005. “Institutions as a
Fundamental Cause of Long-Run Growth.” In Handbook of Economic Growth, 385–472.
Acharya, Viral V., Tim Eisert, Christian Eufinger, and Christian W. Hirsch. 2017. “Whatever It
Takes: The Real Effects of Unconventional Monetary Policy.” SAFE Working Paper 152:
1–67.
Ahearne, Alan, and Naoki Shinada. 2005. “Zombie Firms and Economic Stagnation in Japan.”
International Economics and Economic Policy 2(4): 363–81.
Amway. 2015. Amway Global Entrepreneurship Report 2015: Defining the Entrepreneurial Spirit.
Amyx, Jennifer A. 2004. Japan’s Financial Crisis: Institutional Rigidity and Reluctant Change. Princeton:
Princeton University Press.
Anderson, Kent, and Makoto Ito. 2007. “Insolvency Law for a New Century: Japan’s Revised
Framework for Economic Failures.” In Law in Japan: A Turning Point, ed. Daniel H. Foote.
Seattle: University of Washington Press, 589–621.
Andrade, Gregor, and Steven N. Kaplan. 1998. “How Costly Is Financial (Not Economic)
Distress? Evidence from Highly Leveraged Transactions That Became Distressed.”
Journal of Finance 53(5): 1443–93.
Ang, Yuen Yuen. 2016. How China Escaped the Poverty Trap. Ithaca: Cornell University Press.
ANU TV. 2013. Why Did Japan Stop Growing? Professor Takeo Hoshi at Australia National University.
Canberra. https://www.youtube.com/watch?v=8xpaAPTYAVo (May 28, 2016).
Arbatli, Elif et al. 2016. “Reflating Japan: Time to Get Unconventional?” IMF Working Paper
16(157): 1–45.
Bank of England. 2012. Financial Stability Report.
Bank of Japan Financial Research Institute. 1989. 19 Nihon Kin’yu Shi Shiryo [Japan Financial History
Documents]. Tokyo: Ministry of Finance Printing Bureau.
Barnett, Michael N., and Martha Finnemore. 1999. “The Politics, Power, and Pathologies of
International Organizations.” International Organization 53(4): 699–732.
Berkovitch, Elazar, and Ronen Israel. 1998. “The Bankruptcy Decision and Debt Contract
Renegotiations.” Review of Finance 2(1): 1–27.
Bird, Mike. 2015. “France’s Zombie Economy Is Missing Out on Europe’s Recovery.” Business
Insider. http://uk.businessinsider.com/frances-zombie-economy-is-missing-out-on-
europes-recovery-2015-3 (May 19, 2016).
168
Bloomberg. 2017. “Peterson Inst.’s Lardy Doesn’t Expect China Debt Crisis.”
https://www.bloomberg.com/news/videos/2017-03-23/peterson-inst-s-lardy-doesn-t-
expect-china-crisis-video (June 11, 2017).
Blyth, Mark. 2002. Great Transformations: Economic Ideas and Institutional Change in the Twentieth
Century. New York: Cambridge University Press.
Buckley, Chris. 2017. “In Rare Move, Chinese Think Tank Criticizes Tepid Pace of Reform.”
The New York Times. https://www.nytimes.com/2017/03/27/business/chinese-economy-
reform-critical-report.html (March 28, 2017).
Buckley, Chris, and Keith Bradsher. 2017. “Xi Jinping’s Failed Promises Dim Hopes for
Economic Change in 2nd Term.” The New York Times.
https://www.nytimes.com/2017/03/04/world/asia/china-xi-jinping-economic-
reform.html (March 5, 2017).
Buckley, Chris, and Jane Perlez. 2016. “U.S. to Press China to Curb Industrial Output.” The New
York Times. http://www.nytimes.com/2016/06/06/business/us-to-press-china-to-curb-
industrial-output.html (June 6, 2016).
Bueno de Mesquita, Bruce, Alastair Smith, Randolph M. Siverson, and James D. Morrow. 2003.
The Logic of Political Survival. Cambridge: MIT Press.
Caballero, Ricardo J., and Mohamad L. Hammour. 1994. “The Cleansing Effect of Recessions.”
American Economic Review 84(5): 1350–68.
———. 1996. “On the Timing and Efficiency of Creative Destruction.” The Quarterly Journal of
Economics 111(3): 805–52.
Caballero, Ricardo J., Takeo Hoshi, and Anil K. Kashyap. 2008. “Zombie Lending and
Depressed Restructuring in Japan.” The American Economic Review 98(5): 1943–77.
Caixin. 2016. “Zhou Fangsheng: Jiang Gongzi Buneng Jiejue Guoyou Qiye Wenti [ Zhou
Xiansheng: Lowering Wages Cannot Solve the Problem of State-Owned Enterprises].”
Caixin Wang. http://finance.caixin.com/2016-05-31/100949709.html (June 25, 2016).
Calder, Kent E. 1988. Crisis and Compensation: Public Policy and Political Stability in Japan, 1949-1986.
Princeton: Princeton University Press.
Campana, Paolo, and Ken Okamura. 2016. “Lawyers, Yakuza and Zombie Firms: A Quasi-
Natural Experiment.” Saïd Business School Research Papers 24: 1–31.
Campbell, John Creighton, and Ethan Scheiner. 2008. “Fragmentation and Power:
Reconceptualizing Policy Making under Japan’s 1955 System.” Japanese Journal of Political
Science 9(1): 89–113.
169
Capoccia, Giovanni, and R. Daniel Kelemen. 2007. “The Study of Critical Junctures: Theory,
Narrative, And Counterfactuals in Historical Institutionalism.” World Politics 59(3): 341–
69.
Chan, Szu Ping. 2015. “Japan’s ‘Zombie’ Firms Holding Back Recovery.” The Telegraph.
http://www.telegraph.co.uk/finance/economics/11426119/Japans-zombie-firms-
holding-back-recovery.html (May 11, 2016).
Chandler, Clay. 1999. “In Japan, a Reversal on Reform?: Premier Draws Fire for ‘Go Slow’
Stance on Economy.” Washington Post: E1.
Christensen, Jesper Lindgaard, S. Jackson, Allan Riding, and S. Mensah. 1999. Loan Guarantee
Schemes in Six Countries. Copenhagen: Danish Ministry of Industry.
Cohen, Benjamin J. 2008. “After the Fall: East Asian Exchange Rates since the Crisis.” In Crisis
as Catalyst: Asia’s Dynamic Political Economy, eds. Andrew MacIntyre, T. J. Pempel, and John
Ravenhill. Ithaca: Cornell University Press, 25–44.
———. 2009. “A Grave Case of Myopia.” International Interactions 35(4): 436–44.
———. 2010. “Are IPE Journals Becoming Boring?” International Studies Quarterly 54(3): 887–91.
Collier, Ruth B., and David Collier. 1991. Shaping the Political Arena: Critical Junctures, the Labor
Movement, and Regime Dynamics in Latin America. Princeton: Princeton University Press.
Cooley, Thomas F. 2009. “Zombie Firms and Zombie Banks.” Forbes.
http://www.forbes.com/2009/02/10/recession-tarp-japan-opinions-
columnists_0211_thomas_cooley.html (May 19, 2016).
Cox, W. Michael, and Richard Alm. 1992. Federal Reserve Bank of Dallas Annual Report: The Churn:
The Paradox of Progress.
Culpepper, Pepper D. 2011. Quiet Politics and Business Power: Corporate Control in Europe and Japan.
New York: Cambridge University Press.
Darvas, Zsolt. 2013. “Banking System Soundness Is the Key to More SME Financing.” Bruegel
Policy Contribution 10: 1–18.
De Grauwe, Paul. 2017. The Limits of the Market: The Pendulum between Government and Market.
Oxford: Oxford University Press.
Ding, Jun. 2015. “Xi Jinping: Tuijin Jingji Jiegou Xing Gaige Xuyao ‘5+4+3’ [Xi Jinping: To
Promote Economic Structural Reform Needs ‘5+4+3’].” Ershiyi Shiji Jingji Baodao.
http://epaper.21jingji.com/html/2015-11/11/content_25525.htm (August 3, 2017).
Donnan, Shawn. 2016. “IMF Warns Chinese Response to Debt Needs to Be More
Comprehensive.” Financial Times. http://www.ft.com/cms/s/0/808a6d8c-0be9-11e6-
9cd4-2be898308be3.html (April 27, 2016).
170
Donnan, Shawn, and Tom Mitchell. 2016a. “IMF Sounds Warning on China’s Corporate
Debt.” Financial Times. http://www.ft.com/intl/cms/s/0/3f8dcf22-304c-11e6-bda0-
04585c31b153.html#axzz4BKevR6e8 (June 12, 2016).
———. 2016b. “Renminbi Displaced as US Finds New China Bugbear.” Financial Times.
http://www.ft.com/intl/cms/s/0/dd5c1292-2870-11e6-8ba3-
cdd781d02d89.html#axzz4APedeNw9 (June 2, 2016).
Drazen, Allan, and William Easterly. 2001. “Do Crises Induce Reform? Simple Empirical Tests
of Conventional Wisdom.” Economics & Politics 13(2): 129–57.
Du, Juan. 2016. “Lu Zai Jiaoxia: ‘Jiangshi Qiye’ Bixu Jieshou Xianshi [Road at Our Feet:
‘Zombie Firms’ Must Accept Reality].” Zhongguo Ribao.
http://cn.chinadaily.com.cn/2016-03/29/content_24153537.htm (July 31, 2017).
Fackler, Martin. 2003. “Japan Targets Bad Bank Debt: Effort to Help Lenders Take Over
Corporate Borrowers.” Wall Street Journal: C9.
Feldman, Robert Alan. 2001. Into a Worried World. Tokyo: Morgan Stanley.
de Figueiredo, Rui J. P. 2002. “Electoral Competition, Political Uncertainty, and Policy
Insulation.” The American Political Science Review 96(2): 321–33.
Financial Times. “Definition of Zombie Company.” Financial Times.
http://lexicon.ft.com/Term?term=zombie-company (May 3, 2016).
Fujii, Mariko, and Masahiro Kawai. 2010. “Lessons from Japan’s Banking Crisis, 1991–2005.”
ADBI Working Paper Series 222: 1–19.
Fukao, Mitsuhiro. 2000. “Recapitalizing Japan’s Banks: The Functions and Problems of
Financial Revitalization Act and Bank Recapitalization Act.” Keio Business Review 38: 1–
16.
———. 2007. “Financial Crisis and the Lost Decade.” Asian Economic Policy Review 2(2): 273–97.
Fukuda, Shin-ichi, Munehisa Kasuya, and Jouchi Nakajima. 2006. “Deteriorating Bank Health
and Lending in Japan: Evidence from Unlisted Companies under Financial Distress.”
Journal of the Asia Pacific Economy 11(4): 482–501.
Fukuda, Shin-ichi, and Jun‐ichi Nakamura. 2011. “Why Did ‘Zombie’ Firms Recover in Japan?”
The World Economy 34(7): 1124–37.
Gailmard, Sean, and John W. Patty. 2007. “Slackers and Zealots: Civil Service, Policy
Discretion, and Bureaucratic Expertise.” American Journal of Political Science 51(4): 873–89.
Gandhi, Jennifer, and Adam Przeworski. 2006. “Cooperation, Cooptation, and Rebellion under
Dictatorships.” Economics & Politics 18(1): 1–26.
171
Garside, W. R. 2012. Japan’s Great Stagnation: Forging Ahead, Falling Behind. Northampton: Edward
Elgar Publishing, Inc.
Geddes, Barbara. 1990. “How the Cases You Choose Affect the Answers You Get: Selection
Bias in Comparative Politics.” Political Analysis 2: 131–50.
George, Alexander L., and Andrew Bennett. 2005. Case Studies and Theory Development in the Social
Sciences. Cambridge: MIT Press.
Gerring, John. 2007. Case Study Research: Principles and Practices. New York: Cambridge University
Press.
Global Entrepreneurship Monitor. 2014. “Japan Country Profile.”
http://www.gemconsortium.org/country-profile/76 (June 12, 2016).
Golub, Adam, and Carrie Lane. 2015. “Zombie Companies and Corporate Survivors.”
Anthropology Now 7(2): 47–54.
Gomes, Joao, Jeremy Greenwood, and Sergio Rebelo. 2001. “Equilibrium Unemployment.”
Journal of Monetary Economics 48(1): 109–52.
Goto, Yasuo, and Scott Wilbur. Forthcoming. “Efficiency Among Japanese SME: In the Context
of Zombie Firm Hypothesis and Firm Size.” RIETI Discussion Paper Series.
Gough, Neil. 2016. “As China’s Growth Slows, Banks Feel the Strain of Bad Debt.” The New York
Times. http://www.nytimes.com/2016/04/16/business/dealbook/china-economy-bad-
loans.html (April 22, 2016).
Gourevitch, Peter Alexis. 1986. Politics in Hard Times: Comparative Responses to International Economic
Crises. Ithaca: Cornell University Press.
Government Of Japan Cabinet Office. Multiple years. “Kokumin Seikatsu Ni Kansuru
Seronchosa [Public Opinion Poll on Public Life].” http://survey.gov-online.go.jp/index-
ko.html (October 28, 2017).
Griffin, Naomi N., and Kazuhiko Odaki. 2009. “Reallocation and Productivity Growth in Japan:
Revisiting the Lost Decade of the 1990s.” Journal of Productivity Analysis 31(2): 125–36.
Grimes, William W. 2001. Unmaking the Japanese Miracle: Macroeconomic Politics, 1985-2000. Ithaca:
Cornell University Press.
Guilford, Gwynn. 2014. “Zombies Once Destroyed Japan’s Economy: Now They’re Infecting
China’s.” Quartz. https://qz.com/198458/zombies-once-destroyed-japans-economy-
now-theyre-infecting-chinas/ (February 1, 2017).
Hacker, Jacob, Paul Pierson, and Kathleen Thelen. 2015. “Drift and Conversion: Hidden Faces
of Institutional Change.” In Advances in Comparative Historical Analysis, eds. James Mahoney
and Kathleen Thelen. Cambridge: Cambridge University Press, 180–209.
172
Hall, Robert E., Henry Farber, and John Haltiwanger. 1995. “Lost Jobs.” Brookings Papers on
Economic Activity 1: 221–73.
Hardin, Russell. 1989. “Why a Constitution?” In The Federalist Papers and the New Institutionalism,
eds. Bernard Grofman and Donald Wittman. New York: Agathon Press, 100–119.
Hartcher, Peter. 1998. The Ministry: How Japan’s Most Powerful Institution Endangers World Markets.
Boston: Harvard Business School Press.
He, Fan, and He Zhu. 2016. “Hei, Wo Faxianle Yi Zhi Jiangshi Qiye: Jiangshi Qiye Xilie Yanjiu
Zhi Yi [Hey, I Found a Zombie Firm: Part One in the Zombie Firm Research Series].”
Caizhi Yanjiu.
http://mp.weixin.qq.com/s?__biz=MzA4MzI4MzY1MQ==&mid=401889437&idx=1
&sn=dbd11c5c9a90e62316e9bb926c139811&scene=1&srcid=0112oSrTuXjfOfZDBeKe
Cc0H&from=singlemessage&isappinstalled=0#wechat_redirect (February 1, 2017).
Hendrickson, Mark. 2013. “ZIRP, Zombies and the Japanization of the American Economy.”
Forbes. http://www.forbes.com/sites/markhendrickson/2013/01/03/zirp-zombies-and-
the-japanization-of-the-american-economy/ (December 29, 2016).
Henry, David. 1997. “Japan Forced to Clear Out Deadwood.” USA Today: B3.
Hicken, Allen, Shanker Satyanath, and Ernest Sergenti. 2005. “Political Institutions and
Economic Performance: The Effects of Accountability and Obstacles to Policy Change.”
American Journal of Political Science 49(4): 897–907.
Hill, Peter B. E. 2003. The Japanese Mafia: Yakuza, Law, and the State. Oxford: Oxford University
Press.
Hirose, Sumio. 2009. “Effects of the Bankruptcy Laws Reform on Banks: The Examination of
Recent Japanese Experience.” Public Policy Review 5(2): 201–28.
Hirose, Sumio, and Fumio Akiyoshi. 2010. “Tosan Shori Hosei Kaikaku Ni Yoru Kigyo Tosan
Shori Koritsu-ka No Kensho [Validation of Improving Corporate Bankruptcy Processing
Efficiency by Bankruptcy Processing Legal Reform].” Keizai Kenkyu 61(3): 193–202.
Hoshi, Takeo. 2006. “Economics of the Living Dead.” Japanese Economic Review 57(1): 30–49.
Hoshi, Takeo, and Anil K. Kashyap. 2001. Corporate Financing and Governance in Japan: The Road to
the Future. Cambridge: MIT Press.
———. 2010. “Will the U.S. Bank Recapitalization Succeed? Eight Lessons from Japan.” The
2007-8 Financial Crisis: Lessons from Corporate Finance 97(3): 398–417.
———. 2011. Why Did Japan Stop Growing? Tokyo: National Institute for Research Advancement.
173
Hu, Shuli. 2015. “Shuli Guancha: Diao Jiegou Xu Maizang "Jiangshi Qiye” [Shuli’s
Observation: Structural Adjustment Requires Burying ‘Zombie Firms’].” Caixin.
http://m.opinion.caixin.com/m/2015-10-26/100866623.html (January 28, 2017).
Huang, Yasheng. 2008. Capitalism with Chinese Characteristics: Entrepreneurship and the State. New
York: Cambridge University Press.
Huber, John D., and Charles R. Shipan. 2002. Deliberate Discretion: The Institutional Foundations of
Bureaucratic Autonomy. New York: Cambridge University Press.
Imai, Kentaro. 2016. “A Panel Study of Zombie SMEs in Japan: Identification, Borrowing and
Investment Behavior.” Journal of the Japanese and International Economies 39: 91–107.
International Monetary Fund. 2016a. “Global Financial Stability Report.”
———. 2016b. The People’s Republic of China. Washington: International Monetary Fund.
Ito, Akiko. 2011. “Koteki Shin’yo Hosho Seido Ni Okeru Moraruhazado Ni Kansuru Kosatsu
[A Study on Moral Hazard in the Public Credit Guarantee System].” National Graduate
Institute for Policy Studies.
Japan Exchange Group. 2015. “Criteria for Listing.” Japan Exchange Group.
http://www.jpx.co.jp/english/equities/listing/criteria/listing/ (April 25, 2017).
Jaskowski, Marcin. 2015. “Should Zombie Lending Always Be Prevented?” International Review of
Economics & Finance 40: 191–203.
Jiang, Yujia. 2014. “The Curious Case of Inactive Bankruptcy Practice in China: A Comparative
Study of U.S. and Chinese Bankruptcy Law.” Northwestern Journal of International Law &
Business 34(3): 559–82.
Johnson, Chalmers A. 1982. MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925-
1975. Stanford: Stanford University Press.
Jones, Claire, and Chris Giles. 2012. “Bank Turns Gloomy on Recovery.” Financial Times.
http://www.ft.com/intl/cms/s/0/518c4a60-2e47-11e2-8bb3-
00144feabdc0.html#axzz47YFYd0jQ (May 3, 2016).
Juan, Du. 2016. “‘Zombie Companies’ Adjust to New Reality.” China Daily.
http://www.chinadaily.com.cn/kindle/2016-03/15/content_23874376.htm (June 14,
2016).
Kane, Edward J. 1987. “Dangers of Capital Forbearance: The Case of the FSLIC and ‘Zombie’
S&Ls.” Contemporary Economic Policy 5(1): 77–83.
Kaplan, David E., and Alec Dubro. 2012. Yakuza: Japan’s Criminal Underworld. Expanded edition.
Berkeley: University of California Press.
Katz, Richard. 2003a. “A Bumpy Road to Reform.” Asian Wall Street Journal: A9.
174
———. 2003b. Japanese Phoenix: The Long Road to Economic Revival. Armonk: M.E. Sharpe.
Kaufman, Robert R. 2017. “Great Transformations but No Critical Juncture? Latin America in
the Twenty-First Century.” Qualitative and Multi-Method Research 15(1): 16–17.
Kawai, Masahiro, and Peter Morgan. 2013. “Banking Crises and ‘Japanization’: Origins and
Implications.” ADBI Working Paper Series 430: 1–35.
Kennedy, Scott. 2011. “Overcoming Our Middle Kingdom Complex: Finding China’s Place in
Comparative Politics.” In Beyond the Middle Kingdom: Comparative Perspectives on China’s
Capitalist Transformation, ed. Scott Kennedy. Stanford: Stanford University Press, 3–21.
King, Gary, Robert O. Keohane, and Sidney Verba. 1994. Designing Social Inquiry: Scientific
Inference in Qualitative Research. Princeton: Princeton University Press.
Kinoshita, Nobuyuki. 2013. “Legal Background to the Low Profitability of Japanese
Enterprises.” Columbia Business School Center on Japanese Economy and Business Working Paper
Series 316: 1–88.
Kobayashi, Keiichiro. 2009. “Financial Crisis Management: Lessons from Japan’s Failure.” In
The First Global Financial Crisis of the 21st Century: Part II: June-December, 2008, eds. Andrew
Felton and Carmen M. Reinhart. Centre for Economic Policy Research, 337–41.
Koo, Richard C. 2008. The Holy Grail of Macroeconomics: Lessons from Japan’s Great Recession.
Singapore: Wiley.
Kornai, János. 1979. “Resource-Constrained Versus Demand-Constrained Systems.”
Econometrica 47(4): 801–19.
———. 1992. The Socialist System: The Political Economy of Communism. Princeton: Princeton
University Press.
Kornai, János, Eric Maskin, and Gérard Roland. 2003. “Understanding the Soft Budget
Constraint.” Journal of Economic Literature 41(4): 1095–1136.
Krause, George A. 2010. “Legislative Delegation of Authority to Bureaucratic Agencies.” In The
Oxford Handbook of American Bureaucracy, ed. Robert F. Durant. Oxford: Oxford University
Press, 521–44.
Krauss, Ellis S., and Robert J. Pekkanen. 2010. “The Rise and Fall of Japan’s Liberal
Democratic Party.” The Journal of Asian Studies 69(1): 5–15.
Kume, Ikuo. 2009. “Koteki Shikin Chunyu Wo Meguru Seiji Katei [Political Process of Injection
of Public Funds to the Financial Sector].” In Furyo Saiken To Kinyu Kiki [Bad Loans and the
Financial Crisis], ed. Kazuhito Ikeo. Tokyo: Keio Gijyuku Shuppankai, 215–49.
Kwon, Hyeog Ug, Futoshi Narita, and Machiko Narita. 2015. “Resource Reallocation and
Zombie Lending in Japan in the 1990s.” Review of Economic Dynamics 18(4): 709–32.
175
Lam, Raphael W., and Jongsoon Shin. 2012. “What Role Can Financial Policies Play in
Revitalizing SMEs in Japan?” IMF Working Paper 12(291): 1–23.
Landers, Peter. 1999. “Tokyo Still Needs Plan to Clean Up Bad Loans Stifling Firms’ Growth.”
Asian Wall Street Journal: 1.
Lardy, Nicholas. 2014. Markets over Mao: The Rise of Private Business in China. Washington: Peterson
Institute for International Economics.
———. 2016. “No Need to Panic, China’s Banks Are in Pretty Good Shape.” Financial Times.
http://www.ft.com/intl/cms/s/0/5bfb049a-2287-11e6-9d4d-
c11776a5124d.html?siteedition=intl#axzz4ANqDxVxD (June 2, 2016).
Lawrence, Susan V. 2000. “Mr. Privatization.” Far Eastern Economic Review 163(9): 44–45.
Lee, Jiyeun. 2015. “Zombie Companies in South Korea Spur Restructuring Push.” Bloomberg.com.
http://www.bloomberg.com/news/articles/2015-11-23/zombie-companies-in-south-
korea-spur-restructuring-push (May 3, 2016).
Levinson, Daryl, and Benjamin I. Sachs. 2015. “Political Entrenchment and Public Law.” Yale
Law Journal 125(2): 400–482.
Lewis, Leo, Tom Mitchell, and Yuan Yang. 2017. “Is China’s Economy Turning Japanese?”
Financial Times. https://www.ft.com/content/a3564812-363c-11e7-99bd-13beb0903fa3
(May 29, 2017).
Li, Jin. 2015. “Renmin Caiping: Wuda Lujing Jiejue ‘Jiangshi Qiye’ Qing Tui Nanti [People’s
Financial Review: Five Paths to Solve ‘Zombie Firms’].” Renmin Wang Caijing Pindao.
http://finance.people.com.cn/n1/2015/1217/c1004-27940565.html (July 31, 2017).
Lieberthal, Kenneth G. 1992. “Introduction: The ‘Fragmented Authoritarianism’ Model and Its
Limitations.” In Bureaucracy, Politics, and Decision-Making in Post-Mao China, eds. Kenneth G.
Lieberthal and David M. Lampton. Berkeley: University of California Press, 1–31.
Lin, Li-Wen, and Curtis J. Milhaupt. 2013. “We Are the (National) Champions: Understanding
the Mechanisms of State Capitalism in China.” Stanford Law Review 65(4): 697–759.
Lincoln, Edward J. 2011. “The Heisei Economy: Puzzles, Problems, Prospects.” The Journal of
Japanese Studies 37(2): 351–75.
Lipscy, Phillip Y., and Hirofumi Takinami. 2013. “The Politics of Financial Crisis Response in
Japan and the United States.” Japanese Journal of Political Science 14(3): 321–53.
Mabuchi, Masaru. 1997. Okurasho Wa Naze Oitsumeraretanoka: Seikan Kankei No Henbo [Why Was the
Ministry of Finance Cornered: Changes in Politician-Bureaucrat Relations]. Tokyo: Chuko Shinsho.
176
MacAskill, Andrew, and Manoj Kumar. 2014. “Modi Seeks to Revive India’s Zombie Factories,
Not Abandon Them.” Reuters. http://in.reuters.com/article/india-industry-revival-modi-
idINKCN0I202C20141013 (April 26, 2016).
MacIntyre, Andrew. 2001. “Institutions and Investors: The Politics of the Economic Crisis in
Southeast Asia.” International Organization 55(1): 81–122.
Mahoney, James. 2007. “Qualitative Methodology and Comparative Politics.” Comparative
Political Studies 40(2): 122–44.
Mahoney, James, and Kathleen Thelen. 2010. “A Theory of Gradual Institutional Change.” In
Explaining Institutional Change: Ambiguity, Agency, and Power, eds. James Mahoney and
Kathleen Thelen. Cambridge: Cambridge University Press, 1–38.
Makinen, Julie. 2015. “A Subculture of Entrepreneurship Hatches in Japan.” Los Angeles Times.
http://www.latimes.com/world/asia/la-fg-japan-entrepreneurs-20150329-story.html
(October 6, 2016).
Matsuura, Katsumi, and Yasuko Takezawa. 2001. “Ginko No Chushokigyo-Muke Kashidashi
Kyokyu to Tanpo, Shin’yo Hosho, Furyo Saiken [The Supply of Small Business Bank
Loans, Collateral, Guarantees and Non-Performing Assets].” Yusei kenkyujo disukasshon pepa:
1–22.
McGowan, Müge Adalet, Dan Andrews, and Valentine Millot. 2017. “The Walking Dead?
Zombie Firms and Productivity Performance in OECD Countries.” OECD Economics
Department Working Papers 1372: 1–45.
McGregor, James. 2012. No Ancient Wisdom, No Followers: The Challenges of Chinese Authoritarian
Capitalism. 1st ed. Westport: Prospecta Press.
Miller, Matthew. 2016. “China to Clean-Up ‘Zombie’ Companies by 2020: Xinhua.” Reuters.
http://www.reuters.com/article/us-china-companies-idUSKCN0UV0GH (May 11,
2016).
Moe, Terry M. 1995. “The Politics of Structural Choice: Toward a Theory of Public
Bureaucracy.” In Organization Theory: From Chester Barnard to the Present and Beyond, ed. Oliver
E. Williamson. New York: Oxford University Press, 116–53.
Montinola, Gabriella, Yingyi Qian, and Barry R. Weingast. 1995. “Federalism, Chinese Style:
The Political Basis for Economic Success in China.” World Politics 48(1): 50–81.
Mu, Yibin. 2002. “Impediments to SME Access to Finance and Credit Guarantee Schemes in
China.” Unpublished Manuscript.
Muramatsu, Michio. 1997. “Post-War Politics in Japan: Bureaucracy versus the Party/Parties in
Power.” In State and Administration in Japan and Germany: A Comparative Perspective on Continuity
and Change, eds. Michio Muramatsu and Frieder Naschold. New York: W. de Gruyter,
13–38.
177
Muramatsu, Michio, and Ethan Scheiner. 2009. “The Slow Government Response to Japan’s
Bank Crisis: A New Interpretation.” In Political Change in Japan: Electoral Behavior, Party
Realignment, and the Koizumi Reforms, eds. Steven R. Reed, Kenneth M. McElwain, and Kay
Shimizu. Stanford: Walter H. Shorenstein Asia-Pacific Research Center, 181–98.
Nakamura, Jun-ichi. 2017. Japanese Firms During the Lost Two Decades: The Recovery of Zombie Firms
and the Entrenchment of Reputable Firms. Tokyo: Springer.
Nakamura, Jun-ichi, and Shin-ichi Fukuda. 2013. “What Happened to ‘Zombie’ Firms in Japan?
Reexamination for the Lost Two Decades.” Global Journal of Economics 2(2): 1–18.
Nakata, Hiroko. 2009. “Corporate Bankruptcy, the Japanese Way.” The Japan Times Online.
https://www.japantimes.co.jp/news/2009/04/28/reference/corporate-bankruptcy-the-
japanese-way/ (October 9, 2017).
Nathan, Andrew J. 2003. “Authoritarian Resilience.” Journal of Democracy 14(1): 6–17.
National Federation of Credit Guarantee Corporations. 2011. Credit Guarantee System in Japan
2011. Tokyo: National Federation of Credit Guarantee Corporations.
Nie, Huihua, Ting Jiang, Yuxiao Zhang, and Mingyue Fang. 2016a. “Woguo Jiangshi Qiye De
Xianzhuang, Yuanyin Yu Duice [The Present Situation, Causes and Countermeasures
for Zombie Firms in China].” Hongguan Jingji Guanli 9: 63–68.
———. 2016b. Zhongguo Jiangshi Qiye Yanjiu Baogao: Xianzhuang, Yuanyin He Duice [China’s Zombie
Firms Research Report: Current Situation, Causes and Countermeasures]. Beijing: Renmin
University National Academy of Development and Strategy Anti-Corruption and New
Political Economy Research Center.
Nitani, Miwako, and Allan Riding. 2005. “Promoting Enterprise Development or Subsidizing
Tradition?: The Japan Credit Supplementation System.” International Small Business Journal
23(1): 48–71.
North, Douglass C. 1990. Institutions, Institutional Change, and Economic Performance. New York:
Cambridge University Press.
OECD. 2000. OECD Economic Surveys: Japan 2000. Paris: OECD Publishing.
———. 2005. OECD Economic Surveys: Japan 2005. Paris: OECD Publishing.
———. 2015. Economic Surveys: Japan: Overview. OECD. www.oecd.org/eco/surveys/Japan-2015-
overview.pdf (August 3, 2016).
Onaran, Yalman. 2012. Zombie Banks: How Broken Banks and Debtor Nations Are Crippling the Global
Economy. Hoboken: John Wiley & Sons, Inc.
178
Ono, Arito, Iichiro Uesugi, and Yukihiro Yasuda. 2013. “Are Lending Relationships Beneficial
or Harmful for Public Credit Guarantees? Evidence from Japan’s Emergency Credit
Guarantee Program.” Journal of Financial Stability 9(2): 151–67.
Orlik, Tom. 2011. “China’s Lessons From Mexico and Japan.” Wall Street Journal.
http://www.wsj.com/articles/SB10001424053111904353504576566521762785388
(June 5, 2016).
Osnos, Evan. 2009. “The Forbidden Zone: Letter from China.” The New Yorker 85(21): 54–61.
Otani, Atsushi. 2016. “China Ramps Up Efforts to Remove ‘Zombies’ from Stock Market.”
Nikkei Asian Review. http://asia.nikkei.com/Markets/Equities/China-ramps-up-efforts-to-
remove-zombies-from-stock-market (August 10, 2016).
Packer, Frank, and Mark Ryser. 1992. The Governance of Failure: An Anatomy of Corporate Bankruptcy in
Japan. Columbia University Center on Japanese Economy and Business.
Park, Gene. 2011. Spending Without Taxation: FILP and the Politics of Public Finance in Japan. Stanford:
Stanford University Press.
Peek, Joe. 2009. “The Contribution of Bank Lending to the Long-Term Stagnation in Japan.”
Columbia Business School Center on Japanese Economy and Business Working Paper Series 273: 1–43.
Peek, Joe, and Eric S. Rosengren. 2005. “Unnatural Selection: Perverse Incentives and the
Misallocation of Credit in Japan.” The American Economic Review 95(4): 1144–66.
Pei, Minxin. 2016. China’s Crony Capitalism: The Dynamics of Regime Decay. Cambridge: Harvard
University Press.
Pempel, T. J., and Keiichi Tsunekawa, eds. 2015. Two Crises, Different Outcomes: East Asia and
Global Finance. Ithaca: Cornell University Press.
Pepinsky, Thomas B. 2014. “The Politics of Capital Flight in the Global Economic Crisis.”
Economics & Politics 26(3): 431–56.
Pesek, William. 2014. Japanization: What the World Can Learn from Japan’s Lost Decades. Singapore:
John Wiley & Sons.
Pierson, Paul. 1996. “The New Politics of the Welfare State.” World Politics 48(2): 143–79.
Povel, Paul. 1999. “Optimal ‘Soft’ or ‘Tough’ Bankruptcy Procedures.” Journal of Law, Economics,
& Organization 15(3): 659–84.
Prestowitz, Clyde V. 2015. Japan Restored: How Japan Can Reinvent Itself and Why This Is Important for
America and the World. Tokyo: Tuttle Publishing.
Przeworski, Adam, Michael E. Alvarez, Jose Antonio Cheibub, and Fernando Limongi. 2000.
Democracy and Development: Political Institutions and Material Well-Being in the World, 1950-1990.
New York: Cambridge University Press.
179
Ramseyer, J. Mark, and Frances McCall Rosenbluth. 1993. Japan’s Political Marketplace.
Cambridge: Harvard University Press.
Reuters Staff. 2009. “China Urged to Revamp Rating, Credit Guarantee Systems.” Reuters.
https://www.reuters.com/article/creditratings-china/china-urged-to-revamp-rating-
credit-guarantee-systems-idUSHKG26352520090302 (November 4, 2017).
———. 2016. “UPDATE 1: China’s Non-Performing Loans Hit 11-Year High—Regulator.”
Reuters. https://www.reuters.com/article/china-economy-loans/update-1-chinas-non-
performing-loans-hit-11-year-high-regulator-idUSL3N18935N (November 5, 2017).
Riding, Allan, Judith Madill, and George Haines. 2007. “Incrementality of SME Loan
Guarantees.” Small Business Economics 29(1/2): 47–61.
Rodrik, Dani. 1996. “Understanding Economic Policy Reform.” Journal of Economic Literature
34(1): 9–41.
Rodrik, Dani, Arvind Subramanian, and Francesco Trebbi. 2004. “Institutions Rule: The
Primacy of Institutions Over Geography and Integration in Economic Development.”
Journal of Economic Growth 9(2): 131–65.
Rosenbluth, Frances McCall, and Michael F. Thies. 2010. Japan Transformed: Political Change and
Economic Restructuring. Princeton: Princeton University Press.
Saito, Kuniyoshi, and Daisuke Tsuruta. 2014. “Information Asymmetry in SME Credit
Guarantee Schemes: Evidence from Japan.” RIETI Discussion Paper Series 14(42): 1–39.
Sakai, Koji, Iichiro Uesugi, and Tsutomu Watanabe. 2010. “Firm Age and the Evolution of
Borrowing Costs: Evidence from Japanese Small Firms.” Journal of Banking & Finance
34(8): 1970–81.
Sanderson, Henry, and Henry Foy. 2017. “Call to Tackle China’s Soaring Aluminum Output.”
Financial Times. https://www.ft.com/content/2f255636-0b21-11e7-ac5a-903b21361b43
(March 19, 2017).
Sartori, Giovanni. 1970. “Concept Misformation in Comparative Politics.” The American Political
Science Review 64(4): 1033–53.
Saruyama, Sumio, and Peng Yan. 2016. “Zonbi Segumento To Denki Kiki Sangyo [Zombie
Segment and Electronic Goods Production].” In Japanese Economic Association Autumn
Meeting, Waseda University, 1–26.
Saunavaara, Juha. 2009. “Occupation Authorities, the Hatoyama Purge and the Making of
Japan’s Postwar Political Order.” The Asia-Pacific Jounal 7(39): 1–21.
Schaede, Ulrike. 2013. “System Change and Corporate Reorganization in Japan: The Strategic
Logic of Business Groups and Main Banks, Revisited.” In Syncretism: The Politics of Economic
180
Restructuring and System Reform in Japan, eds. Kenji E. Kushida, Kay Shimizu, and Jean C.
Oi. Stanford: The Walter H. Shorenstein Asia-Pacific Research Center, 77–113.
Scheiner, Ethan. 2006. Democracy without Competition in Japan: Opposition Failure in a One-Party
Dominant State. New York: Cambridge University Press.
Schuman, Michael. 2015. “Zombie Factories Stalk the Sputtering Chinese Economy.” The New
York Times. https://www.nytimes.com/2015/08/30/business/international/zombie-
factories-stalk-the-sputtering-chinese-economy.html (June 12, 2017).
Schumpeter, Joseph. 2006. Capitalism, Socialism and Democracy. 6th ed. New York: Routledge.
Schwartz, Frank J. 1998. Advice and Consent: The Politics of Consultation in Japan. New York:
Cambridge University Press.
Scott, James C. 1998. Seeing like a State: How Certain Schemes to Improve the Human Condition Have
Failed. New Haven: Yale University Press.
Sekine, Toshitaka, Keiichiro Kobayashi, and Yumi Saita. 2003. “Forbearance Lending: The
Case of Japanese Firms.” Monetary and Economic Studies 21(2): 69–92.
Shepsle, Kenneth A. 1986. “Institutional Equilibrium and Equilibrium Institutions.” In Political
Science: The Science of Politics, ed. Herbert F. Weisberg. New York: Agathon Press.
Shikano, Yoshiaki. 2008. Nihon No Chusho Kigyo: CRD Deta Ni Miru Keiei To Zaimu No Jitsuzo
[Japan’s SME: The Real Images of Management and Finances as Seen in CRD Data]. Tokyo: Toyo
Keizai Shinposha.
Shimizu, Kay. 2008. “Private Money as Public Funds: The Politics of Economic Downturn.”
Ph.D. Dissertation. Stanford University.
———. 2013. “The Survival of Regional Banks and Small and Medium Enterprises:
Maintaining Low Unemployment under Economic Stress.” In Syncretism: The Politics of
Economic Restructuring and System Reform in Japan, eds. Kenji E. Kushida, Kay Shimizu, and
Jean C. Oi. Stanford: The Walter H. Shorenstein Asia-Pacific Research Center, 147–72.
Small and Medium Enterprise Agency. 2005. “Shin’yo hokan seido no arikata ni kansuru kento
ko iinkai torimatome (an) ni taisuru iken no gaiyo to sore ni taisuru kangaekata [Thoughts
and Outline of Opinions about Subcommittee Compilation (Draft) of Considerations
about the Way of the Credit Enhancement System].”
http://www.chusho.meti.go.jp/koukai/shingikai/kihon/2005/download/050712pubco
mme.pdf.
———. 2008. 2008 White Paper on Small and Medium Enterprises in Japan: Improvement of Productivity
and the Challenge of Community Revitalization. Tokyo: Ministry of Economy, Trade and
Industry.
181
———. 2016a. 2016 White Paper on Small Enterprises in Japan: Ongoing Business and New Challenges.
Tokyo: Ministry of Economy, Trade and Industry.
———. 2016b. Chushokigyo Kin’yu Shin’yo Hokan Seido No Kokusai Hikaku Ni Tsuite [International
Comparison of SME Finance and Credit Enhancement System]. Tokyo: Ministry of Economy,
Trade and Industry.
SME Policy Council Finance Working Group. 2015. Chushokigyo Shokibo Jigyo-Sha No Hatten Ni
Shisuru Jizoku Kanona Shin’yo Hokan Seido No Kakuritsu Ni Mukete Chukan-Tekina Seiri
[Intermediate Summary Concerning the Establishment of a Sustainable Credit Enhancement System That
Contributes to the Development of SMEs and Microenterprises]. Tokyo: Small and Medium
Enterprise Agency.
http://www.chusho.meti.go.jp/koukai/shingikai/kihonmondai/2015/download/151216
kihonmondai.pdf (August 2, 2016).
———. 2016. Chushokigyo Shokibo Jigyo-Sha No Jigyo No Hatten o Sasaeru Jizoku Kano Na Shin’yo Hokan
Seido No Kakuritsu Ni Mukete (Gaiyo) (An) [Toward Establishing a Sustainable Credit Enhancement
System to Support the Development of Small Businesses and Small-Scale SME (Overview) (Proposed)].
Tokyo: Small and Medium Enterprise Agency.
http://www.chusho.meti.go.jp/koukai/shingikai/kihonmondai/2016/download/161220
kihonmondai3.pdf (April 2, 2017).
Smith, Noah. 2016. “Japan Must Let Zombie Companies Die.” Bloomberg View.
http://www.bloombergview.com/articles/2016-01-20/japan-must-let-zombie-
companies-die (May 7, 2016).
Song, Jiyeoun. 2014. Inequality in the Workplace: Labor Market Reform in Japan and Korea. Ithaca:
Cornell University Press.
Song, Jung-a. 2015. “South Korea Takes Aim at Zombie Companies.” Financial Times.
http://www.ft.com/intl/cms/s/0/e0d74c5c-8456-11e5-9dc0-
186bb1146746.html#axzz44BRB1KBH (April 22, 2016).
Stallings, Barbara. 1992. “International Influence on Economic Policy.” In The Politics of Economic
Adjustment: International Constraints, Distributive Conflicts, and the State, eds. Stephan Haggard
and Robert R. Kaufman. Princeton: Princeton University Press, 41–88.
Starr, Paul. 2015. “Built To Last?: Policy Entrenchment and Regret in Medicare, Medicaid, and
the Affordable Care Act.” In Medicare and Medicaid at 50: America’s Entitlement Programs in the
Age of Affordable Care, eds. Alan B. Cohen, David C. Colby, Keith Wailoo, and Julian E.
Zelizer. New York: Oxford University Press, 319–32.
Steele, Stacey. 2006. “Insolvency Law in Japan.” In Insolvency Law in East Asia, ed. Roman
Tomasic. Burlington: Ashgate, 13–62.
———. 2018. “Trends and Developments in Chinese Insolvency Law: The First Decade of the
PRC Enterprise Bankruptcy Law.” American Journal of Comparative Law: 1–51.
182
Steinfeld, Edward S. 1998. Forging Reform in China: The Fate of State-Owned Industry. Cambridge:
Cambridge University Press.
Stothard, Michael. 2013. “Companies: The Rise of the Zombie.” Financial Times.
http://www.ft.com/intl/cms/s/0/7c93d87a-58f1-11e2-99e6-
00144feab49a.html#axzz46YJNJc11 (April 22, 2016).
Stothard, Michael, and Chris Giles. 2012. “Zombie Companies Stalk UK Economy.” Financial
Times. http://www.ft.com/intl/cms/s/0/d1ecf0d0-316f-11e2-b68b-
00144feabdc0.html#axzz46YJNJc11 (April 22, 2016).
Streeck, Wolfgang, and Kathleen Thelen. 2005. “Introduction: Institutional Change in
Advanced Political Economies.” In Beyond Continuity: Institutional Change in Advanced Political
Economies, eds. Wolfgang Streeck and Kathleen Thelen. Oxford: Oxford University Press,
1–39.
Subler, Jason, and Kevin Yao. 2013. “China Vows ‘Decisive’ Role for Markets, Results by
2020.” Reuters. https://www.reuters.com/article/us-china-reform-
idUSBRE9AA0YB20131112 (August 9, 2017).
Summers, Lawrence H. 2012. “U.S. Economic Prospects: Secular Stagnation, Hysteresis, and
the Zero Lower Bound.” Business Economics 49(2): 65–73.
Svolik, Milan W. 2012. The Politics of Authoritarian Rule. Cambridge: Cambridge University Press.
Takezawa, Yasuko, Katsumi Matsuura, and Masahiro Hori. 2004. “Chusho Kigyo Kinyu
Enkatsuka to Tosan, Daii-Bensai No Sougo Kankei [Facilitation of Funds and Small
Business Failures].” ESRI Discussion Paper Series 87: 1–21.
Tan, Yuyan, Yiping Huang, and Wing Thye Woo. 2016. “Zombie Firms and the Crowding-Out
of Private Investment in China.” Asian Economic Papers 15(3): 32–55.
Tanner, Murray Scot. 1999. The Politics of Lawmaking in Post-Mao China: Institutions, Processes, and
Democratic Prospects. Oxford: Oxford University Press.
The Economist. 2013. “Europe’s Other Debt Crisis.” The Economist.
http://www.economist.com/news/leaders/21588366-its-not-just-sovereign-borrowing-
there-are-too-many-zombie-firms-and-overindebted (March 11, 2017).
Tokyo Shoko Research. 2012. “Ginko 116-Gyo 2012-Nen 3 Tsuki-Ki Tandoku Kessan Besu
Chushokigyo-to Kashidashi-Kin Zandaka Chosa: 5-Nen-Buri Ni Zoka Ni Tenjita
Chushokigy-to Kashidashi-Kin [Loan Balance Survey of SME from 116 Banks on a
Monthly Non-Consolidated Basis for the Fiscal Year Ending March 2012: Loans to SME
Increase for the First Time in Five Years].” http://www.tsr-
net.co.jp/news/analysis_before/2012/1220954_2004.html (April 3, 2017).
183
———. 2017. “‘Minji Saisei-Ho’ Tekiyo Kigyo No Tsuiseki Chosa (2000-Nendo - 2015-Nendo)
[Follow-Up Survey of Firms Using the ‘Civil Rehabilitation Law’ (2000-2015)].”
http://www.tsr-net.co.jp/news/analysis/20170113_07.html (November 3, 2017).
Toyoda, A. Maria. 2011. “Reforming Government Financial Institutions.” In The Evolution of
Japan’s Party System: Politics and Policy in an Era of Institutional Change, ed. Leonard J.
Schoppa. Toronto: University of Toronto Press, 150–76.
Truex, Rory. 2014. “The Returns to Office in a ‘Rubber Stamp’ Parliament.” American Political
Science Review 108(2): 235–51.
Tsai, Lily L. 2017. “Bringing in China: Insights for Building Comparative Political Theory.”
Comparative Political Studies 50(3): 295–328.
Tsebelis, George. 1995. “Decision Making in Political Systems: Veto Players in Presidentialism,
Parliamentarism, Multicameralism and Multipartyism.” British Journal of Political Science
25(3): 289–325.
Uesugi, Iichiro. 2008. “Chusho Kigyo Ni Okeru Tota Wa Seijo Ka [Was Selection Among SME
Normal?].” In Kensho Chusho Kigyo Kin’yu: Konkyo Naki Tsusetsu No Jissho Bunseki [Verifying
SME Finance: Empirical Analysis of “Groundless Prevailing View”], eds. Tsutomu Watanabe and
Iichiro Uesugi. Tokyo: Nihon Keizai Shinbun Shuppansha, 17–48.
———. 2017. “Chushokigyo-Muke Shin’yo Hosho Seido No Kaikaku Ni Tsuite [About SME
Credit Guarantee System Reform].” RIETI.
http://www.rieti.go.jp/jp/columns/a01_0478.html (June 15, 2017).
Uesugi, Iichiro, and Koji Sakai. 2005. “The Special Credit Guarantee Program in Japan.”
Unpublished Manuscript: 1–30.
Vogel, Steven K. 1996. Freer Markets, More Rules: Regulatory Reform in Advanced Industrial Countries.
Ithaca: Cornell University Press.
———. 2006. Japan Remodeled: How Government and Industry Are Reforming Japanese Capitalism. Ithaca:
Cornell University Press.
Wang, Zuofa. 2015. “The Political Economy of the Implementation of the Bankruptcy Law of
China.” Asian Journal of Law and Economics 6(2): 249–81.
Wehrfritz, George, and Hideko Takayama. 2002. “King of the Zombies.” Newsweek International:
40.
Wildau, Gabriel. 2016a. “China Moves to De-Politicize Management of State-Owned
Enterprises.” Financial Times. http://www.ft.com/cms/s/0/53dbd9c2-3f49-11e6-8716-
a4a71e8140b0.html?siteedition=intl#axzz4GuNeoJ2u (August 10, 2016).
184
———. 2016b. “China’s State-Owned Zombie Economy.” Financial Times.
http://www.ft.com/intl/cms/s/0/253d7eb0-ca6c-11e5-84df-
70594b99fc47.html?siteedition=intl#axzz46YJNJc11 (April 22, 2016).
Wong, Joseph. 2011. Betting on Biotech: Innovation and the Limits of Asia’s Developmental State. Ithaca:
Cornell University Press.
Xu, Peng. 2008. “Corporate Governance in Financial Distress: The New Role of Bankruptcy.”
In Corporate Governance in Japan: Institutional Change and Organizational Diversity, eds. Masahiko
Aoki, Gregory Jackson, and Hideaki Miyajima. New York: Oxford University Press, 179–
204.
“Yamazaki Tai.” 2017. Wikipedia.
https://ja.wikipedia.org/w/index.php?title=%E5%B1%B1%E5%B4%8E%E6%B3%B0
&oldid=65160732 (October 28, 2017).
Yamori, Nobuyoshi. 2011. “Kigyo To Kin’yukkan Soho Ni Motomerareru Shin’yo Hosho Izon
Kara No Dakkyaku [Break Out of Dependence on Credit Guarantee Sought by
Companies and Financial Institutions].” Kin’yu Zaisei Jijo 62(12): 22–24.
———. 2013. “Japanese Banking Regulations and SME Finance under the Global Financial
Crisis.” Japanese Journal of Monetary and Financial Economics 1(1): 59–90.
———. 2015. “Japanese SMEs and the Credit Guarantee System after the Global Financial
Crisis.” Cogent Economics & Finance 3(1): 1–18.
———. 2017. “Shin’yo Hosho Seido Kaikaku Ni Taisuru Gokai o Tadasu [Correcting
Misunderstandings about the Credit Guarantee System Reform].” Kin’yu Zaisei Jijo: 38–
41.
Yanagawa, Noriyuki, Sumio Hirose, and Fumio Akiyoshi. 2005. Tosanshoriho-Sei No Kino To Kigyo
Kin’yu-Jo No Shomondai Ni Kansuru Saikento: Kigyo Saisei Sokushin No Kanten Kara No Kosatsu
[Bankruptcy Law, Corporate Finance, and Corporate Revival Process in Japan]. Tokyo: University of
Tokyo. Working Paper.
Yeung, Godfrey. 2009. “How Banks in China Make Lending Decisions.” Journal of Contemporary
China 18(59): 285–302.
Ying, Qianwei, Danglun Luo, and Lifan Wu. 2013. “Bank Credit Lines and Overinvestment:
Evidence from China.” International Journal of Business and Finance Research 7(2): 43–52.
Yusuf, Shahid, Kaoru Nabeshima, and Dwight H. Perkins. 2006. Under New Ownership: Privatizing
China’s State-Owned Enterprises. Washington: The World Bank.
Zhang, Guohua. 2009. “Study on the Small and Middle-Sized Enterprises Credit Guarantee
Legal System.” Journal of Politics and Law 2(1): 71–74.
185
Zhang, Ping, and Ying Ye. 2010. “Study on the Effective Operation Models of Credit Guarantee
System for Small and Medium Enterprises in China.” International Journal of Business and
Management 5(9): 99–106.
Zhongguo Xinwen Wang. 2015. “Fagaiwei: Jiakuai Taotai Jiangshi Qiye Youxiao Huajie
Guosheng Channeng [National Development and Reform Commission: Speed Up the
Elimination of Zombie Firms to Effectively Resolve Excess Capacity].” Zhongguo Xinwen
Wang. http://www.chinanews.com/cj/2015/10-12/7564476.shtml (January 28, 2017).
Zhongzheng Wang. 2016. “Zhuanjia Cheng Jiangshi Qiye Shi Zaocheng Channeng Guosheng
Zhongyao Yuanyin [Experts Say Zombies Are an Important Cause of Overcapacity].”
Xinhua Wang. http://news.xinhuanet.com/fortune/2016-07/28/c_129184520.htm (July
31, 2017).
Zhu, Dailin, and Xuan Hua. 2016. “Pochan Ting: Zai Sifa Cengmian Da Diao ‘Jiangshi Qiye’
[Bankruptcy Courts: Destroying ‘Zombie Firms’ at the Judicial Level].” Nanfang Zhoumo.
http://www.infzm.com/content/119114 (July 31, 2017).
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Supply and demand of economic hierarchy: the Northeast Asia case
PDF
And the dead shall walk the earth: Zombies and the politics of death
PDF
Anti-foreign boycotts as a tool of economic coercion: the case of China
PDF
The strategy of firms towards political elites and consumers in polarized western democracies
PDF
The politics of the growth of regional disparity in Japan
PDF
Essays on political economy and corruption
PDF
Changing political attitudes and behavior in a diverse America: incorporating individual and contextual determinants
PDF
The business of nationalism: how commodification sustains bilateral tensions
PDF
Examining global financial regulatory reform in the G20 and the FSB
PDF
Shaking the foundations of violent civil conflict: institutions, disasters, and the political economies of state-rebel interaction
PDF
Power, profits, and politics: energy security and cooperation in Eurasia
PDF
The political economy of Brazilian trade policy: domestic determinants, world and regional strategies
PDF
Economic policy-making in a changing environment: the politics of success in Peru vis-à-vis China
PDF
The partial Good Samaritan states: China and Japan in the international relations of autocracy and democracy
PDF
The political economy of implementation: intellectual property rights protection across the world
PDF
Developing countries in the digital era: state and business interactions for industrial upgrading
PDF
Social policy and presidential ideology in Latin America: the political economy of social spending and anti-poverty programs
PDF
Financial crises and trade policy in developing countries
PDF
Social construction of the experience economy: the spatial ecology of outdoor advertising in Los Angeles
PDF
Reconstituting the fiscal bargain: the politics of taxation in Latin America's emerging markets
Asset Metadata
Creator
Wilbur, Scott
(author)
Core Title
Creating destruction: the political economy of zombie firms
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Political Science and International Relations
Publication Date
02/13/2018
Defense Date
11/28/2017
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
comparative political economy,finance policy,Japan,OAI-PMH Harvest,SME,zombie firms
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Katada, Saori N. (
committee chair
), Sandholtz, Wayne (
committee chair
), Hondagneu-Sotelo, Pierrette (
committee member
), Kang, David (
committee member
), Munck, Gerardo (
committee member
)
Creator Email
sawilbur@usc.edu,scott.wilbur@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-472513
Unique identifier
UC11268211
Identifier
etd-WilburScot-6029.pdf (filename),usctheses-c40-472513 (legacy record id)
Legacy Identifier
etd-WilburScot-6029.pdf
Dmrecord
472513
Document Type
Dissertation
Rights
Wilbur, Scott
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
comparative political economy
finance policy
SME
zombie firms