University of Southern California Dissertations and Theses

Crowded with potential: housing and social mobility strategies among China's educated migrants

(USC Thesis Other)

Crowded with potential: housing and social mobility strategies among China's educated migrants

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
CROWDED WITH POTENTIAL:
HOUSING AND SOCIAL MOBILITY STRATEGIES AMONG CHINA’S
EDUCATED MIGRANTS
by
Julia Gabriele Harten

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
PUBLIC POLICY AND MANAGEMENT

August 2020

Copyright 2020 Julia Gabriele Harten

ii
ACKNOWLEDGEMENTS
There are many who helped me along the way. I want to take a moment to thank them.
First, I would like to thank my committee chair Annette Kim. The doctoral program was a
journey of challenges, disappointments, moments of joy – and above all, personal and professional
growth. Annette Kim went above and beyond to help me reach my goals and has shown me, by
her example, what a truly great scholar is. I cannot begin to express my thanks to her for the
countless hours spent advising and encouraging me. Without her help and guidance, I would not
be where I am today, and I will be forever grateful for the chance to have worked with her.
I want to express my deepest appreciation to my other committee members Lisa Schweitzer
and Chris Webster. Lisa Schweitzer's wit and wisdom are an inspiration, and her belief in me and
my work has been an invaluable source of assurance throughout the rockier parts of this journey.
I am grateful to Chris Webster for his support and the instrumental feedback provided along the
way.
I must also express my thanks to Eric J. Heikkila, who supported and believed in me from
the beginning. Our collaboration led to my very first co-authored journal paper, and working with
him has been a great pleasure and a privileged. I am also grateful to have had the chance to work
with Jorge De la Roca, who is as brilliant as he is kind and generous. I wish to thank Marlon
Boarnet and Richard Green, whose advice and care have meant a lot. Thank you also to Pamela
Clouser McCann, Joshua Goldstein, and Grace Ryu for all that you have taught me.
I would like to extend my gratitude to Chen Jie at Shanghai Jiaotong University, Li Limei
at East China Normal University, and Han Zheng at Tongji University for their advice, guidance,
and support. Many thanks also to my mentors and letter writers who I met and worked with prior
iii
to USC. I could not have made it to the doctoral program without the support of Salmai Qari at the
Berlin School of Economics, Nicola Fuchs-Schündeln at the Goethe University Frankfurt, and
Reint Gropp at the Otto-von-Guericke University Magdeburg. Finally, I gratefully acknowledge
the financial support by the Lincoln Institute for Land Policy, the Lusk Center for Real Estate, the
U.S. Department of Education, the USC Price School, and the USC Graduate School.
This work would not have been possible without the generosity and trust of my
interviewees. Although I am not mentioning their names here to protect their identity, it is thanks
to their willingness to share their stories that I could learn and write about my dissertation topic in
such detail. I am incredibly grateful to have met each and every one of them.
I also want to thank the outstanding research assistants who have helped me starting from
the first round of on the ground data collection in summer 2016. The diligence, alertness, ingenuity,
and attention to detail of Li Xiangyu, Liang Yuqi, Liu Yuan, Wang Le, and Zhang Ruiqi were
critical to the quality of my data. I feel lucky to now call them my friends. Special thanks also to
Sharon and Judy Chen, Andrew, Edward, Lee Feng, and many others with whom I share friendship
and fellowship and whose generous introductions to their friends and acquaintances were crucial
to this project. I am extremely indebted to my friends Angela Meng Yin, Micky Lu, and Benny
Weilun Zhang, who were always willing to lend a helping hand and who answered countless
questions about Chinese language and culture. Finally, thank you to my dear friend Jan Thorwirth
who always provided a home away from home in Hong Kong.
I would also like to thank the many wonderful people (and future colleagues!) I have met
during my time at USC Price. I cannot begin to express how grateful I am to have shared this
journey with Huê-Tâm Jamme. These past five years, we went through so much, and you were
iv
there for all of it. You have been my sounding board, writing companion, late-night Gateway-
buddy, and in addition to having a brilliant mind, you are a close friend. Thank you also to the
incredibly smart and talented Maria Francesca Piazzoni. I have learned so much from you and feel
so thankful to call you my friend. Thank you, Gene Burinskiy, Heejin Cho, Soyoon Choo, Andrew
Eisenlohr, Sahil Gandhi, Rogier Holtermans, Nathan Hutson, Zeewan Lee, Colin Leslie, Anthony
Orlando, Gregory Randolph, Seva Rodnyansky, Madi Swayne, Bo Wen, Lee White, Linna Zhu,
Yolanda Yingying Zhu, for the shared moments of joy, dread, worry, learning, and laughter – the
countless hours spent at Gateway are cherished memories thanks to you.
Friendship and family are what sustained me during this time. Thank you to Victoria Rocha,
my long-term roommate and dear friend. Your grit, integrity, and positivity inspire me, and your
humor, friendship, and the occasional adventure have kept me sane during the last five years. Many
thanks also to my dear friend Johanna Stephan whose passion for Chinese language and culture
helped me stay committed and believe that I, too, may someday converse in Mandarin as elegantly
as she does. I am incredibly grateful for my friends of many years, Birte Rendla, Isabella Tröster,
Vanessa Walz, and Julia Weiss, whose friendship has helped me through the hard times: distance
means nothing when someone means so much.
Last but definitely not least, I would like to thank my parents Birgit Susanne and Paul
Alexander Harten, as well as my siblings Isabella, Felix, and Valentina, and my grandparents
Wilma and Helmut Weber. None of this would have been possible without your love and nurturing.
Words cannot express how deeply grateful I am for everything you have given me. Finally, to my
husband and partner, Christos Thrampoulidis, thank you for your understanding, for your trust,
your patience, and your love. Thank you for always being by my side.
v
TABLE OF CONTENTS
Acknowledgements ......................................................................................................................... ii
List of Tables ................................................................................................................................ vii
List of Figures .............................................................................................................................. viii
Abstract .......................................................................................................................................... ix
Chapter 1. Introduction: The Other Chinese Migrants ................................................................... 1
1.1. Literature ............................................................................................................................. 4
1.2. Research Questions and Methodology ................................................................................. 7
1.3. Overview of the Structure .................................................................................................... 9
Chapter 2. A Hidden Informal Housing Market ........................................................................... 11
2.1. Introduction ........................................................................................................................ 11
2.2. Mixed Method Data Collection .......................................................................................... 13
2.2.1 Online Advertisement Data .......................................................................................... 13
2.2.2 “Real” Market Data ...................................................................................................... 15
2.3. Findings ............................................................................................................................. 20
2.3.1 No Housing for Young and Educated Migrant ............................................................ 20
2.3.2 Trading Space for Access ............................................................................................ 24
Chapter 3. Deconstructing the Social Meaning of Group Renting ............................................... 33
3.1. Introduction ........................................................................................................................ 33
3.2. Accessing the Field ............................................................................................................ 36
3.2.1 Fieldwork as an Outsider: What Did I Get to See? ...................................................... 36
3.2.2 My Bed Space Renting Experience ............................................................................. 41
3.3. Overcrowded Living ........................................................................................................... 43
3.3.1 Crowded Spaces ........................................................................................................... 44
3.3.2 Everyday Strategies: Navigating Crammed Conditions .............................................. 47
3.3.3 Interpersonal Relations and Overcrowding: Negotiating Sharing in Tight Quarters .. 52
3.4. Group Renting as Time-Space Strategy ............................................................................. 55
3.4.1 Making Saving Possible ............................................................................................... 55
3.4.2 Gaining a Foothold ...................................................................................................... 59
3.4.3 The Odds Stacked Against Them: Hukou, Education Stratification, and Class .......... 64
3.4.4 A Seemingly Successful City-Starter Strategy ............................................................ 71
vi
3.5. Mobility Strategies Are Multi-Generational ...................................................................... 73
3.5.1 The Weight of Filial Piety in an Aging Society ........................................................... 73
3.5.2 Class Matters ................................................................................................................ 80
3.6. The Promise of Upward Mobility ...................................................................................... 82
3.6.1 Expectations and Reality: Cognitive Dissonance Among the Elite ............................. 82
3.6.2 Internalizing and Reproducing “Sacrificing” as a Social Institution ........................... 87
3.6.3 Questioning the Viability of the Social Contract ......................................................... 91
3.7. Class Consciousness? ........................................................................................................ 94
3.7.1 Shanghai Divided ......................................................................................................... 94
3.7.2 Facing the Elephant in the Room: Class Relations in Contemporary China ............... 98
Chapter 4. Is Shanghai a Special Case? ...................................................................................... 101
4.1. Introduction ...................................................................................................................... 101
4.2. Research Design and Data Collection ............................................................................. 104
4.2.1 Survey Design ............................................................................................................ 104
4.2.2 Survey Distribution .................................................................................................... 107
4.3. Findings ........................................................................................................................... 109
4.3.1 Likely Oversampling of High-Skilled Workers ......................................................... 109
4.3.2. Tier 1 Cities Are Different ........................................................................................ 112
4.3.3 Group Renters Are Mostly Non-Elite Graduates ....................................................... 117
Chapter 5. Summary and Reflections ......................................................................................... 119
5.1. Summary .......................................................................................................................... 119
5.2. Reflections and Open Questions ...................................................................................... 122
5.2.1 Hidden Informality and Planning ............................................................................... 122
5.2.2 Informality and Inequality ......................................................................................... 126
References ................................................................................................................................... 129
Appendix ..................................................................................................................................... 154
Appendix A: The Veracity of Online Advertisement Data for Housing and Informality
Research .................................................................................................................................. 154
Appendix B: Online Social Media (WeChat) Survey Instrument ............................................ 157
Appendix C: Supplemental Material Chapter 4 ...................................................................... 163
Difference in Mean Tests: Shanghai vs. Other Tier 1 Cities (First Job Location) ............. 163
Difference in Mean Tests: Tier 1 Cities vs. Non-Tier 1 Cities (First Job Location) .......... 167

vii
LIST OF TABLES
Table 1: Timeline Mixed Method Data Collection ........................................................................................ 8
Table 2: Desired Tenant Traits, Web-Scraped Advertisement Data (n=3147) ............................................ 21
Table 3: Descriptive Statistics for Field Survey and Scraped Data
+
............................................................ 25
Table 4: Rents by Level of Crowding – Comparing Market Survey versus Web Scraped Data ................. 27
Table 5: Hedonic Regression Results
ß
, Field Survey Data .......................................................................... 29
Table 6: Average Daily Online Advertisements for Low Rent Shared Housing in China’s Most
Populous Cities .................................................................................................................................. 103
Table 7: Descriptive Statistics for All Observations (N=15,345) .............................................................. 109
Table 8: Descriptive Statistics by City Type ............................................................................................. 113
Table 9: Cross-Tabulation of First Job Location and Local Hukou .......................................................... 114
Table 10: Cross-Tabulation of First Job Location and Hukou ................................................................... 115
Table 11: Number of Group Renters by First Job Location, Education, and Hukou ................................. 117
Table 12: Comparative Descriptive Statistics: Mean Values of Online Data versus Field Data ............... 154
Table 13: Hedonic Regression Results,
ß
Web Scraped Online Advertisement Data ................................ 155
Table 14: Difference in Mean Test, Share of Tier 1 City Hukou Holders, Shanghai vs. Other Their 1
Cities .................................................................................................................................................. 163
Table 15: Difference in Mean Test, Share of Elite Graduates, Shanghai vs. Other Their 1 Cities ........... 164
Table 16: Difference in Mean Test, Share of Those Earning Less Than 5,000 RMB/ Month, Shanghai
vs. Other Their 1 Cities ..................................................................................................................... 164
Table 17: Difference in Mean Test, Share of Those Spending Less Than 1,000 RMB/ Month on Rent,
Shanghai vs. Other Their 1 Cities ..................................................................................................... 165
Table 18: Difference in Mean Test, Share of Group Renters, Shanghai vs. other Their 1 Cities .............. 165
Table 19: Difference in Mean Test, Share of Room Shares, Shanghai vs. other Their 1 Cities ................ 166
Table 20: Difference in Mean Test, Share of Those with Family-Provided Housing, Shanghai vs.
Other Their 1 Cities ........................................................................................................................... 166
Table 21: Difference in Mean Test, Share of Those with a Bachelor’s Degree or above, Tier 1 vs.
Non-Tier 1 Cities ............................................................................................................................... 167
Table 22: Difference in Mean Test, Share of Elite Graduates, Tier 1 vs. Non-Tier 1 Cities ..................... 167
Table 23: Difference in Mean Test, Share of Those Earning Less Than 5,000 RMB/ Month, Tier 1 vs.
Non-Tier 1 Cities ............................................................................................................................... 168
Table 24: Difference in Mean Test, Share of Those Earning Less Than 3,000 RMB/ Month, Tier 1 vs.
Non-Tier 1 Cities ............................................................................................................................... 168
Table 25: Difference in Mean Test, Share of Spending Less Than 1,000 RMB/ Month on Rent, Tier 1
vs. Non-Tier 1 Cities ......................................................................................................................... 169
Table 26: Difference in Mean Test, Share of Group Renters, Tier 1 vs. Non-Tier 1 Cities ...................... 169
Table 27: Difference in Mean Test, Share of Room Sharers, Tier 1 vs. Non-Tier 1 Cities ....................... 170
Table 28: Difference in Mean Test, Share of Those with Family-Provided Housing, Tier 1 vs. Non-
Tier 1 Cities ....................................................................................................................................... 170
Table 29: Difference in Mean Test, Share of Those with First Job Trial Periods, Tier 1 vs. Non-Tier 1
Cities .................................................................................................................................................. 171

viii
LIST OF FIGURES
Figure 1: Mapping Web Scraped and Field Survey Data ........................................................................... 18
Figure 2: Left: 50 People Group Renting in a Four-bedroom Apartment, Right: Pictures Taken during
Site Visits ........................................................................................................................................... 28
Figure 3: Pictures of the Room where I Rented a Bed, September 2018 ................................................... 46
Figure 4: The Shanghai Rainbow Chamber Singers Perform “My Body Feels so Drained” ..................... 92
Figure 5: Meme Satirizing the Housing Struggles of Young Adults in Urban China ................................ 93
Figure 6: Hukou and Class - Structural Disadvantage for Educated Migrants ......................................... 122
Figure 7: Selected Promotional Material to Distribute the Survey Offline .............................................. 162

ix
ABSTRACT
In China’s cities, some educated migrants rent individual beds in overcrowded shared apartments
in the city center. This dissertation studies the hidden, informal housing phenomenon known as
“group rentals” in Shanghai, China and asks what its emergence tells us about changing dynamics
of social mobility and contemporary inequality. Based on remotely collected online advertisement
data, three summers of fieldwork, a cell phone social media survey, and several years living,
studying, and working in China, I employ a mix of methods to generate an in-depth investigation
into group renting, not just as a housing, but as a social phenomenon. I find that group renters are
mostly recent migrants with degrees from second-tier colleges who trade-off space for access to
jobs and opportunities. Beyond providing affordable access, group rentals function as a time-space
strategy for Shanghai newcomers who need to keep expenses low; either due to sending
remittances or in order to gain a foothold in the city. Group rentals as incubators for city starters
are part of a longer-term, multi-generational social mobility strategy. Hukou, stratified education,
and socio-economic background structure migration, employment, and housing trajectories for
recent college graduates. Even in the face of structural disadvantage, powerful social constructs
uphold this generation’s belief in social mobility and, for the moment, thwart class-based
resentment. This dissertation’s findings highlight the centrality of cities for social mobility,
increasingly complicated by new dynamics of inequality unfolding at the education-migration
nexus. It urges planning to consider the spatial implications of skill-based sorting and provokes
further inquiry into the future of urban access.

Keywords: Crowding, Migrant Housing, Rental Housing, Urban China, Social Mobility,
Educated Migrants, International Planning
1
CHAPTER 1. INTRODUCTION: THE OTHER CHINESE MIGRANTS
When I first meet Meifeng
1
in the summer of 2018, she is a young woman in her early 20s. After
she graduated from a private college in her hometown of Xiamen, Fujian Province, she decided to
move to Shanghai for her first job. She worked a regular office job, walked or took the bus to go
to work, liked to watch movies in her free time, and made friends in the new city she was excited
to be in. Her life was no different from that of millions of young people in Shanghai – except that
when she got home after work, “home” was a two bedroom, one bathroom apartment shared with
21 other women.
During my dissertation research I met many young men and women who, like Meifeng,
were renting just a bed in Shanghai. In what is called “bed space rentals” ( chuangwei
chuzu) or “group rental housing” ( qunzu fang), tenants rent individual beds in rooms and
apartments full of strangers. Landlords illegally convert individual commercial and residential
units in high-rises into extremely overcrowded dormitories.
2
The crammed living arrangements
violate regulations. But because the overcrowding happens behind closed doors, it can stay mostly
hidden.
3

The emergence of this housing phenomenon in urban China is curious in several regards.
For one, China has long been regarded as having achieved historic urbanization without the usual,
accompanying dark side (UN-Habitat, 2003). During the West’s industrial revolution, and more

1
All names are pseudonyms to protect the identity of my informants.
2
Overcrowding and group renting has also been documented in urban villages (Wu, 2016).
3
In 2011, Shanghai and Beijing released regulations mandating a minimum living space requirement of five square
meters per person and a maximum room capacity of two people per room. In recent years, government crack-downs
on group rental housing have intensified (Beijing Municipal Commission of Housing and Urban-Rural Development,
2011; Shanghai Municipal Government, 2011; Wu, 2019).
2
recently in Global South cities, when millions of people migrate to cities, the under-capitalized
newcomers often have no choice but to find shelter in “slums.” While China’s planned cities are
ostensibly devoid of large, open squatter settlements, group rental housing is just one of several
informal housing markets, which have recently been documented in surprising locations. Migrants
have been found to eke out living space underground, on rooftops, in interstitial spaces – and now
behind closed doors (Huang & Yi, 2015; Kim, 2016; Tanasescu, Wing-Tak & Smart, 2010).
China’s hidden informality certainly reflects the country’s political economy and land use control,
but it is arguably an example of a broader new trend in global patterns of informality. As property
regimes increase enforcement in crowded urban centers around the world, informality has become
more hidden and dispersed within the formal built environment (Kim, 2019).
Another notable feature is that group rentals can mostly be found in central city apartment
buildings. I first learned about group rentals because beds for rent were advertised online. Scrolling
through rental listings on popular websites for classified ads showed almost exclusively addresses
in inner city Shanghai. By contrast, most of the literature on migrant housing locates them on the
urban periphery, in urban villages (Song, Zenou & Ding, 2008; Wu, Zhang & Webster, 2013) and
in factory or on-site construction housing (Pun & Smith, 2007; Swider, 2016). The locations of
group rentals suggest that they are catering to a different, newly emerging housing need; Group
renters appear to require a different kind of urban access.
Third, looking through group rental advertisements, landlords seemed to specifically target
college-educated Shanghai newcomers. Migrants with higher education degrees is a relatively new
phenomenon in China. Up until a series of higher education reforms and in particular a push for
mass education in the later 1990s, college education had been the privilege of few and mostly
urban youth (Mok & Qian, 2018). Since the efforts to increase access to higher education, there
3
has been a steady growth in college graduates, a growing share among them from rural
backgrounds (Bai, 2006; Hu & Hibel, 2014; Zhong, 2011). But education was supposed to produce
high-skilled workers whose increased productivity would earn them high wages. The central
location would suggest that group renting affords them access to the city’s expensive downtown
central business district. Group renters are thus likely to be white collar workers, more than factory
workers. If group rentals accommodate predominantly educated migrants, what does this say about
the labor market outcomes for this demographic?
Finally, the online advertisements bring together strangers to share very tight quarters and
limited facilities. Residential crowding and shared living has been a common affordable housing
strategy in many cities around the world and throughout history (Gibbon & Bell, 1939; Lubove,
1963; HAD, 2017). Previously considered the domain of student housing (Hochstenbach &
Boterman, 2017; Maalson, 2020; Rhodes, 2002), or a kin-based safety net in times of economic
hardship (Mykyta & Macartney, 2011; Seltzer, Lau & Bianchi, 2012; Wiemers, 2014), house and
room sharing has recently re-emerged as a mainstream housing strategy among urban young
professionals in cities around the world (Clapham et al., 2014; Clark et al., 2017; Maalson, 2020;
Mykyta, 2012; Heath & Kenyon, 2001). Group rentals may thus be part of a global shift in how
urban space is consumed, where, and by whom.
In sum, the emergence of group rental housing appears to indicate that something has
changed in China’s urban development. Informality is often just the visible symptom of deeper-
rooted social inequalities (Perlman, 1979; Roy, 2004; 2009). Hence, this dissertation asks whether
bed space renting as a new informal housing phenomenon, at a more fundamental level, indicates
that the mechanisms of inequality are changing at the intersection of education and migration in
contemporary urban China.
4
1.1. Literature
China is marked by large regional inequalities. Although over the last five decades, China has seen
unprecedented economic growth, lifting millions out of poverty, inequality has actually increased
sharply since the reforms in the late 1980s and the ensuing transition to a market-based economy
(Naughton, 2006; World Bank & State Council, 2014; Zhu, 2012). Inequality, however, is often
socially acceptable – if there is equal access to opportunity and the real possibility of social
mobility (Chetty et al., 2014; Whyte, 2010).
Broadly speaking, the literature discusses two main strategies for social mobility: education
and migration. The positive association between education and earnings, and with that the potential
for moving up the social ladder, is a central theme in labor economics (e.g., Katz & Autor, 1999;
Goldin & Katz 2007). At a macro level, human capital development also is a key economic
development and growth strategy (Lucas, 1989). Despite growing evidence of structural barriers
– along the lines of factors such as race, gender, and socio-economic background – many societies
continue to hold on to and invest in education as a vehicle for social mobility (Boudon, 1974;
Chetty et al., 2014; Cole & Omari, 2003; Ellis & Lane, 1963; Haveman & Smeeding, 2006;
Higginbotham & Weber, 1992). The neighborhood effects literature adds the role of location to
the discussion about access to and success in education and in turn social mobility (Ainsworth,
2002; Garner & Raudenbush, 1991; Luswig et al., 2013).
Of course, location is central to the second strategy, the notion that links (labor) migration
and social mobility. In their influential model, Harris and Todaro (1970) explain migration flows
as decisions based on expected income differentials. Physical mobility means the optimization
across space, relocation for better economic opportunity (De Haas, 2010; Moretti, 2011).
5
Mostly, the same themes and issues are also covered in the literature on post-reform China.
There is a growing body of literature studying the nature and change of returns to education (Gao
& Smyth, 2015; Hu & Hibel, 2014; 2015; Mok & Wu, 2016; Zhong, 2011). Expanding access to
education has been an integral part of China’s development strategy (Bai, 2006; Meng, Shen &
Xue, 2013). In China, too, education – via increasing productivity and wages – has been regarded
as promoting economic mobility. In a recent meta-analysis, Churchill and Mishra (2018) find that
the literature generally agrees that in China returns to education are positive and rising, with a
premium for the most educated. The role of location in access to education is also increasingly
documented. Large regional disparities intersect with China’s institutional framework – in
particular the Chinese household registration system hukou ( )
4
– and pose structural barriers
to educational access (Fu & Ren, 2010; Liu, 2005; Montgomery, 2012; Wu, Zhang & Waley, 2016).
China has been transformed by millions of people betting on the second strategy, migrating
for social mobility. Migration and urbanization lie at the heart of the country’s economic successes
and have brought upward mobility for millions of migrants and their families (Bian & Logan, 1996;
Nee, 1996). Despite ample evidence of differential treatment of migrants in urban labor markets,
even hukou-based discrimination, regional differences, especially rural-urban disparities, remain
so large that migration continues to have tangible economic consequences (Chen, 2011; Meng &
Zhang, 2001; Xiao & Bian, 2018). In particular, the urban wages of rural farmers have had
spillover effects for economic development in their native places: Migrants' remittances have
funded private housing construction, small business development, and the human capital

4
Hukou, the Chinese household registration system, ties access to citizenship rights – including public housing and
education – to the place of registration, typically the place of birth. Importantly, it distinguishes between rural and
urban status and functions like an internal passport system (Chan, 1996; 2010; 2013; Solinger, 1999).
6
development of migrant children (Cai, 2003; Du, Park & Wang, 2005; Murphy, 2002; Rozelle,
Taylor & DeBrauw, 1999; Sargeson, 2002).
But what about the interaction of migration and education? Generally, empirical evidence
suggests that mobility is rising with education, suggesting a higher ability to optimize across
geographies (Malamud & Wozniak, 2012; Machin, Pelkonen & Savanes, 2012; Molloy, Smith &
Wozniak, 2011).
Additionally, there is a sizable literature in urban and labor economics that looks at the
location decisions and returns to education for educated workers; in other words, the interaction
of location, migration, and education after education has been obtained. The urban wage premium
literature seeks to tease out the effect of density of economic activity on workers’ productivity
(Eeckhout, Pinheiro & Schmidheiny, 2014; Glaeser & Mare, 2001; Gould, 2007). Within this
literature, recent works have found differential effects based on skill levels and education, with the
most educated fetching the largest density-productivity boosts (Bacolod, Blum & Strange, 2009;
De la Roca & Puga, 2017). Similarly, the human capital externalities literature focuses on
analyzing the influence of aggregate levels of human capital on individual wages (Moretti, 2004,
2011; 2012; Rauch, 1993). Within this literature, recent work has documented the rise of skill-
based spatial sorting, with high-skilled workers increasingly concentrating in fewer, larger cities
(Diamond, 2016; Moretti, 2011).
For China, the relationship between education and migration is severely understudied; in
part because being both college-educated and a migrant is a fairly new occurrence. There exists
some literature on how education impacts migration patterns (Lu & Xia, 2016; Luo & Xing, 2016),
but the literature on interregional differentials in the returns to education is sparse (Liu, 2007). To
7
the best of my knowledge, the question of how education and migration intersect in producing
labor market access and outcomes has not been studied. The implicit assumption in literature and
policy seems to be that education overrides migrant status and most disadvantages associated with
hukou. Educated migrants now renting mere beds in overcrowded dormitory-style housing,
however, suggests new dynamics of precarity. Could it be that structural inequalities related to
hukou and class are carrying over into young adulthood and the intersection of education,
urbanization, and urban employment? What does this mean for contemporary inequality and social
mobility?
1.2. Research Questions and Methodology
It is against this background that this dissertation studies group rental housing not just as a housing
sub-market, but as a social phenomenon. The overall objective of this dissertation is to analyze
group renting and the structural forces that produce it with the aim to understand what the
emergence of group renting is telling us about the changing mechanics of social mobility. In
particular, I ask how education intersects with migration in the making of contemporary inequality
in China. I approach the overarching research aim through a series of secondary questions, each a
building block towards the overall research goal:
(1) Whose housing needs is the “group rental” housing market in Shanghai meeting? Do they
indicate educational and/or regional stratification?
(2) What does this housing market offer? What are the tradeoffs that the renters in this housing
market are making?
8
(3) Why do tenants rent in this market? Moving beyond individual rational choices, are there
structural socio-cultural forces producing this group renter demographic and supporting
this new housing market?
(4) How do tenants make sense of their experience and about why they are living in group
rentals? What individual and collective processing is involved to cope with the housing
conditions? Are they developing class consciousness?
(5) Is group renting in Shanghai a regional anomaly or a local instance of a larger phenomenon?
Answering these research questions addresses an important gap in the literature. The migration-
education nexus is understudied in China. More broadly, both education and migration have been
studied as levers of social mobility, but not much is known about the interplay of education and
migration against differences in socio-economic background. This dissertation probes the role of
cities in the striving for social mobility, paying attention to new dynamics of stratified access.
To address these research questions, I use a multi-phase, sequential mixed-method design.
Table 1 illustrates the multi-method data collection.
Table 1: Timeline Mixed Method Data Collection
Time
Web Scraping Online Advertisements February – June 2016
Groundtruthing Field Survey June – July 2016
Fieldwork (Interviews & Participant Observation) July – September 2018
Follow-up Interviews & Mobile Survey June – July 2019

Given the range of research questions, different data and methods are needed to address
the different types of questions. Generally, questions about broad patterns are addressed through
quantitative approaches, questions about social meaning and motivation through a qualitative
9
approach. A detailed discussion of the different methodologies can be found in the respective
chapters.
1.3. Overview of the Structure
Chapter 2 – A Hidden Informal Housing Market – analyzes group rentals as a niche housing market
in Shanghai and addresses research questions one and two.
5
Drawing on web scraped online
advertisements in combination with market survey data collected in the field, I find that this
informal housing market serves young, educated migrants who trade off minimal personal space
and amenities for superior locational access to employment centers. The calculus is reflected in
extremely fine-scale pricing of crowding, down to the individual bed.
Chapter 3 – Deconstructing the Social Meaning of Group Renting – draws on ethnographic
research to delve into the social meaning of group rental housing beyond what can be learned with
the tools of housing market analysis. Becoming a group rental renter myself and developing
personal relationships with renters with whom I could follow up with subsequent updates allowed
me to study a broader range of forces that had led to their being in our group rental. Addressing
research questions three and four, I find that group renting is motivated by the need to save money
– either to send back home to the group renters’ families or because a new dynamic of structural
inequality is producing lower starting wages for most educated migrants. Either way, group renting
is a low-cost stepping stone, a means to buy space and time while they set up better situations for
themselves. In the long term, the group renters’ mobility strategies were motivated by a bid for
upward mobility and financial security. These were high-stakes goals because they would allow

5
Most of the work presented in this chapter is joint work with Annette M. Kim and J. Cressica Brazier. For clarity of
discussion I will use the first person pronoun.
10
the young men and women to support their elderly family members in a rapidly aging society with
a tenuous pension system and public safety net.
Living with them also allowed me to see their non-verbal coping mechanisms as well as
their individual and collective sensemaking in order to understand the common societal narratives
in relation to group renting. I find that group renters and other struggling job market entrants
process their experience by focusing on temporariness. The coping happens in a societal contexts
in which narratives about sacrificing and individual responsibility to achieve social mobility are
constantly reproduced. Having witnessed more class disruption than continuity, my interviewees
did not voice class-based sentiments. They still see their opportunities as plentiful, and even more
so compared to their parents’ generation.
Chapter 4 – Is Shanghai a Special Case? – addresses the fifth and final research question.
I draw on survey data on the educational background, employment and housing trajectories of job
market entrants across all major Chinese cities. This data, collected via a cell phone social media
survey, allows me to place the case of Shanghai in relation to other Chinese cities and with that
get at the question of whether group renting in Shanghai is merely a regional anomaly or in fact a
local instance of a much larger phenomenon. I find that regardless of where job market entrants
took their first job, about one fifth of survey respondents reported living in group rentals.
Furthermore, while Shanghai seems roughly comparable to other large cities, a distinctive pattern
of difference emerges between job market entrants working in one of China’s four biggest cities
and everyone else.
Finally, chapter 5 – Summary and Reflections – summarizes the findings and reflects on
lessons and open questions for planning and policy; for China and beyond.
11
CHAPTER 2. A HIDDEN INFORMAL HOUSING MARKET
2.1. Introduction
This chapter approaches the group renting phenomenon through a housing market analysis lens.
Specifically, I address the first two research questions: Whose housing needs is the “group rental”
housing market in Shanghai meeting? And what are the tradeoffs that the renters in this housing
market are making?
Group renting as a housing sub-market has not always existed in urban China. Certainly,
the internet has facilitated in organizing the house sharing among strangers, but from a housing
research perspective, both questions can be addressed by inquiring about the unmet demand that
is giving rise to the informal arrangements. In particular, residential crowding and shared living
have a long history as affordable housing strategies; crowding achieves affordability in otherwise
unaffordable locations by increasing the intensity of use. But who needs to access locations that
would otherwise be out of reach? Who is willing to sacrifice personal space and privacy towards
this end?
Given that this is group rental housing, I hypothesize that it is migrants who rent bed spaces.
During China’s economic and political transition, privatization of public housing, backlog demand,
rapid urban expansion, economic growth, and speculative investment brought about fast and
dramatic increases in urban homeownership, housing stock, and prices (Logan, Fang & Zhang,
2009; 2010; Wang & Murie, 1996; Yu, 2006). But it was almost exclusively urban residents with
urban hukou that benefitted from these changes and became majority homeowners.
6
Migrants on

6
Additionally, cultural norms link leaving home with marriage for young adults. If young people work locally, they
hence typically remain in the parental home until they establish their own family (Ting & Chiu, 2002).
12
the other hand, have for the most part been excluded from urban homeownership due to low
incomes and institutional barriers. They are therefore majority renters. In Shanghai, for instance,
the share of migrant renters is 78% (Shanghai Municipal Government, 2017; Wu, 2004).
Most of the existing work discusses migrants living on the urban periphery, in urban
villages (Song et al., 2008; Wu et al., 2013) and in factory or on-site construction housing (Pun &
Smith, 2007; Swider, 2016). But group rentals are regular high-rise units, illegally turned into
overcrowded dormitories. Instead of being spatially isolated, they are embedded in the regular
urban built environment. The spatiality points to a different, newly emergent housing need, one
that has not been captured by the literature or the formal market.
Since migration in China is primarily employment-driven (Cui, Geertman & Hooimeijer,
2014), I hypothesize that proximity to a different kind of jobs is driving the need for a different
kind of locational accessibility. The migrant population is huge – now counting more than 285
million people and roughly 40% of Shanghai residents – and is becoming more and more diverse.
Not all are laborers but more and more come to the city with degrees in higher education (Xinhua,
2018a; Shanghai Municipal Statistics Bureau, 2011). Hence, I hypothesize that it is not just
migrants, but specifically migrants requiring a different kind of urban access, who rent beds in
group rental housing.
To address the two central research questions in this chapter– Who is the group rental
market for? And what does this market offer? – I draw on web scraped online advertisement data
and corresponding market data collected in the field. In particular, I use quantitative text analysis
of the online advertisements’ descriptive text bodies together with experiences and conversations
in the field to explore who landlords and brokers are targeting as potential renters and what
13
demographic group rentals are housing. Through mapping I analyze the spatial distribution of
advertised and real bed spaces for rent. In addition to studying spatial patterns, I also draw on
descriptive statistics and hedonic regression analysis in order to tease out what the market data can
reveal about group renter preferences.
2.2. Mixed Method Data Collection
2.2.1 Online Advertisement Data
I first learned about the market for bed spaces by looking through the “shared housing” ( hezu)
section on 58.com, China’s largest online marketplace for classified ads.
7
Narrowing the search to
the sub-category “bed space for rent” ( chuangwei chuzu), I saw hundreds of thousands
of daily ads offering beds for rent across all major Chinese cities.
The rise of urban big data, such as online advertisement data, holds great promise to reveal
more of what is going on in our cities, especially what is happening informally. While the majority
of housing market research has focused on formal homeownership, rental housing has been
perennially understudied, to a large degree because of lack of data. Sensors, social media, and
online classified advertisements, however, create digital traces susceptible to urban analysis at
unprecedented granularity and scope. Leveraging these has already started to produce fresh
insights into rental housing market operations (Boeing & Waddell, 2017; Hogan & Berry, 2011;
Hu et al., 2019; Kim, 2016; Li et al., 2019; Liu et al., 2015a).
Classified advertisements have long been used to study the development of real estate
markets (Kim, 2004, 2007; Fraser, 2000). With advertisements going online, this strategy becomes

7
In 2013, 58.com already reported unique monthly page views averaging around 329 million (Reuters, 2015; Shu,
2015).
14
even more feasible (Boeing & Waddell, 2017; Kim, 2016; Rae, 2015). In urban China, using online
markets for research is particularly promising because mobile internet penetration is pervasive and
widely affordable in Chinese cities (CNNIC, 2018).
8

While advertisements do not reveal the ultimate transaction prices, the offer price can still
be used to study relative values as long as the negotiated price is not correlated with the error term.
There could be a problem if, for instance, there was an undetected variable that made the spread
between offered and negotiated differ systematically for a subset of the market. However, I expect
the negotiated price to vary little from the offer price given the low investment in rental market
transactions, as well as the general shortage of affordable housing in Shanghai (Chen, Hao &
Stephens, 2010; Mak, Choy & Ho, 2007; Yang, Wang & Wang, 2015).
For this dissertation, I chose 58.com to web scrape online ads for bed space rentals in
Shanghai. After comparing different sites, 58.com emerged as the site which hosted the most
unique ads, provided the most comprehensive information in each ad, and had the most organized
html script. Choosing Shanghai as the study location was motivated in part because Shanghai,
together with Beijing, had by far the largest number of daily bed space advertisements.
Additionally, my years of experience living in Shanghai gave me local knowledge and access to a
big social and professional network that I could leverage for this research project.
As advertisements were easily identifiable through the “bed space for rent” category,
9
I set
up a web harvesting algorithm to systematically collect and store the advertisement information.

8
With widespread usage, online data has the potential to be less biased by structural barriers to access (Boeing, 2020;
Mossberger et al., 2012). Furthermore, posting ads online is currently still free.
9
Fortunately, I happened to scrape online market data before the government censored the use of the term
“bed space for rent”. In July 2016, the category was removed.
15
Web scraping enables the automatic extraction of structured data sets from human-readable
content. It thus automates an otherwise messy and laborious manual data collection (Mitchell,
2018). The web scraper was written in the Python programming language using the scrapy web
scraping framework (Scrapy Community, 2016). The data collection protocol largely followed the
methodology developed in Kim (2016).
After fine-tuning the script over an eight-week pilot study, I configured the web scraper to
collect data in roughly 10 day intervals between February and June of 2016, with an initial count
of 33,084 ads for Shanghai only.
I applied conservative criteria for identifying and deleting duplicate postings, as well as
removing advertisements that lacked key variables of interest.
10
The final data set counts 3,450
unique advertisements, retaining roughly 10% of the original “raw” data.
11
Because the
advertisements’ HTML script contained geographic references, spatial variables could be
generated.
12

2.2.2 “Real” Market Data
I then proceeded to groundtruth the internet data with fieldwork. With the excitement of new data
streams comes the need for greater critical thinking about this data. In the urban development

10
On 58.com each posting is assigned a unique ad ID, but the platform also allows for its users to re-post a listing.
Additionally, users occasionally re-submit an ad under a new ID by simply copying and pasting a previous ad,
sometimes with slight modifications of content. We only retained listings that had unique ID numbers, unique prices,
neighborhoods, apartment floors, and total number of floors in a building. The largest number of ads we deleted were
those that did not specify how many people were sharing a room (19,040 ads deleted) because one of our major
objectives was to price crowding levels. If the most crowded situations did not provide this information in the ads, our
dataset would be undercounting them.
11
In an analysis of online rental advertisements on Craigslist for a cross-section of US cities, Boeing & Waddell (2017)
end up with a similar share of final observations after raw data cleaning.
12
Online advertisements are geo-referenced. However, the geodetic datum formulated by the Chinese State Bureau of
Surveying and Mapping uses an obfuscating algorithm leading to coordinate shifts of 100 to 700 meters. We employed
a linear transformation to correct the offset in Shanghai.
16
research literature about smart cities, there has been a troubling absence of discussion about its
limitations (Kitchen, 2013; 2014; Shelton, Poorthuis & Zook, 2015; Kim, 2018). Big data
proponents stake claims to greater objectivity, neutrality, and accuracy. With enough volume, so
the argument goes, data can speak for itself, transcending context, heralding the “end of theory”
(Anderson, 2008).
But no data is value free no matter how great its volume and velocity. Social inequity and
politics in the material world also shape the geography of the digital world (Beracha & Wintoki,
2013; boyd & Crawford, 2012; Das, Ziobrowski & Coulson, 2015; Kim, 2019; Markham, 2005;
Nobel, 2018). Hence, I needed to test the relevance and veracity of the online dataset by going into
the field.
To this end, I went to Shanghai in the summer of 2016 and started inquiring after randomly
selected online advertisements.
13
We booked appointments to see apartments throughout Shanghai.
Without exception, real estate brokers asked to meet us on the street and took us to housing
situations different from those advertised online. The rentals were different in terms of all
important housing characteristics: location, price, crowding levels, and amenities. It became clear
that the internet ads were being used to locate potential renters more so than to advertise specific
units.
14

Discovering this obfuscation compelled me to embark on collecting a second set of “real”
market data through surveying the field. To guide the sample selection within the Shanghai region,
I visited government designated economic centers, as well as unofficial “activity centers”

13
The initial site visits were conducted together with Dr. Annette Kim and research assistant Xiangyu Li.
14
Note that the data collection process alone already highlights a fundamental insight about using online content for
housing research: there is a significant distinction between the object of study (the actual group rental housing market)
and its representation in online communication. For a more detailed discussion see appendix.
17
identified by research using mobile phone signal data (Zhong et al., 2017). These locations,
combined with the clusters identified by mapping the web scraped data, informed the market data
collection sites during June and July 2016 (Figure 1)
I assembled a research team of six – three female and two male local graduate students in
addition to myself – to systematically call brokers and landlords listed in the online ads and book
appointments to see units. All inquiries were made in teams of two such that one person would
talk to the landlord and tenants while the other took notes and pictures. I rotated to accompany
different research assistants in order to ensure quality of data collection.
Field research teams recorded actual locations, rents, crowding levels, amenities, and
tenant genders. We also discussed the terms of the contract and rental regulations, took photos and
field notes, and engaged in conversations with current tenants and brokers to learn more about the
living situation and common renter demographics. This participant observation helped
contextualize both the field survey and the advertisement data. The site visits brought to life the
spatiality of the crowded living conditions in this market. Interviews with landlords and house
managers were a chance to verify preferences stated online.
.

18

Figure 1: Mapping Web Scraped and Field Survey Data

Source: Online advertisement data collected from 58.com; field survey location selected based on government designated economic centers, mapping of
cell phone signaling data (Zhong et al., 2017) and pilot web scraped online advertisements; site visits during June and July 2017. “Spatial selection” here
refers to spatial boundary of data used in the hedonic price model analysis.

19
In total, inquiring about 132 online advertisements led to roughly 177 apartment
observations as brokers/landlords frequently showed additional apartments upon request.
15
In
summary, the data collection generated three distinct datasets:
(1) Field survey data collected during field work, 177 apartments
(2) Matched online advertisements that were the basis for fieldwork, 132 apartments
(3) Scraped online advertisement data, 3,450 apartments
Because this is a market for bed rentals, rather than entire apartments, I created multiple
observations from each apartment visit. Typically, within the same apartments, prices per would
differ depending on factors such as how many people shared a room. The online advertisements
also often mentioned a range of rents and a range of crowding levels. Informed by fieldwork, for
the scraped dataset, I hence matched the lowest rent price with the highest level of crowding listed
and the highest rent with the lowest level of crowding listed, and conservatively deleted any values
in between that did not state specific price levels. The data processing led to the following three
datasets with the bed as the unit of analysis:
(1) Field survey data, 747 beds
(2) Matched online data, 194 beds
(3) Scraped data, 4,209 beds

15
For a subset of the online data we can and compare one-to-one actual versus advertised group rentals. A detailed
discussion of the veracity of the online data set can be found in the technical appendix.

20
2.3. Findings
2.3.1 No Housing for Young and Educated Migrant
I start the exploration of group renting as a housing phenomenon by addressing the question of
who the target demographic of this niche market is. Recall that the online advertisements scraped
from 58.com all contained rich descriptive text, often relating in surprising detail what kind of
renter the advertisers were looking for. This communication between landlords/brokers and a
target tenant group can give a first indication of who this market is intended for. I mined this text
data by automating the counting of key phrases and then grouping them into meaningful
categories.
16

Analysis of this online content suggests landlords are seeking to attract a young and
educated target group. As Table 2 details below, a quarter of the descriptive advertisement texts
used words referencing youth – 70% if one counted the word “student.” Almost a fifth (18%)
mention a specific age limit; the maximum age allowed ranging from 28 to 32. Phrases such as
“recent graduate,” and “looking for a job”, or “already employed” were used in about 70% of the
ads. In roughly 20% of the ads, white collar workers were listed specifically as preferred. Around
one third (32%) also mentioned traits that were unwanted such as drunkenness, gambling,
noisiness and night shift workers; 11% explicitly stated seeking “compliant” ( fucong) tenants.
Fieldwork confirmed this target tenant profile sketched out online. Many of renters in the
units we visited were indeed recent graduates, predominantly coming from smaller cities and lesser

16
Recently, similar text analysis methods have been deployed on scraped ads of American rental housing markets to
display how their texts code racial and economic segregation in the US housing markets (Boeing, 2019).

21
known universities. Out of the 84 tenants we spoke with, two thirds (67%) worked in white collar
or service jobs.
Table 2: Desired Tenant Traits, Web-Scraped Advertisement Data (n=3147)
Tenant Traits Chinese Term Frequency
Looking for a job
,
2041
Currently employed/ in training , 2405
White collar worker 607
Student/ University graduate
, ,
2193
Young , 781
Age restriction 564
ID

878
Newcomer 175
Obedient 360
Unwanted traits: drinking, gambling,
noisiness, working night shifts
, , , 987

Furthermore, the characteristics about being quiet and keeping regular hours indicate how
this informal housing situation requires furtive practices. Management of renters was clearly a
concern. While only 18% of online ads mentioned specific age limits, more than half of broker
and landlords we talked to (57%) would not accept tenants over 30. They explained the strong
preference for young renters with perceived greater compliance. For the same reason, several
voiced a preference for male tenants; they, too, were seen as “easier to manage.”
17
The renters’
compliance is needed for the subterfuge of these illegal rentals in the midst of formal units.

17
Note how perceived “difficult women” contrast with cultural expectations of women as docile, demure, and quiet.
Scholars have documented the ways in which gender stereotypes are invoked in workplace settings, as a means of
regulating women’s behavior and to justify differential treatment of women. What is instrumentalized in China is
often an appeal to “feminine virtues”, including being compliant (Hanser, 2005; Lee, 1998; Pun, 2005; Rofel, 1999).
The landlord preferences are reflected in the gender distribution among actual tenants: only one third of the beds we
saw during site visits were rented to women. The gender angle on this housing phenomenon is developed further in a
forthcoming article.

22
Site visits also made the high demand for bed spaces apparent. Brokers and landlords would
reserve a bed for hours at most and typically rent out vacancies the same day. The ability to state
disqualifying traits and be selective about renters, together with rapid changes in occupancy
strongly indicate a seller’s market. Shanghai’s group rentals appear to be filling a missing housing
market for young migrants, educated at second-tier institutions.
We know that informal housing emerges when a housing need is not met in the formal
market (De Soto, 1989; 2000; Durst & Wegmann, 2017; Harris, 2018; Ward, 2019), but we also
know that the conditions that give rise to informality do not develop in a vacuum. Rather,
informality is always also the product of socio-political context and government actions (AlSayyad
& Roy, 2003; Feler & Henderson, 2011; Perlman, 1979; Roy, 2005; 2009).
Key to understanding the emergence of group rental housing is to take note that bed space
rentals house migrants. In contemporary urban China, private rental housing is the direct
consequence of transition and, most importantly, migration. After years of supplying heavily
subsidized public rental housing for all urban citizens under socialist rule, urban governments
switched to a policy of privatization and owner-occupation as part of the country’s economic and
political transition (Huang & Clark, 2002; Yang & Chen, 2014). Urban economic opportunity
fueled what has been called the “largest migration in human history” (Chan, 2013). A total of 286
million people migrated cityward since the country eased mobility restrictions and moved towards
an urban system of production (Chan, 2013; Xinhua, 2018a). The Yangtze River Delta, with
Shanghai at its heart, is China’s largest migrant receiving region (Xinhua, 2018a). In Shanghai,
roughly 40% of city dwellers are migrants (Shanghai Municipal Statistics Bureau, 2011).

23
With rapid urban expansion and the new housing policy framework, the urban housing
stock grew quickly and housing prices skyrocketed as homeownership became a major investment
vehicle (Logan et al., 2009; 2010; Wang & Murie, 1996; Yu, 2006). Urban locals profited in
particular since they were the first to become private homeowners when public housing was sold
off into the private market (Chen & Hao, 2010; Mak et al., 2007; Yi & Huang, 2014). Chinese
familism and cultural attachment to homeownership further powered quick acceptance of urban
homeownership (Ting & Chiu, 2002). Today, homeownership in China is at 90% – extremely high
in comparison, for instance, to roughly 64% in the US, and only just over 51% in Germany
(Goodman & Mayer, 2018; Cui, Deng & Lu, 2019).
18

It is thus the demand for housing generated by the millions of urban newcomers that
stimulated the development of (informal) private rental markets. To this day, the Chinese
household registration system hukou continues to regulate access to urban private property markets
and government assistance. Given these restrictions and without access to public rental housing or
homeownership subsidies, the housing choice set for migrants is effectively confined to private
renting (Huang & Tao, 2015; Liu, Wang & Tao, 2013; Shi, Chen & Wang, 2016; Wu, 2002; 2004).
Earlier work documents migrant settlements mostly on the urban periphery (Wu, 2008).
The informal and semi-formal housing markets that have developed in response to massive
increase in demand with rural-to-urban migration include urban village rental housing (Song et al.,
2008; Wu et al, 2013; Wu, 2016a) and employer-provided housing in factories or on construction
sites (Pun, 2005; Pun & Smith, 2007; Swider, 2016). However, employer-provided housing is
declining and renting has become the dominant tenure form for migrants. In parallel, migrants are

18
In Shanghai and Beijing, homeownership rates have jumped from an average of around 20% in 1994 to more than
60% in 2000, and to more than 80% today (Tang & Coulson, 2017; Yu, 2006).

24
increasingly found dwelling throughout Chinese cities (Li & Duda, 2010; Li & Wu, 2008; Liao &
Wong, 2015). In Shanghai, most migrants (78%) are now renters (Shanghai Municipal
Government, 2017).
Beyond aggregate spatial patterns the notoriously poor living conditions, very little is
known about the actual price levels, trade-offs, needs or living arrangements of migrants (Li et al.,
2018; Huang & Yi, 2015; Kim, 2016; Shen, 2015; Zhang & Chen, 2014). Much of the existing
literature treats migrants as a single large group; Literature that reflects the diversity within the
migrant population is just emerging (Cui, Geertman & Hooimeijer, 2014; Xinhua, 2018a). Works
that distinguish between skilled and unskilled migrants’ housing needs find that skilled migrants
disproportionally concentrate in areas with a larger number of professional jobs, which in Shanghai
means the urban core (Cui, Geertman & Hooimeijer, 2014; 2016). Group renting seems to be one
niche (informal) rental market, among others, that caters to a specific migrant demographic: young
and educated.
2.3.2 Trading Space for Access
This section asks what the market data collected online and in the field can tell us about the renters’
revealed housing preferences. I draw on descriptive statistics, mapping, as well as hedonic price
model regression for analysis.

25
Table 3 presents descriptive statistics for the web scraped and field survey data. The table
shows that advertisers tend to paint a rosier picture online, but mean values are by and large
comparable.
19

Table 3: Descriptive Statistics for Field Survey and Scraped Data
+

Field Survey (n=747) Scraped Data (n=3147)
20

Location
Distance to CBD (km) 4.41 4.34
Distance to closest
subway stop (km)
0.43 0.38
Number of subway stops
(within 800m radius)
1.57 1.67
Number of subway lines
(within 800m radius)
2.27 2.35
Apartment
Rent (RMB) 850 688
People per room 6.6 6.6
People per apartment 24.2 -
Size (sqm) - 145.41
Number of bedrooms 3.24 2.95
Number of bathrooms 1.61 1.98
Cooking dummy 0.32 0.03
Tenant Gender
Female 0.32 0.02
Male 0.47 0.11
Co-ed 0.21 0.87
+
The individual bed is the unit of analysis.

19
Refer to technical appendix for detailed discussion on how online and “real” market data compare and what we can
learn from this comparison about the limitations and opportunities of online advertisement data for research. The
remainder of this section uses the market data collected in the field as the benchmark for analysis. In addition to greater
reliability in terms of data quality, the field survey also allowed me to collect and test more variables than the
information provided in the online ads.
20
Note the number of observations here: Instead of 4,209 beds, I only include 3,147 observations for analysis. I come
to this final data set after testing whether spatial non-stationarity is in issue in Shanghai. Given the large urban area,
averages across space may obscure important local variations (Redfearn, 2009). Hence, I separated the web-scraped
data set by concentric distances to the central business district and analyzed descriptive statistics and hedonic
regression results using OLS estimation at various intervals. I found listings advertising group rentals at a distance
farther than 6.5 km from the center appear to be a fundamentally different kind of rental: smaller, more crowded, and
with fewer bedrooms and bathrooms. . Since rentals in the peri-urban area appears to be a separate market, I confined
the analysis to the spatially restricted sample within the 6.5 km circle, resulting in a dataset of the size n=3,147.

26
Descriptive statistics suggest the importance of locational access in the market for bed
spaces. The rentals tend to be in central locations – on average, 4.4 km (2.7 miles) away from the
city center. Additionally, the group rental units typically have good access to public transit, with
an average of at least one subway stop and two subway lines reachable within walking distance.
Spatial analysis lends further weight to locational access being the hypothesized central
driving force. Location is the primary determinant of value in any real estate market, but mapping
both advertised and actual locations shows a distinctive spatial pattern: each set of data cluster in
a ring-shaped form around what is considered to be Shanghai’s urban core (Figure 1). The historic
core is dominated by luxury office and retail spaces, but bed space rentals appear to cluster just at
the edge, quickly dissipating as one moves away from the center (Li & Wu, 2008). While trading-
off the consumption of space and locational access is at the heart of the monocentric city model
(Alonso, 1960; Mills, 1967; Muth, 1969), group rentals push this fundamental calculus in urban
economics to the extreme. The spatial pattern suggests that minimizing the consumption of space
by renting just a bed may only be worth it if it affords very close spatial proximity to downtown.
This calculus is negotiated at the smallest spatial scale. Descriptive statistics (Table 3)
indicate that typically 24 people are crowded into three-bedroom apartments with one or two
bathrooms. A bed in a room shared between an average of six or eight people, costs 850 RMB per
month on average, roughly 127 USD at the time of study.
21
Plotting rent prices by crowding levels
further reveals a fine-scaled pricing of density; Crowding is priced down to the room level.

21
In 2016, the average monthly income of migrant workers in Shanghai was 5,328 RMB (Chen, 2017). This income
might be skewed lower end because it includes uneducated migrants, but in either case this implies that group rentals
provide an affordable option, roughly 16% of monthly income.

27
Table 4 shows the mean rent price per bed, depending on the total number of beds per room.
Note that numbers are even because beds are usually bunk beds. The data clearly indicates the
rational pricing of crowding: the more beds per rooms, the cheaper the rent. This inverse linear
association between rent price and crowding suggests that group renters engage in tough trade-offs
on tight budgets.
Table 4: Rents by Level of Crowding – Comparing Market Survey versus Web Scraped Data

People per Room Frequency Mean Rent per Bed in RMB (~USD)
Field Survey Scraped Field Survey Scraped
1 4 0 1250 (~180) -
2 35 238 1007 (~146) 1397 (~202)
4 153 364 1078 (~156) 1051 (~152)
6 249 1,137 931 (~135) 727 (~105)
8 188 1,137 668 (~97) 535 (~77)
10 86 266 625 (~90) 509 (~74)
12 32 0 589 (~85) -

Also note the sharp drop in average rent for beds in rooms shared by more than six people.
Fieldwork showed that the density levels were achieved either through beds placed in awkward
common spaces such as hallways or extreme crowding in rooms with very little space per person.
Surveying the field, we saw beds in kitchens, storage rooms, and walk-in closets – improvised
spaces with extremely small square meter counts. Figure 2 illustrates the crammed conditions with
a sketch drawn from memory after visiting a particularly crowded unit in the summer of 2016, as
well as a series of photos taken during site visits.

Figure 2: Left: 50 People Group Renting in a Four-bedroom Apartment, Right: Pictures Taken during Site Visits

Source:
Left: Sketch drawn from memory drawn after site visit
on June 28
th
, 2016.
Right: Pictures taken by field research team during site visits in
June and July, 2016.

28

29
These trade-offs can be further explored through hedonic price model analysis. The
underlying assumption of hedonic price models is that the coefficient estimates for housing
characteristic variables express the respective contribution to the rent price. Under this model
assumption, coefficient estimates reveal marginal willingness to pay, or the tradeoffs that market
participants make (Chen and Hao, 2010; Kim, 2016; Redfearn, 2009; Rosen, 1974).
22

Table 5: Hedonic Regression Results
ß
, Field Survey Data
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Location
Distance to CBD (km) -0.037***
(-5.07)
-0.026***
(-4.11)
-0.026***
(-4.46)
-0.023***
(-3.91)
-0.023***
(-3.93)
-0.021***
(-5.42)
Distance to closest
subway stop (km)
-0.078
(-1.27)
-0.063
(-1.22)
-0.002
(-0.04)
-0.088*
(-1.75)
-0.090*
(-1.85)
-0.081**
(-2.52)
Number of subway lines
(within 800m radius)
0.029**
(2.51)
0.031***
(3.22)
0.014
(1.58)
0.039***
(4.14)
0.036***
(3.91)
0.006
(1.00)
Apartment
People per room -0.075***
(-17.43)

-0.074***
(-17.62)
-0.073***
(-17.93)
-0.033***
(-10.92)
People per apartment -0.018***
(-20.37)

Number of bathrooms

0.131***
(7.07)
0.118***
(6.57)
0.047***
(3.83)
Lower bunk dummy

0.150***
(7.11)
0.112***
(8.00)
Cooking dummy

0.487***
(30.6)
N 743 743 743 743 743 743
Adj. R-squared 0.058 0.332 0.399 0.374 0.418 0.741
F-statistic 16.25 93.17 122.7 89.48 88.02 305.12
ß
In all model specifications the logarithm of the monthly rent price per bed is the dependent variable.

Table 5 summarizes the estimation results. Overall, the model performs well as reflected
by adjusted R-squares. Again, locational access emerges as an important feature in the market for

22
Bayer, Ferreira & McMillan (2007) address the issues of supply and demand affecting the interpretation of hedonic
model estimates. Mean and marginal willingness to pay may be similar or vastly different depending on the group of
“buyers” and the relative supply of a given housing feature in the market. If “buyers” are relatively homogenous, as
appears to be the case in the market for bed spaces, then coefficient estimates are likely to reflect the marginal
willingness to pay of the average renter.

30
bed spaces. In line with housing market theory, the coefficient estimate for distance to center is
highly significant, negative, and stable across specifications, especially after the introduction of
the crowding variables.

Existing work established that Shanghai’s housing markets generally
display a monocentric structure and distance to center functions as a proxy for access to
employment (Chen & Hao, 2010; Li et al., 2019). Here, being one kilometer further from the center
reduces the rent by 2-4%.
Access to transit, proxied for by distance to the closest subway station is also statistically
significant and in the expected direction: Being one kilometer farther from the subway stop reduces
rent by 8-9%. The premium for proximity to transit access may, however, be more about time
savings rather than commuting cost savings since recent work on migrants’ job housing-balance
suggests that the majority of low income workers walk to work (Kim, 2016; Li & Liu, 2016). Note
also that the estimates on the number of subway lines accessible within walking distance are
positive and significant in most specifications. Such subway hubs may signify access to
employment centers more than access to transportation (Guihare & Hao, 2008).
The hedonic models also pick up on the granular pricing of personal space. Crowding is a
significant price structuring variable. Crowding is measured at the apartment level (“people per
apartment”) and at the room level (“people per room”). However, pairwise correlation analysis
shows that the total number of people per apartment is correlated with both the number of people
per room and the number of bathrooms. Hence, I exclude the variable for people per apartment in

31
the preferred models and only include the number of people per room and bathrooms. I find that
adding another bed to a room decreases the rent price by 3-8%.
23

The pricing of apartment level amenities further reflects the crammed conditions. Given
the crowded quarters, the number of bathrooms is a valued amenity. Having an additional
bathroom increases the rent by 5-13%. Additionally, in 44% of the apartments we visited, rent
prices differed for upper versus lower bunks. In tight spaces having the lower bunk functionally
means having somewhere to sit and comfortably accessing to storage space below the bed. The
lower bed garners a 11-15% price premium.
The preferred model (Model 6) shows that the ability to cook almost doubles the rent and
increases the R-squared to 74.1%. In roughly half of the apartments we visited, the kitchen was
unavailable for cooking because it had been converted into an additional “bedroom,” typically
within very crowded overall situations. Given the field experience, I interpret this variable to be a
proxy for better apartment living situations, in addition to the cost savings of cooking at home.
24

Note that coefficient estimates on variables describing the living conditions – crowding
and apartment level amenities, most impressively cooking variable – have an economically larger
effect on price than those describing locational access. Given the tight, ring-shaped spatial
clustering of group rentals and extremely dense transit coverage in central Shanghai, I attribute

23
I also estimate Models 4-6 with the number of people per apartment instead of people per room and number of
bathrooms. I find comparable coefficient estimates, adj. R-squared, and F statistics.
24
Note that none of the online advertisements mention a price distinction between upper and lower bunk beds.
Cooking, too, was mentioned in only 3% of the adds.

32
this finding to a lack of significant spatial variation with regard to location access of group rental
apartments.
25

25
I also run hedonic regressions on the online data set and find the analysis is somewhat impacted by discrepancies
between advertised and real conditions, as well as omitted variable bias. Please see the appendix for a detailed
discussion.

33
CHAPTER 3. DECONSTRUCTING THE SOCIAL MEANING OF GROUP RENTING
3.1. Introduction
The previous chapter explored group rental housing as an informal housing sub-market. Drawing
on web scraped online advertisements in combination with market survey data collected in the
field, I find that group rentals house young and educated migrants who trade off minimal personal
space and amenities for superior locational access to employment centers. This trade-off on small
budgets is further reflected in a fine-scale pricing of crowding, such as different prices depending
on the number of people share a room as well as rent price differentials for upper versus lower
bunks. I also show that while scraping online advertisements was a quick way to amass data –and
in fact the first means for me to “see” the hidden housing market – local knowledge and fieldwork
were needed to more accurately decode revealed group renter preferences.
The housing market analysis already unveils some key elements about the group rental
phenomenon. In particular it outlines the tenant group as primarily educated migrants and
identifies proximity to the downtown employment center as a key driver in this market.
However, the findings presented in chapter 2 also stimulate new questions, questions
regarding motives, sensemaking, and meaning, which cannot be addressed with the data or
analyses employed in chapter 2. In this chapter, I hence draw on ethnographic fieldwork to address
research questions three and four – Why do tenants rent in this market? And how do tenants process
their experiences?.
Another way of posing research question three – Why do tenants rent in this market? – is
to ask about the specific housing need that group rentals are filling, beyond what could be learned

34
from the housing market data. In relation to research question three, I hypothesize that group
renting is part of a longer term strategy. Analyzing the descriptive texts of online advertisements
revealed the frequent use of phrases such as “newly arrived in Shanghai”, “recent graduates”, or
“looking for work” (Table 2). All of these phrases describe transitions: from a different city to
Shanghai, from studying to working, from one job to another. Group renting could be one of the
ways this tenant group manages such moves. Relatedly, the choice of migrating to Shanghai,
widely known to be highly competitive in terms of labor and housing markets, over a smaller, less
competitive city strikes me as pointing towards a longer term goal motivating the group renters’
decisions.
I further hypothesize that the group renters’ dual identity as migrants and college graduates
matters. There exists a large literature documenting the institutional barriers associated with the
hukou system (Fan, 2002; 2007; Li & Liu, 2016; Li, Li & Chen, 2010; Solinger, 1999), including
unequal distribution of access to education (Fu & Ren, 2010; Montgomery, 2012) and social
capital (Liu, 2005; Liu, Wang & Tao, 2013; Wang, Li & Deng, 2017). Furthermore, while the first
generation of migrants were almost exclusively laborers, the migrant population has since
diversified (Wang et al., 2017; Xinhua, 2018). Educated migrants likely rent bed spaces due to a
newly emerging set of obstacles. Research question three thus also asks about the structural forces
that give rise to the housing need, which is in turn filled by group rentals.
Research question three inquires about the motives behind group renting, but research
question four asks about the social cognitive processes involved in the group rental experience.
Studying cognitive processes is about understanding the shared cognitive schemas through which
people construct and interpret human experiences (D’Andrade, 1981; 1992; 1995; Quinn et al.,
1992).

35
With research question four, I ask about mental processing and the presence of social
constructs as a window into the underlying social contract. As group renting appears to be a
common enough experience among a certain demographic, I hypothesize that there exists a societal
narrative, a set of social institutions that give meaning to group renting and help with sensemaking.
The fact that group rentals are able to hide within the built environment, although clearly in
violation of city planning ordinance and involve the participation of many actors, further lends
weight to this hypothesis that there are social narratives that facilitate their coordination. Exploring
the individual sensemaking as well as the social context in which it takes place, I am linking my
questions about changing mechanics of social mobility to a broader inquiry into the nature of class
relations and class consciousness in contemporary China.
To be able address these questions, I needed to collect a different kind of data and employ
a different method. In the summer of 2018, I returned to Shanghai for ethnographic fieldwork, in
particular semi-structured interviews and participant observation. The goal of this part of the
dissertation is to get to what Geertz (1973) in reference to Gilbert Ryle has called a “thick
description” – that is, an investigation into the social meaning of group renting.
Towards this end, over the course of three months, I met 14 current and former group
renters for one to two hour semi-structured interviews.
26
I also rented a bed in a crowded unit for
about two weeks. During that time, I was able to observe the daily negotiation of overcrowded
living and got to know the seven other women I shared a room with. In order to be able to compare
and contrast, I also interviewed 16 recent college graduates who did not rent just a bed, but rented
a room of their own, or shared it with just one other person. In addition, I spoke to several landlords

26
I received IRB exempt status for this research and got verbal consent from everyone I talked to.

36
and brokers, of whom I got to know one very well. In the summer of 2019, I followed up with nine
of my interviewees from 2018, five group renters and two non-group renters.
To analyze the interview and field notes data I used thematic content analysis. In this
qualitative data method, the material is processed through identifying first codes and later themes
and patterns (Hsieh & Shannon, 2005). Codes are obtained through a simultaneous processes of
extracting meaning and breaking down the text into meaningful segments. Chosen such that the
full set of codes systematically “map[s] the informational terrain of the text” (MacQueen et al.,
1998: 33), codes add information to the material, rather than reducing it (Sandelowski, 1995; Tesch,
2013). To produce themes from the initial codes, I used both grounded theory coding
27
(Padgett,
2012; Tesch, 2013) and constant comparative analysis
28
(Glaser & Strauss, 1967; Strauss & Corbin,
1994). Constant comparative analysis, which systematically decontextualizes and recontextualizes
data to find patterns, can be helpful when categories that are suspected to matter are known a priori
(Tesch, 2013). Here, I compare and contrast based on housing arrangement (group renting versus
other), gender, level of educational attainment, and status as current versus former group renter by
the time of the interview.
3.2. Accessing the Field
3.2.1 Fieldwork as an Outsider: What Did I Get to See?
Every city has within it a multiplicity of cities and city experiences, depending on who is
experiencing, and who is telling the story (Miraftab & Kudva, 2014). During my fieldwork I

27
After “open coding” in the first, inductive stage, grounded theory coding moves into “focused coding,” which uses
potential themes that have formed during the first stage to fine-tune coding by going back to the raw data and going
over coded and un-coded segments in a more directed, selective manner. Finally, axial coding aims at mapping out
the dimensions of potential themes by finding the relationship between codes (Padgett, 2012).
28
Constant comparative analysis separates groups by different categories to uncovers how themes vary across different
characteristics (Strauss & Corbin, 1994).

37
learned that there were (at least) two more Shanghais than I had previously been aware of: one
where everyone knew about group rentals and one where nobody had a clue about their existence.
I eventually got to know the group rental Shanghai through talking to group renters,
through hearing their stories, and through talking to brokers, landlords, and immediate neighbors.
But I first knew the other Shanghai, the one where nobody knew about them. The divide started to
dawn on me before I even started meeting with people, when I started preparing for fieldwork by
thinking about strategies to recruit informants.
Before going to Shanghai in the summer of 2018, I worried about how to locate group
renters and how to build trust. I worried, because I was quite certain that my own social network
– usually the best way to meet new people and establish trust – would be of little help this time.
Most of the people I had gotten to know over the years were students, and eventually
graduates from top schools, like Tongji, Shanghai Jiao Tong, the universities I had attended as a
foreign exchange and Chinese language student. Many of them spoke English, could sympathize
with me learning Chinese. Some of them had also travelled or lived abroad. We became friends
because shared experiences helped bridge cultural differences. My friends in Shanghai were (or
were in the process of becoming) part of the cultural elite. I hadn’t fully appreciated, however, just
how much their privilege insulated their experiences.
Watching the research assistants I hired for our initial field work in 2016 react to what they
saw with horror and disbelief had been my first indication of this divide. Over the course of roughly
three weeks my local student research assistants and I were calling up and visiting group rentals
daily in order to collect data on the group rental housing market directly from the field. We would

38
meet to debrief at the end of each survey day. But the most impressive scenes were shared as they
happened, via our common WeChat
29
group.
Some site visits left us stunned because of their absurdity. We saw group rentals in former
kindergartens, where the spaces had been packed with beds without bothering to paint over the
childish murals of fairytales, flowers, and butterflies; or a pet shop-turned-group rental where the
owner had changed businesses so recently that unsold pet supplies were still stacked up against
one wall of the shop floor. Other site visits were as disturbing as they were saddening – like the
group rental that housed 34 people in tight quarters above ground, but offered an even cheaper
option, renting just a mattress on the floor in a room without windows, below ground to another
16 people. There were also decent, lower density arrangements that seemed well taken care of, but
the majority of the places we visited were cramped and stuffy, full of beds and bodies.
The research assistants, who were all graduate students at elite universities in Shanghai,
were just as shocked – if not more – as I was. They told me, more than once, how they had
“absolutely no idea” people lived like this right in the center of Shanghai.
A couple of my friends had even been renting in a building with multiple group rentals for
years without knowing about them. We found out only because I stayed in one of them. While my
friends shared a two bedroom apartment on the sixth floor between four women, the group rental
I stayed at housed 22 women within roughly the same layout, just six stories below, on the ground
floor. When one of my friends helped me move in to the group rental, she couldn’t believe what
was happening literally in front of their door steps. Even when they were close – either spatially

29
WeChat ( weixin) is China’s most popular social media app, which includes the ability to send and receive
money as well as pay in stores and for public transit, as well as booking planes, trains, and hotels (among others). The
number of active registered users reached over 1 billion in 2018 (Jao, 2018).

39
or in terms of demographics – group renters were essentially invisible to their more privileged
peers.
During my fieldwork I got to see both worlds, the one where nobody knows and the one
where everybody knows. Accessing the former was relatively easy. I first came to Shanghai as a
foreign exchange student almost a decade ago to spend the summer and fall of 2010 at Tongji
University. Since then I have returned to Shanghai almost every year: for Chinese language studies,
internships, and to visit friends. Over the years, I have built a large social network in Shanghai and
language fluency has helped me maintain and expand this circles of friends, colleagues, and
acquaintances.
Accessing the other Shanghai, the one where everybody knew about group rentals, was
much more difficult. The limitations of my immediate social network aside, landlord and brokers
went to considerable length to hide their illegal rental activity. In fact, it was only through
technology and the online advertisements that I even got a first glimpse of this world. Although
there are a number of personal accounts of former group renters on online forums, group renting
is rarely in the news; and when it is, media coverage casts group renting in a narrative of chaos
and lawlessness, focusing on public safety hazards and criminal landlords, describing group rentals
as rare incidents of code violation, not a wide-spread social phenomenon (e.g., China News, 2014;
Hu, 2014; Wu, 2016b).
To meet and talk to group renters, I needed to mobilize the social networks of my friends
and acquaintances, but even then recruiting interviewees proved challenging. Most of the people I
know claimed not to know anyone who was group renting. In the end, all of the 14 current and
former group renters I did meet in 2018 were friends or relatives of friends, but in some cases there

40
were up to three degrees of separation between my interviewees and me. I met only one of them,
Ninghong, without a direct personal introduction. Ninghong had posted about looking for a new
roommate in a Tongji alumni WeChat group. I texted her, asking whether she would be willing to
meet with me to discuss her housing situation and, to my surprise, she agreed. In our conversation
I found out that she had been intrigued by my research and had once also considered a career in
academia. In this one case, sharing interest for research and both having attended Tongji was
enough to establish a connection.
In all other cases, however, the personal introduction turned out to be crucial. For one,
group renting is illegal, which puts group renters in a vulnerable position when talking about it to
a stranger. Additionally, many were reluctant, some even embarrassed to talk about their
experiences. Having a friend vouch for me hence went a long way.
Typically, the first contact would be via WeChat. Most of my interviewees did not speak
English and felt wary about talking to a foreigner. Through texting I could put them at ease about
my Chinese language skills and also answer any questions they had about the research and what
kind of things I would ask them about. If I could still sense uneasiness, I would suggest to invite
the person who introduced us to join us for the initial interview, which often helped created a more
relaxed setting. In general, I met all of my interviewees in public spaces, often over food or for
coffee or tea. I also compensated all of them for their time.
My identity as a student, I believe, was key in allowing me to traverse both worlds. After
all, most of the young people I wanted to reach were (recent) college graduates and could
sympathize with student research. Being close in age and fluent in Chinese also helped.

41
Being a foreigner, on the other hand, cut both ways. On the one hand, my interviewees
often assumed ignorance and were thus perhaps more elaborate in their answers and explanations,
more understanding with my many questions of clarification. I was also generally afforded the
benefit of the doubt, a privilege that allowed me to ask perhaps more exploratiory questions. On
the other hand, by the same logic, it is very possible that interviewees held back on some details
of their stories, for instance for fear of being misunderstood if indeed I lacked sufficient contextual
knowledge. And of course, I may very well have missed some of the nuances in what they were
communicating to me.
Still, I believe not being Chinese was neither good nor bad for this research, rather it likely
introduced different kinds of biases. Ethnographies are often conducted by researchers that are to
some varying degree outsiders to the communities they work with. At least in this case, the outsider
role was made very explicit through my physical appearance.
3.2.2 My Bed Space Renting Experience
Apart from interviewing, I also rented a bed myself. Through this immersive participant
observation, I wanted to learn about the everyday life in group rentals, I wanted be able to better
contextualize what I was learning through interviews. Although my experience would never truly
mirror those of the young people I talked to, I hoped to understand group renting on a different
level, using my own body. Staying in a group rental myself, I wanted to understand what it is really
like living with so many strangers in such tight quarters. How are daily routines affected by
crowded living? What is the general atmosphere? Is it tense? Is it safe?
It took me weeks, however, to find a group rental where I could rent a bed. By the summer
of 2018, the website 58.com, which I had used to web scrape online advertisements listing beds

42
for rent two years earlier, had developed an app which included a messaging service. Not needing
to call to inquire about available beds, I could initially conceal my identity as a foreigner when
texting the contacts listed in ads, which I identified as potential group rentals from the
advertisements’ pictures. Inevitably, however, at some point during the text message exchanges
they would ask what I was doing in Shanghai. When I answered I was a student from outside of
China, it was always the end of the exchange.
It took me several weeks and texting with 19 brokers and landlords until I finally found the
group rental where I was eventually accepted as a short-term renter. After texting back and forth
for only a couple of minutes, the young man who managed the all-female group rental I ended up
staying at agreed to meet me outside a convenience store close to the group rental apartment. We
chatted for a few more minutes before he decided he would rent to me and started explaining the
payment and house rules.
I rented a bed for about two weeks in September 2018. “My” group rental housed 22
women, but I mostly got to know only the seven other women in my room. As the next section
(Section 3.3) discusses in more detail, the tight spatiality of the group rental complicated talking
to and getting to know my fellow renters. Because there was no other furniture besides bunk beds,
one’s bunk was really the only place to be. At the same time, the bed was also the only space that
was not shared, the only private space in a sense. Without anywhere to hang out, it was difficult to
know when it was ok to strike up a conversation. The women’s work schedules also made it
difficult to talk a lot. Most of them worked six days a week. In the mornings, they rushed to get to
work and at night they typically came home after dinner, mostly just to shower and to rest. In
addition to the altered spatial dynamics, the window of time to talk was also very limited; even
when the other group renters were “home”, the women often seemed exhausted.

43
From the beginning, I thus mostly left it up to my roommates to initiate conversations. In
fact, the lines of communication opened only slowly and on their terms. My first night at the group
rental nobody really talked. Nobody said much, not to me but also not really to each other. I only
asked a couple of questions about where I could put my things, got curt answers, and let it be.
After all, I did not really know what to expect since my informants’ experiences had been quite
varied.
The next night, one of the women in my room made a joke about a fight over the bathroom
unfolding down the hall, and as I joined in on the laughter, the ice broke. Having laughed together,
the women seemed to warm up and asked me a couple of questions about myself and why I was
there. The shared moment seemed to ease a tension about me being there and from then on they
resumed chatting with each other. Every now and then they threw a couple of questions my way
and including me in their conversations, but mostly they just ignored me. I only got to know one
of them, Meifeng, a little better. Since she did not start work until 1:00 pm, we spent several
mornings together at the group rental and on some of them we talked.
3.3. Overcrowded Living
Before addressing research questions three and four – Why do tenants rent in this market? And
how do tenants process their experiences? – this section introduces group renting through the lived
experience and the everyday. In this section, I describe the crowded spaces, associated every day
challenges, and some of the response and coping strategies of group renters. I draw mostly on my
participant observation renting a bed myself but also bring in what I learned through site visits and
in conversations with group renters.

44
3.3.1 Crowded Spaces
It is difficult to bring alive the spatiality of the group rentals with words, even pictures do not quite
seem to capture what the crowded environments were really like.
Most places we visited during summer 2016 had no furniture other than beds, just beds
everywhere, and bodies, everywhere. Our market survey showed that the average group rental
housed 24 people in three bedroom apartments, most commonly with 6 or 8 people sharing
individual rooms. But to house 24 people in such tight corners meant also placing beds in hallways
and living rooms, using flimsy curtains to create what must only be an illusion of privacy. In
general, the more crowded the places, the more we saw of such improvised spaces: beds not only
in hallways and living rooms, but also in kitchens and closets, balconies turned into storage rooms,
and kitchens turned into bathrooms.
There were less crowded, more pleasant group rentals, too. In 2016, we saw a number of
group rentals that only had four beds to a room and no beds in the common areas. Instead, living
rooms and hallways would be real common spaces – with couches, tables, and chairs; some even
had a shared TV. Visiting these less dense rentals, one could feel the difference in one’s body.
Because they were less crammed, natural light could fill the rooms and the air felt almost fresh,
not damp and stuffy as in the more crowded units. These less crowded group rentals also had
closets and shelves to store people’s belongings and some kitchens could be used for cooking.
Overcrowding as a housing issue has a long history in China. Up until the “Reform and
Opening” ( gaige kaifang) policy changes, the majority of urban households lived in
poor housing conditions (Li, Wang & Chang, 2018). The average living space per person was only
3.6 sqm (38.8 sqft) before economic transition, with almost half of urban households living in

45
“severely overcrowded arrangements,” meaning less than 2 sqm (21.5 sqft) of living space per
person according to Zhang & Chen (2014).
30

Although marketization of housing supply led to great improvements in terms of average
dwelling quality and per capita floor space, the crowding problem is documented to persist (Li et
al., 2018; Zhang & Chen, 2014). Overcrowding issues in Shanghai have notably consistently been
worse than national averages (Zhang & Chen, 2014; Chen, 2016). In an effort to crack down on
overcrowded housing, the Shanghai Municipal Government (2011) issued “The Administration
Rules of Rental Housing in Shanghai”, which specifies a minimum living floor area space of 5
sqm (53.8 sqft) per tenant and no more than two people sharing a room. All the group rentals we
visited violated this rule.
The group rental where I rented a bed was on the crowded side. I shared a two bedroom
apartment with 21 other women. The tiny kitchen had been turned into a storage area, crammed
with stuff except for a narrow path leading to the sink that was still used for hand-washing clothes
or brushing teeth. The two bedrooms each had 8 beds, but only my room had a door to close. There
were six more beds in what used to be the living room and one shared bathroom.
The bunk beds, the only place to sit, to be when “at home”, were simple wooden structures;
And they were tiny. One of the landlords told me that he had a guy who regularly built them for

30
Meaning and measures for overcrowding are cultural and highly context-dependent. Myers & Lee (1996), for
instance, define overcrowding in the North American context as more than 1 adult person per room. A recent US
Department of Housing and Urban Development report (Blake, Kellerson & Simic, 2007) states an overcrowding
standard of more than two people per bedroom, while UN-Habitat (2010) defines overcrowding as more than three
persons per room. As an alternative to measuring overcrowding according to people per room, there are also space-
based standards, which are equally wide-ranging: The WHO (2000) defines a minimum floor space of 7-9 sqm (75-
97 sqft) per person as a threshold, while the recent US Department of Housing and Urban Development report proposes
a threshold of roughly 15 sqm (161 sqft) per person (Blake et al., 2007). Literature reviews frequently find that there
is no singular, widely accepted standard measure (Blake et al., 2007; Office of the Deputy Prime Minister, 2004).

46
him. They were extra narrow (“to make the best use of the space”) and made out of cheap wood
(“When there’s a police raid and I don’t hear about it in time and they come and destroy my beds,
it doesn’t cost me too much to replace them”). How small they really were, I only realized when I
laid down to sleep in “my bed” the first night at the rental. My top bunk had no banister and was
so narrow, I was afraid I might roll out at night. Of course I am relatively tall (5'8) but when I laid
down, my head all the way up to the top of the bed and with my feet stretched out, my toes would
touch the edge of the very thin mattress that let me feel every single one of the wooden slats
beneath it.
Figure 3: Pictures of the Room where I Rented a Bed, September 2018

There was one built-in open shelf in the small hallway next to the bathroom and it was
littered with stuff, mostly the women’s beauty and personal hygiene products: a collection of hand
mirrors, hair and toothbrushes, various tubes and bottles with lotions, face masks, shampoos, and
makeup. Otherwise, it was a struggle to store one’s belongings and I learned from watching my

47
fellow renters. We kept things in suitcases and small boxes underneath the lower bunks, in bags
hanging from the bed poles and on hangers dangling from the slats of the top bunk. Everyone also
kept a number of frequent-use items, such as rolls of toilet paper and laptops, in the beds. At first
I wrongly assumed that this was to keep valuables safe, but later I learned it was just a practical
move – there was simply nowhere else to keep things conveniently within ones reach.
3.3.2 Everyday Strategies: Navigating Crammed Conditions
Keeping frequent-use items in the bed was only one of many tricks I learned during my group
rental experience. For instance, everyone only used very small towels, hand and face towels,
because they dried faster. Also, the women typically washed only a few pieces of clothing at a
time, almost every night, because there was nowhere to hang a full load of laundry. On my second
day I bought myself a small plastic basket for my essential bathroom toiletries: toothbrush,
toothpaste, shampoo, and of course toilet paper, everyone had their own. Each of the women had
a bathroom basket and it made bathroom switches much more efficient.
In general, group renting required sacrifices with regard to all the basic human needs: going
to the bathroom, sleeping, and eating.
With only one bathroom shared between 22 women, mornings and especially evenings
were always busy. As it is common in China to bath at night, evenings were veritable shower rush
hours – and occasion for many a squabble. In fact, a scene with one young woman in the shower
and an older one banging on the door from outside, desperately pleading with her to unlock the
door so that she could relieve herself was cause for much mockery and laughter from the women
in my room. It turned out that what broke the ice on my second night was somewhat of a running
gag.

48
The bathroom was no pleasant place. It was always damp and musty, not too surprising
given the 15 plus showers every day. By the time I got to use it at night, which was usually around
midnight, the bathroom would be covered in hair and the trash can overflowing with used toilet
paper.
31

Still, the bathroom was really the only time one could get a few minutes alone. Taking a
shower, the running water all but drowning out the noise from the rest of the apartment, felt like
taking a break from the continual commotion on the other side of the bathroom door. The constant
little battles between women defending their 10 to 20 minutes of privacy and others needing to use
the facilities or needing to get ready to go to bed and get rest did not surprise me.
Finding sleep was my biggest struggle. I was used to small spaces – during my
undergraduate degree in Berlin I rented a walk-in pantry with a lofted bed for nine months – but I
was not used to sharing my space with so many people. At night the light would typically stay on
way past midnight. There was constant noise, people talking, watching TV at full volume on their
cell phones. Mornings were the worst for me. Starting from around 5:00 am, alarm clocks would
go off in what felt like 5-minute intervals, a strange cacophony mixing in with the sounds of people
hustling and bustling about to get ready for work. Additionally, the neighboring building, a former
hospital, was being renovated; Construction started every morning at 7:30 am. I needed a sleeping
mask and ear plugs to at least try and block out all the different stimuli and find sleep.
For my roommates, on the other hand, sleeping was not an issue. My first night at the group
rental, I got there around 9 pm and the only middle aged woman in our room – everyone just called
her “auntie” ( ayi) – was already fast asleep. She had a little hand towel draped over her eyes

31
It is common in China not to flush used toilet paper to not risk blocking the plumbing.

49
and was snoring loudly, much to the indignation of everyone else. Apart from a few snappish
remarks and eyerolls, nobody seemed to mind much and, on the contrary, kept on talking at normal
volume. The lights went off just after 2 am that night, but ayi never woke up once. As I found out
over the subsequent nights, when she went to sleep, she slept like a rock.
The younger women, too, seemed to be quite tolerant to noise. On several other evenings
I found Jing, who everyone called “big sister Jing”, fast asleep with her day clothes still on, all the
while things took their usual course: people coming in and out of the room, chatting, lights on until
late. During one of the mornings when Meifeng and I were alone at the apartment – Meifeng’s
regular work hours were between 1 and 9:30 pm – I asked her when the renovation and
construction next door had started. She paused for a moment and then simply said “Oh, I guess it’s
still going on. I can’t actually remember when it started. To be honest, I don’t hear it anymore.”
My first few days at the group rental, I was nothing short of amazed by the flexibility,
tolerance, and resilience these women displayed, sharing such tight quarters and rarely having a
quiet moment to themselves. I knew that college students usually stayed in university dormitories,
where undergraduates typically shared rooms between six to eight people.
32
I suspected that this
experience must have helped coping with group renting and asked Meifeng about it on another one
of our mornings together. Meifeng had just graduated from a private college in Xiamen, Fujian
province, and sort of agreed “Sure it’s similar. But this is worse.”
But then again, I, too, somewhat got used to the situation after a couple of days. Seven days
into my stay, I was able to sleep using neither sleeping mask nor earplugs. The accumulated lack

32
When I lived in a Chinese university dormitory, I had to share the room with only one other person. At my host
university, this was a foreign exchange student privilege, otherwise only afforded to graduate and doctoral students.

50
of sleep likely played its part. I had always wondered about people napping in the middle of busy
McDonald’s or Starbuck, or on parked Scooters and on the metro. Ten days into my stay, I fell
asleep for an unplanned 45 minute nap, on top of my notebook, in the middle of the afternoon in
a crowded and noisy coffee shop.
33

Group renters also made adjustments with regards to their eating and recreational activities.
Only about half the group rentals we visited in 2016 allowed their tenants to cook. This didn’t
necessarily mean that the kitchen could be used, but that tenants could have personal rice cookers
or hotplates. Where I stayed, I rarely saw people eating anything else but snacks. Ayi had a water
kettle and a small hotplate and would occasionally prepare instant noodles. When I moved in, she
had been there for less than a month. She told me it didn’t take her long to get used to sharing the
space with the other women, who turned out to be mostly around her daughter’s age. “The only
thing I miss is preparing a proper meal. I just cannot get used to eating outside all the time.” Others,
like Meifeng, got most of their meals from the convenient store next door.
Almost all the women I shared the group rental with worked six days a week. Free time
was already limited, but many of them further limited themselves on recreational activities. On
workdays, everyone typically just came back to the apartment, which essentially meant spending
time in their beds. Days off were spent outside. Meifeng, for instance, had gone “shopping” with
colleagues a couple of times, even though usually none of them would actually buy anything. Since
her regular workday was from 1 to 9:30 pm and she frequently worked overtime, she had mornings
rather than evenings off. She complained it never felt like much of a “time off” since she often
slept until 10 or 11 am, sometimes finding the time to watch TV. “I never do anything. I thought

33
Knowing about the short-term nature of my stay and its purpose was another big part. I discuss this in a later section.

51
maybe I would like to join a gym, but it’s too expensive.” When I asked Yiran, a personal trainer
at a high-end gym who lived in a group rental organized by his employer, about how he used his
free time, he didn’t quite understand my question at first. He worked every day from morning till
midnight and only got two days off per month. Eventually he said that he liked watching movies
on his phone or going to the park. Only once had he gone to the Bund,
34
a 90 minute trip away
from his group rental in the Southwest of the city, to take pictures to share with his friends and
family back home.
What I observed and experienced has been documented in the literature: overcrowding in
residential environments generates noise, unwanted social interaction, and interference with sleep
patterns and daily routines (Conley, 2001; Evans et al., 2003; Evans & Lepore, 1992). Lacking
adequate facilities, occupants of overcrowded spaces may compete for shared bathrooms and
adjust their rest and recreational activates. Because residents in crowded shared living constantly
have to negotiate their habits and schedules with those of others, they have been documented to
display elevated stress levels (Campagna, 2016; Hartig, Johansson, & Kylin, 2003; Li & Liu, 2018).
More broadly, social epidemiologists have highlighted an association between cramped living
conditions and adverse effects on psychological well-being (Dunn, 2002; Evans, 2004; Quinn et
al., 2010; Sandel & Wright, 2006). Consequently, lack of adequate living space is generally
considered a key criterion for housing deprivation (WHO, 2000; Blake et al., 2007; Department of
Communities & Local Government, 2014; Eurostat, 2014; UN-Habitat, 2009; 2010).

34
The Bund ( waitan) is Shanghai’s famous waterfront, lined with colonial heritage buildings. It offers a view of
the city’s skyline in the financial center Lujiazui in Pudong, on the other side of the Huangpu River, which runs
through central Shanghai. It is one of Shanghai’s most famous landmarks and tourist attractions.

52
3.3.3 Interpersonal Relations and Overcrowding: Negotiating Sharing in Tight Quarters
Group renters developed different a range of tactical responses
35
in order to deal with life among
dozens of strangers. From what I heard, and saw – from interviews, site visits, and my own
experience – people opted to either keep to themselves or cautiously bonded.
Many of the group rentals we visited during our market survey in 2016 appeared eerily
muted despite the amount of people in them. People would barely look up as landlords or brokers
showed us the places. There would be short exchanges, for instance about using the bathroom, but
other than that the low hum of fans, sometimes ACs, and the sounds of games or TV shows coming
from tiny smart phone speakers would make for the typical soundscape. Longwei, who lived in a
group rental with 29 other men as a temporary solution while he made a career change, told me
“People usually don’t talk. We all grow up knowing that in the big city everyone minds their own
business ( guan ziji). There are a lot of foul characters and dishonest people ( huairen,
pianzi], it’s safest to just keep to yourself.” Mingzhu, who lived in a group rental while she
worked as a street vendor before meeting her husband, getting married, and becoming a housewife
and mother, finds more cutting words: “I tried not to talk to anyone, to look at anyone. The other
people, who are they? For all I know they could be criminals, (fengxie).
36
”
Several of the group renters I talked to, including Mingzhu, also reported abusive landlord
behavior, adding to the frosty atmosphere and transactional feel of many of the shared living
arrangements. Mingzhu talked about how her landlord had installed individual meters in each room
to monitor the tenants’ electricity use and charge them accordingly. He also didn’t return her

35
The emotional and mental energy group renters invested in coping and processing the renting experience as part of
their respective biographies is addressed in section 3.6.
36
This is a reference to Chinese medicine and denotes a pathogen.

53
deposit when she moved out. Some of the places we visited in 2016 had rules in an intimidating
level of detail on what was and, more importantly, what was not allowed plastered all over walls
and doors. Rules included policies on when to shower, what time to turn off the lights, and when
and where to take phone calls. During one of the site visits, the man showing us the group rental
woke up a girl who was by all appearances fast asleep by shaking her and yelling “I am still waiting
on this month’s rent! Don’t you think I forgot!”
The group rental where I rented a bed was nothing like that. Although I learned that there
was a constant coming and going among the tenants – in my room, all but one woman had moved
in within the last three months – my fellow renters had cultivated a friendly atmosphere.
Our landlord, a young guy who I estimated to have been in his mid-twenties, contributed
to this. When he agreed to let me rent a bed, he said he would make sure to prepare the others that
I was coming. He managed our tenant WeChat group and could be called anytime someone was
locked out or otherwise needed help. He also came on alternate days to sweep and take out the
trash. When he came to clean on my second day, he asked me how I slept and how everything was
working out for me and his concern seemed genuine.
But it was really how the women treated each other that made all the difference for living
with 21 strangers. Even though they had only known each other for a few weeks, the women
seemed at ease with each other. They helped each other making sense of their new experiences,
for instance by talking about Shanghai colloquialisms and comparing how things would be phrased
or called back where they each came from. They gave each other advice on romantic relationships
and clothes, shared stories about their workdays and their families. They teased Meifeng about

54
being an only child
37
and made fun of Ayi who snored like a chainsaw. On several occasions, I
overheard the only two middle-aged women fondly referring to the other women as small children
( xiao haizi).
38

No matter the general atmosphere in the group rentals, whether cold and anonymous or
warm and convivial, anywhere the use of smart phones was ubiquitous. Both in the places I visited
and where I stayed, people were constantly on their phones: they watched TV, played online games,
talked to their parents, exchanged messages and pictures with friends and family, or scrolled
through social media. In her ethnography of young migrant women working in Beijing, Wallis
(2013) describes how mobile phones were often the first thing they bought. For the young migrants,
mobile phones enabled what she calls “immobile mobility” (Wallis, 2013: 6f.). The phones were
a means to overcome spatial, temporal, and physical boundaries; even when physical mobility was
restricted.
39
The women in my group rental had limited time and money to spend on recreational
activities. They lived far from their families and rarely got to see them. After work, they typically
just came back to the apartment. Keeping in touch with friends and family via the phone was
comfort. Focusing on the phone to read, watch TV, or play games was a way to shut out the
constant stimuli of their busy environments, a way to cope and mentally escape their bodies and
surroundings.

37
While the One-child Policy was strictly, and at times violently enforced in urban areas, enforcement was more lax
in the country side. In particular, if the first child was a girl, a second child would be allowed after five years. Residents
of remote areas and ethnic minorities were granted further concessions (Hesketh & Zhu, 1997).
38
I, too, was often on the receiving end of these small acts of kindness. For instance, Ayi insisted on “cooking” me
instant noodles on several occasions since she was worried that she never saw me eat anything while I was there. Or
when I left the group rental, Meifeng helped me return my key so that I could my deposit back from the landlord,
which I did.
39
Since Wallis’ (2013) research, smart phones have become virtually universal in urban China and even indispensable
as most payments and even ID controls are done electronically. Mobile internet penetration is pervasive and widely
affordable in China’s cities (CNNIC, 2018).

55
All in all, group renting required flexibility and tolerance in the face of challenging
conditions. In order to share very little space with a lot of people, group renters came up with a
number of adaptive strategies to manage everyday life. Importantly, figuring out daily routines in
tight quarters also included navigating inter-personal relations with strangers. Numerous
conversations with group renters conveyed that nobody really liked living there. The group renters
I met were getting by and making do.
But to what end? What were they doing this for? My housing market analysis showed that
bed space renters traded off personal space to gain locational access at affordable cost. But what
was driving this calculus? The following chapter explores the motivation behind these choices.
3.4. Group Renting as Time-Space Strategy
By asking what group rentals allowed their tenants to do, this section delves deeper into group
renters’ motivations in order to fully unpack the housing need giving rise to the phenomenon.
Beyond shining light on what opportunities group renting afforded, this section also discusses the
socio-cultural forces at work in producing a housing need not met in the formal market.
3.4.1 Making Saving Possible
Money was a concern for anyone I met and spoke with, but the reasons why differed. With Yiran,
the personal trainer, money played a prominent role in our interaction from the beginning. I got to
know Yiran through a friend, who had been taking personal boxing classes with him in order to
lose her pregnancy weight. The high-end gym Yiran worked at was on the underground level of a
newly opened luxury mall in Shanghai’s quickly developing Southwestern edge – the new heart
of a growing neighborhood which, according to my friend, mostly consisted of young professionals
and young middle class families like her own.

56
To talk to Yiran, I had to pay the fee of a private training lesson. My friend negotiated this
on his behalf, because she didn’t want him to get in trouble with his boss. We met at a Starbucks
on a different floor of the same luxury mall so that he could get back to work quickly once we
were done. My friend stayed with us through the interview. By then she had seen Yiran four times
a week for the last five months and she thought her staying might help him feel more comfortable.
Yiran, 27 years old when I first met him, had come to Shanghai about 15 months prior to
when we met, right after Chinese New Year of 2017. He had worked in various personal trainer
jobs before in his hometown of Dongguan, an industrial city of roughly 8 million inhabitants
between Guangzhou and Shenzhen. He now lived in his second group rental after his employer,
who organized accommodation for most of their staff, had relocated them to a slightly bigger
apartment that he now shared with 11 other coworkers. He worked every day from morning until
midnight, including weekends, and only got 2 days off per months.
Of the approximately 20,000 RMB
40
he made every month, he paid 500 RMB in rent and
aimed to save about half of his salary. It felt like a confession when he told me that actually, most
months the savings ended up being closer to 5,000 RMB. The main reason Yiran lived in a group
rental and rarely made his personal savings goal was that every month he sent several thousands
of RMB to his parents, who were taking care of both his grandparents; more if there were medical
bills. He told me: “This [the group rental] is the best way to keep my expenses low. And besides,
I am single, I work all the time, what space do I need? Sure it would be nice if it was cleaner but
it’s fine. Plus I don’t know anyone in Shanghai. Why would I live alone?”

40
This is the equivalent of 3,077 USD at the time, a comparatively high salary. For reference, my interviewees with
Master’s degrees from elite universities made between 10,000 and 30,000 RMB/ month in entry-level positions.

57
When I met Ninghong, another group renter in her 20s, she was sharing her bedroom with
two people and the whole three bedroom apartment with a total of 9 others. She, too, shared living
space to keep expenses low in order to send money back to her family. Despite the surface-level
similarities in living situation and motivation, her story was quite different.
While Yiran started working full time as soon as he finished high school, Ninghong had
attended and received Bachelor and Master degrees in Geophysics from Tongji University ( tongji daxue), one of the best universities in the country – a huge achievement by any account,
but especially given that she was from a small town in rural Henan.
To this day young people from poorer and especially rural upbringings still face
tremendous obstacles when they want to pursue higher education. Regional, and in particular rural-
urban disparities in the distribution of incomes and educational resources are systematically
disadvantaging rural students (Ding, 2002; Hannum & Wang 2006; Li & Sicular, 2014; Zhang &
Kaubur 2009; Zhou, Moen & Tuma, 1998). The hukou system practically prescribes
41
where one
receives pre-college education. Due to higher public and private investment in education in urban
areas, those with urban hukou tend to have better access to higher quality education and with that
are generally better prepared for the national university entrance exam ( gaokao) (Choi, 2016;
Fu & Ren, 2010; Gao & Smyth, 2015; Montgomery, 2012). Conversely, rural students’ limited
access to quality education adversely affects their chances of getting into a top college. On average,

41
Changing hukou registration is extremely difficult and rare (Chan, 2013; Solinger, 1999). More recently municipal
governments have been urged to support the access to education for migrant children and some cities have taken action
(Chen & Feng, 2013). If migrant parents can afford it, they will send their children to one of the growing number of
private schools, catering to (upper) middle class families, regardless of hukou.

58
they score lower on the extremely competitive exam (Hannum, Wang & Adams, 2008; Yang,
Huang & Liu, 2014; Zhang, Li & Xue, 2015).
42

As a result, rural students are disproportionally funneled into colleges of lesser quality
43

(Li et al., 2012; Ling, 2015; Postiglione, 2015). Despite the increase in access to higher education
for rural students
44
, there is overwhelming evidence that social background matters for college
enrollment (Chan & Ngok, 2011; Hu & Vargas, 2015; Luo, Guo & Shi, 2018; Qin et al., 2018;
Yeung, 2013). In fact, two studies found that rural students were the minority (17-18%) among
incoming students at the country’s two top universities, Peking University and Tsinghua
University (Huang, Xin, & Hou, 2014; Mok & Wu, 2016). By contrast, the vast majority of
students at non-elite institutions are rural and often the first in their families to pursue post-
secondary education (Chan & Ngok, 2011).
As Ninghong was getting ready to graduate from Tongji university, she had been promised
a spot as a staff researcher in her advisor’s lab, but the arrangement fell through at the last minute.
When she told me about the frantic job search that followed, disappointment and bitterness still
sounded fresh three months later. By the time she had learned the research job wouldn’t work out

42
Note that most top schools are in cities and favor local students in admissions (Hu & Vargas, 2015; Li et al., 2012).
Wealthy urban families have been known to buy secondary property – known as “school district apartments” (
xuequ fang) – solely for the purpose of getting access to elite middle- and high-schools (e.g., Wu et al., 2016).
43
China’s postsecondary education institutions can be separated into four-year universities ( benke) and three-
year specialized colleges ( zhuanke), leading to a bachelor’s degrees and diplomas. The specialized colleges, or
polytechnicals ( xueyuan), can further be distinguished into upper-level specialized colleges ( dazhuan) and
vocational-technical colleges ( gaozhi), most of which are either fully private or public-private partnerships
(Postiglione, 2015). Additionally four-year universities can be further distinguished into elite and regular institutions.
In 1996, the government designated around 100 universities key national institutes of higher education, known as
“Project 211” (211 eryiyi gongcheng). Later, “Project 985” (985 jiubawu gongcheng) further selected a
group of initially 9 schools of distinction, akin to the ivy league in the United States (Xinhua, 2017a). Importantly,
211 schools receive 70% of all government research funding (People’s Daily Online, 2008).
44
In 2014, the share of rural college students had grown to 59 per cent (Ma & Yang, 2015).

59
she was late to the job market for entry-level positions
45
and ended up taking a position as a
business analyst in a company that produced computer parts. Ninghong, who was 24 when we met,
had felt a strong pressure from her family to “not waste any more time” and firmly locked her
dream of pursuing a Ph.D. away. “Maybe if I work for a while and earn money, I can show that
my education has been worth it and that I am grateful for my family’s support all these years.
Maybe, after I find a husband,
46
I can return to school.”
In her current position, she earned a net income of 8,000 RMB/ month out of which she
spent 1,200 RMB on rent and at least 1,500 RMB on food. With roughly 50 RMB or 7.50 USD
per day, her budgeting was on the frugal side. Every months she tried to save 3,000 RMB for
herself and sent the same amount back to her family. To be able to do that she tutored high-school
kids for several hours each Saturday, in addition to her full-time job.
3.4.2 Gaining a Foothold
Both Yiran and Ninghong talked about plans for “after,” plans for life after group renting; They
assumed that their current situation was temporary. In fact, all group renters intended for group
renting to be a transitory means to an end only. A couple of the people I talked to were renting
beds just while they were in town for summer internships or language classes, or while they studied
for a certification exam. But for most, group renting functioned as a bridge, a deliberate stepping
stone in a longer-term strategy.

45
It is common knowledge that in China both labor and housing markets are marked by strong seasonality. Known as
“ ” (jin san yin si, literally meaning “golden three silver four”), March ( san yue) and April ( si yue)
are peak recruitment months because they follow on the heels of Chinese New Year when bonus and taxes have been
paid. The second “ ” (gao feng qi, peak period) is around June, July, and August.
46
In China, women with high levels of education often are at the receiving end of societal disapproval for their choices.
A common saying goes: “There are three kinds of people in the world: Men, Women, and Women with Ph.Ds.”

60
For Longwei, for instance, renting a bed allowed him to eventually land on his feet after
his small business had failed. Originally from a small town in Jiangsu, he had graduated from
Zhejiang Communication College ( zhejiang chaunmie xueyuan) in 2015 with a
degree in in Music production and composition. Afterwards, he worked as a sound manager and
production assistant before investing his entire savings to open up a gym in Guangzhou with a
friend from college. When he lost everything, Shanghai was a fresh start. He got a job as a real
estate marketing agent for a firm specializing in marketing Australian investment property to rich
private investors in China. He came to Shanghai with nothing more than a suitcase. The group
rental bought him time and space until he eventually gained a foothold in the new city and his new
career.
Those who had recently graduated from college or university were using group renting as
part of a strategy to master the transition into stable employment. The majority of freshly minted
graduates started their jobs with trial periods ( shiyong qi), which included reduced pay –
reductions as high as 30% from what for many were already low starting salaries to begin with.
Trial periods typically lasted one to six months and also preempted job market entrants from
getting bonuses. Data I collected through a cell phone social media survey in 2019 (Chapter 4)
showed that 92% of respondents went through a trial period in their first job after graduation.
Those who did not start their employment with a trial period had done internships with their
employers while they were still students.
What’s more, in the tight rental markets of China’s biggest cities, a common policy charges
new renters “ ” (ya yi fu san), meaning that to move in one must pay one month’s rent as
deposit, as well as the first three months of rent up front. As migrants, group renters could not fall

61
back on living with their parents, but renting in the formal market required a significant advance
payment. With very low incomes and few options, the Shanghai newcomers hence saw group
renting as a temporary fix, a way to buy time while they saved up for a deposit, waited until they
got into a better position financially to be able to upgrade to better accommodation that would also
costs more, or both.
Lihua, for instance, had just graduated from a Master’s degree in hospitality management
from Fu Jen Catholic University ( taiwan furen daxue) when we met. Although the
school is a top private university in Taiwan, it is lesser known in mainland China. Lihua attended
the university on a government program that sponsored mainland Chinese students’ university
attendance in Taiwan. Originally from a mid-sized city in Jiangsu province, she had been raised
by a single mother after her parents got divorced.
After graduation, she had taken a job as a trainee at an international high-end hotel and
restaurant chain. When we talked she was clearly frustrated about what her first job could afford
her in terms of accommodation: “I have lived in a dormitory for the last seven years! I thought
finally, when I start working, finally I will be able to afford something better. But I had to realize,
in Shanghai that’s not possible, not for me. If I rented something myself, I’d have to spent 90% of
my salary on rent.” Her salary during the trial period was 3,500 RMB (ca. 540 USD). When her
employer offered her a bed in a room shared with three others for 300 RMB per month, she
accepted. After the three month trial period she would be reassessed and was hoping to be placed
in the management trainee track that would come with a significant salary increase.
Another one of my interviewees recalled hearing about colleagues at her work who had
rented a bed when they just started out at the company: “At the time they could have afforded

62
more, but not a whole apartment by themselves, they told me. So they were stuck, they had no
choice but to stay in a group rental and save up to move out.”
As the name suggests, trial periods are used to “test out” job market entrants and inform
decisions on whether or not they would be offered a permanent position. China has a long history
of “face time” office culture and “volunteering” to work overtime is often seen as showing
commitment and a “fighting spirit” ( pinbo jingsheng).
47
My roommate Meifeng, for
instance, chose the group rental we stayed at mainly because it was what she could afford within
walking distance of her office.
48
Meifeng told me she chose the place quickly, within one day,
because she wanted to be able to “always be on time, work hard, show my talent to the boss.”
49

One of the brokers who lived in a group rental with his colleagues told me: “Price is most important,
but after that it is location. They…we
50
can’t afford to ride the subway for half an hour, one hour
each way to work. The work day is already so long! We need some time to rest, to sleep.”
For Haotian, both the job and the group rental were temporary. In 2010, he graduated – as
the first in his family – with a degree in electrical engineering from what is now called Wuhan
University Luojia College ( wuhan daxue luojia xueyuan).
51
Because he was
unable to find a job in his field, he ended up taking a six-months contract with a nuclear power

47
Recently protests against the so called “996 work day system” (996 gongzuo zhi), which requires employees
to work from 9 am to 9 pm, 6 days per week, have made headlines in China (e.g. People’s Daily, 2019).
48
Our group rental was a 10 minute bike ride away from Xujiahui, a subcenter in the Southwest of inner city Shanghai.
Recent work on migrants’ job–housing balance suggests a preference for short walking commutes among low-income
workers in China (Harten et al., 2020; Kim, 2016; Li & Liu, 2016).
49
She came home one night, bewildered, defeated, and told us: “Do you know what happened to me today? I was less
than 10 minutes late and the boss deducted 100 yuan from my salary. Can you believe it? 100 yuan! And being late
wasn’t even my fault. How unfair is that.”
50
He kept correcting himself from time to time to “we” during our conversation.
51
The college, formerly known as Wuhan Qingchuan College ( ), is private and was created through a
joint venture between several private companies and Wuhan University in 2006. Wuhan University is an elite
university and the name change most likely an attempt to carry over some of this prestige to the second tier institution.

63
company in Dalian as a stopgap solution. Working on the electricity system construction he earned
a monthly salary of 3,000 RMB (ca. 450 USD) and shared a room with five others in employer-
provided accommodation. By the end of the short-term contract, he had found a permanent position
at an IT company in Beijing.
Over the course of my fieldwork I also spoke to dozens of group rental landlords and
brokers. They, too, were very clear on what they offered: flexible, short-term, affordable
accommodation, close to where their mostly young and white collar tenants worked. They had
watched hundreds of tenants pass through their apartments and knew they filled in for a missing
market for young people in transition. Because group renters were almost exclusively job market
entrants, to some degree they also profited off of the inexperience of the newly arrived. As Hong,
a former real estate broker who lived in a group rental for his first three years in Shanghai, told
me: “This market caters to beginners in the city. They only stay until they learn the big city life.”
My informants stories carve out the key function of group rental housing: temporary, cheap
housing in the big city; one part in a city-starter strategy that enables group renters to make and
save money and/ or buys them time while they work on setting up better situations for themselves.
But, not all the young people I met rented beds. I also spoke with several other recent graduates
who rented a room by themselves or shared it with only one other person, typically a coworker or
friend from college. So why were group renters struggling while their peers transitioned into life
off-campus and employment much more smoothly? Are there structural forces, social institutions
that create these disparate experiences?

64
3.4.3 The Odds Stacked Against Them: Hukou, Education Stratification, and Class
The group renters I met were part of what the government calls the “new generation migrants” ( xin sheng dai nongmingong): born after the reforms of 1978, more educated, and
more likely to concentrate in China’s largest cities (Qin, Wang & Lu, 2018; Xinhua, 2018a; Zhao,
Liu & Zhang, 2018).
Born in the 1980s and 1990s,
52
they grew up in a China marked by extraordinary change.
Starting from 1978, a series of reform policies, commonly summarized under the “Reform and
Opening” umbrella, kicked off the country’s economic and political transition and unleashed
decades of rapid economic growth. After years of central planning, a change in leadership within
the Chinese Communist Party (CCP) initiated the gradual implementation of market reforms in an
effort to salvage the crippling domestic economy.
The first stage of reforms in the late 1970s and early 1980s started with de-collectivizing
agricultural production. The successful transition to a rural market economy set the precedent for
introducing market forces more broadly (Brandt & Rawski, 2008; Mühlhahn, 2019; Unger, 2002).
Private businesses were legalized and the country gradually opened up to international trade and
foreign investment through Special Economic Zones (Chen, Jefferson & Zhang, 2011; Ge, 1999).
In a second wave of reforms, the CCP started lifting price controls, protectionist policies, and
regulations in the mid 1980s. Pursuing a “dual-track” model, most state-owned enterprises were
also privatized, while the government maintained monopoly power over key industries. Joining

52
China calls them “80 , 90 ” (ba ling hou, jiu ling hou), “ ” meaning after.

65
the World Trade Organization in 2001 marked China’s full integration into the world economy
(He et al., 2016; Naughton, 2006).
The parallel processes of industrialization, marketization, decentralization, and
globalization gave rise to unprecedented economic growth (He & Zhu, 2007; Wei, 2000). At the
beginning of the reforms, China was one of the poorest countries in the world. Since then, more
than 500 million people have been lifted out of poverty and the country’s GDP has grown at an
average rate of 10 percent a year for three consecutive decades (World Bank & State Council,
2014; Zhu, 2012). Within the time-span of a single generation, China has developed from an
impoverished agrarian society into the world’s second largest economy (Mühlhahn, 2019).
Migration and urbanization lie at the heart of these successes. While pre-reform mobility
restrictions had essentially prohibited any migration, when constraints were lifted in 1985, millions
of rural peasants started moving to the cities. Years of rural and urban underemployment in the
centrally planned economy had generated an extreme labor surplus (Chan, 1996; 2010; Davin,
1998). As the country resumed large-scale infrastructure investment and construction, and shifted
out of agriculture and into urban production this labor force was in high demand (Chan, 2013; He
et al., 2016). In 2017, the total number of migrant workers reached 286 million people (Xinhua,
2018a).
The early years after the reforms, in particular, were a period of wealth creation where the
benefits of economic growth were shared broadly among the Chinese people (Biao & Shen, 2009;
Bian, 2002). When private businesses were first legalized in the early 1980s, even families with
limited resources could start a business and, with some luck, not only make a living but become
wealthy. Furthermore, the wages of millions of rural farmers-turned-urban migrant workers helped

66
fuel development in their hometowns. Social mobility was real and tangible (Bian & Logan, 1996;
Nee, 1996). As Li (2010) puts it: “[n]ever in history have so many people made so much economic
progress in one or two generations” (Li 2010: 3).
However, the distribution of benefits has changed. By the late 1990s, as taxation, regulation,
and competition increased, most small businesses were barely providing a living (Fong, 2004).
Additionally, China’s export-driven economy was hit hard by the Asian financial crisis of 1997.
Restructuring and layoffs affected workers the most and China entered a period of wealth
concentration, which also saw the re-emergence of clear class relations (Biao & Shen 2009). While
Deng Xiaoping’s “let some people get rich first” ( rang yi bufen ren xian fu
qilai) had been acceptable as long as the ripple effects were still felt widely, people started to feel
the competition for a piece of the pie intensifying. For parents of young children during that time,
these developments meant focused efforts on giving them the best possible chances to succeed
(Fong, 2002; 2004).
Besides “Reform and Opening”, the other 1978 reform with far-reaching consequences
was the One-child policy ( dusheng zinv zhengce). Mandating that each couple only
have one child, the policy induced an abrupt demographic transition and drastically altered the
structure of the Chinese family (Cheung & Yeung, 2015; Deutsch, 2006; Hesketh, Li & Zhu, 2005).
Large families, with little to invest in each child, were quickly replaced by small families that
could now invest much more, and in one child only. Those born in the 1980s, and especially those
born in the 1990s, grew up with much more resources available to them than any generation before
them; but the pressure to become high-achievers also grew at the same pace (Fong, 2002; 2004).

67
After years of government investment in flattening class structures, the race to the top was
getting crowded and education emerged as a promising vehicle for social mobility. For decades,
Maoist China had pursued a politics of anti-intellectualism, which came to a head with the
atrocities of the Cultural Revolution ( wenhua da geming, 1966-1976). As the
economy was restructuring, the government now advertised a modernized China, led to the top of
the global economy by a new generation of highly skilled workers. Consequently, the government
switched course and issued a sequence of major education policy reforms to rapidly revive and
expand the atrophied system of higher education (Bai, 2006; Meng et al., 2013). In a society
imprinted with Confucian ideals about education
53
the push for mass education found fertile soil.
Education reforms started in 1977 with the restoration of the central university entrance
exam, (gaokao). The aim of the reforms was to restructure education opportunities away
from political affiliation and towards meritocracy while expanding access (Gao & Smyth, 2015;
Wan, 2006). It wasn’t until 1999, however, that the government drastically increased the target
post-secondary enrollment rate and adopted a course of rapid growth in higher education
institutions and admitted students (Cebolla-Boado, Hu & Soysal, 2018; Mok & Qian, 2018).
54
By
2016, higher education enrollment had reached 42.7 per cent – up from 0.26 per cent in 1949,
when the People's Republic of China was founded, and 1.55 per cent in 1978, at the begin of the
reforms (Sun, 2017). In 2019, the number of freshly minted college graduates surpassed the 8
million mark (Xinhua, 2019a).

53
A famous Chinese saying is “ ” (wanban jie xiapin, weiyou dushu gao), translating to
“all things are inferior, only education is valuable.”
54
In 1998, there were 1,984 higher education institutions, with a total enrolment of over 3.4 million students. By 2017,
these numbers had risen to 2,631 and over 27.5 million, respectively (Ministry of Education, 2017a; 2017b; Yang,
2005).

68
The drastic upsurge in college graduates suddenly flooding the labor market did not fail to
leave its mark. When the first education reforms rolled out, the job market for college graduates
had been strong. As the national economy changed to a more complex system of production, the
demand for high-skilled labor was constantly rising as well. The recent levelling off of wages for
those with tertiary education as well as reported increases in graduate unemployment indicate that
perhaps supply has now caught up with or even outstripped demand (Appleton, Sing & Xia, 2014;
He & Mai, 2015; Qin et al., 2018).
Another contributing factor: the quick expansion had also come at the expenses of quality
and has led to significant horizontal stratification within higher education. Some of the newly
established colleges turned into diploma mills, yielding poorly qualified students who – on paper
– still had a university education (He & Mai, 2015; Mok, 2009). This “credential inflation” created
uncertainty about the value of a post-secondary degree (Bai, 2006; Hu & Hibel, 2014; Knight,
Deng & Li, 2017).
The labor market responded with generally low starting salaries and trial/probation periods.
Only elite universities, which are extremely selective, retained strong signaling power for their
graduates (Hu & Vargas, 2015; Li et al., 2012). Compared to non-elite institution degrees, elite-
college degrees were fetching a wage premium of up to 28 per cent (Hu & Hibel, 2015; Hu &
Vargas, 2015; Li et al., 2012; Zhong 2011). According to the China Wage Development Report
(2013–2014) issued by the Institute for Labor and Wage Studies under the Ministry of Human
Resource and Social Security the average monthly starting salary of all higher education graduates
in 2012 was 3,039 RMB; But the average monthly starting salary for graduates from non-elite
colleges was only 2,323 RMB, much closer to the average for graduates with secondary education
only (2,274 RMB per month) (ILO, 2016).

69
The difference was noticeable with the people I met and interviewed. Those with degrees
from elite universities consistently reported starting salaries of at least 5,000 RMB per month; even
over 8,000 RMB if they graduated with a Master’s degree and/ or with a degree in the sciences or
business. By contrast, the graduates from second tier colleges, such as Longwei, Meifeng, and
Haotian, all started out making less than 4,000 RMB
55
per month.
Notably, the wage gap has been widening together with the employment opportunity gap
(Mok & Wu, 2016). While overall graduate unemployment has increased significantly since 2003,
the share or rural graduates and graduates from non-elite institutions has been estimated to be as
high as 30 to 40 percent at graduation (Cebolla-Boado et al. 2018; Hu & Vargas, 2015; Knight et
al., 2017; Mok & Qian, 2018; Mok & Wu, 2016).
56
Given competitive labor markets, taking low
paying jobs to add work experience to their resumes, as Lihua and Haotian did, can be a strategic
choice.
Until most recently, Shanghai has been among the top employment destinations for recent
graduates. Recent graduates moved to or stayed in “tier 1 cities,”
57
because the jobs they were
seeking concentrated in the country’s biggest and economically strongest cities (Cui, Geertman &
Hooimeijer, 2014; Zhaopin, 2014; 2015; 2017). For instance, large foreign companies, which are
popular employers as they pay higher salaries on average, disproportionally locate in tier 1 cities

55
Note the discrepancies in absolute values in comparison to the data in ILO (2016) are in part due to the time lag and
in part due to the difference between country-wide averages versus regional outliers. Shanghai is among the top three
with regards to the highest average wages in China (Yu, 2019).
56
Rural college graduates have also been reported to be more likely to work in temporary jobs and jobs they are
overqualified for (Chan, 2015; He & Mai, 2015).
57
Beijing, Shanghai, Guangzhou, and Shenzhen are tier 1 cities ( yixian chengshi). They all have a GDP
surpassing $US300 billion. Beijing and Shanghai are also directly controlled by the central government and have
populations over 20 million. A group of fast growing large and mid-sized cities are now called the “new tier 1 cities
( xin yixian chengshi). Most recently they have overtaken the group of tier 1 cities as the number one
destination for recent college graduates (Zhaopin, 2017).

70
(Yu, 2019). 2010 census data shows that the top 1% of cities received 45.5% of all migrants. Out
of these, Shanghai, with a migrant population of almost 11 million, or 40% of its residents, was
the largest recipient (Liu et al., 2015b; Qin et al., 2018; Xing & Zhang, 2017).
These same cities, however, have for years now pursued modernization and “global city”
policies that come mostly at the expense of migrants (e.g., Timberlake et al., 2014). In particular,
Beijing and Shanghai implemented population caps in 2017 to thwart so called “big city disease”
( da chengshi bing) (Roxburgh, 2018). Part of this plan involves what Sun (2018) coined
the “3D of de-urbanization”: (1) dilution, i.e., the relocation to suburbs, (2) destruction, i.e., the
cracking down on illegal housing, and (3) distinction, i.e. using the housing market to “filter for
talents”, the idea being that high housing costs will keep those with low earning potentials out.
In Shanghai, this “filtering” logic is propped up by persistent unaffordability on the one
hand (Chen, Hao & Stephens, 2010; Mostafa, Won & Hui, 2006), and a reformed hukou and
residence permit system on the other hand. The new point-based system favors so called “talents”
( rencai) and wealthy migrants while effectively closing the door to everyone else (Li, Li &
Chen, 2010; Zhang, Wang & Lu, 2019). As one of my friends cynically put it: “with this new
system you can know exactly how valuable you are to this city.”
While I also met a number of laborers and elite-graduates, most of the group renters I met
in Shanghai were what I have come to think of as “second class college graduates” – not because
of their skills and knowledge, but because of the way the labor and housing markets treated them.
They came to Shanghai with degrees from smaller universities and specialized colleges, hoping to
make good on the promise of education as a social lever. Often from rural or small town
backgrounds, they have gone much farther than their parents could have dreamed of when they

71
were growing up. Most of them were the first in their families to attain higher education. With
them, they carried the hopes and expectations of families who trusted that by investing in their
children’s education, they would pave the way for them to earn money with their brains instead of
their hands. After arriving in Shanghai, however, these job market entrants were confronted with
the harsh reality of high housing costs, long work hours, and low salaries – made even harsher by
unwelcoming government policies. The odds stacked against them, still they found ways to stay.
3.4.4 A Seemingly Successful City-Starter Strategy
But did their persistence pay off? Did those intending for group rentals to be an incubator of sorts,
buying them space and time until they got a foothold in the city, actually ended up in better places
eventually, both in terms of housing and employment?
Career wise, the answer anecdotally seems to be yes. When I met up with my friend who
had been taking private boxing classes with Yiran one year later in 2019, she told me that he had
recently left the gym she trained at to open up his own studio. Together, we scrolled through his
WeChat feed and found it brimming with enthusiastic pictures and short videos documenting the
renovation and opening of his gym, as well as a few first classes that appeared to be packed. From
checking in with him, my friend knew he had moved out of the group rental and was now sharing
a room at the back of his gym together with his business partner.
I also met up with Longwei again. One year later, he recounted how he moved out of the
group rental after three months and went on to share a room with a coworker from his real estate
marketing job. After staying in this job and this apartment for eight and six months, respectively,
he started a new job at a small importer of Korean cosmetics, where he was now in charge of their
online marketing presence. When he told me about his new job and his responsibilities, he was

72
extremely excited. He had a Korean girlfriend and thus learned some Korean. Because of his
language skills, he now often played a key role in bringing together Korean and Chinese clients.
When we met, the company’s online retail platform had just been launched – all under his directive
– and he was proudly showing off his work. He now rented his own room, close to his new office
in an inner Shanghai suburb.
Lihua, the hotel and restaurant trainee, too, seemed to have landed in a stable career. After
her trial period she did get placed in the management trainee track but eventually changed jobs to
work in the marketing department of a food and beverage company. When we talked, she said she
was grateful for her first job because the experience helped her land her current job – where she
was much happier. She now shared a room with a friend from college.
My interviewees’ subsequent housing choices revealed how, after group renting, they were
now in positions enabling them to make different tradeoffs. For most, that meant buying a little
more privacy and comfort with a lot more time spent commuting. Meifeng and Jing, for instance,
moved out shortly after I left the group rental we stayed at. Together they moved to an apartment
a 60 minute transit ride away from our original place. They now shared a three bedroom apartment
with two bathrooms between 12 people. In this new place, they shared the room between only four
instead of eight women previously. When I met up with Meifeng in 2019 she told me she didn’t
mind the commute at all. She could usually nap on her way back and forth from work because of
her off peak hour work schedule. She told me her body had even gotten so used to the trip that she
woke up automatically when it was her time to get off. Lihua, too, told me she had been more than
happy to trade in her central city dormitory for the shared room in the inner suburbs.

73
The road to a room of their own, could be long, however. For Haotian, the electrical
engineer, for example, it took him five years, one more group rental, two more rooms shared with
co-workers, and accepting a 50 minute commute until he was finally able to rent a room just by
himself.
3.5. Mobility Strategies Are Multi-Generational
The previous section detailed what group rentals allowed their tenants to do: Yiran kept his
expenses low to save up for opening his own business, Longwei rented a bed while he regrouped
and started a new career, Lihua and Haotian lived in group rentals while they gained low-paying,
but valuable work experience that eventually landed them stable employment in their fields, and
Meifeng and Jing rented beds close to work until they gained enough job security so that they
could move to better, but less accessible living arrangements and commute.
Section 3.4 revealed the key functioning of group rentals as a time-space strategy for
migrants who needed to save expenses to send money back to their families or until they could
reach a more secure foothold in a highly competitive environment. But why was “making it” in
the big city such a high-stakes goal? To what end did the group renters struggle?
3.5.1 The Weight of Filial Piety in an Aging Society
Except for Ninghong and Yiran, the need to support their families financially at first glance wasn’t
obvious as a significant factor in these young people’s decisions. Yet, the more I talked to them
and dug deeper on the question “What is it all for?”, the more I started to grasp an unspoken, and
at other times rather explicit, intergenerational contract between the young people I talked to and
their aging family members.

74
In China, the number of people aged 60 and above reached 249 million at the end of 2018.
With a senior population share of 18 per cent, China is now the country with the largest and fastest-
growing aged population in the world (Xinhua, 2019b). China had first crossed the 10 per cent
mark in 2000, making it an aging society by internationally recognized standards (Liu & Sun,
2016). By 2050, it is expected that over one third of the population will be of age 60 or older
(Banister, Bloom, & Rosenberg, 2012).
The root causes of population aging are the same as elsewhere: low fertility rates, greater
longevity, and the cumulative effect of past changes in birth and death rates (Banister et al., 2012;
Zhang, Guo & Zheng, 2012). Both the sharp decline in fertility and the sharp increase in life
expectancy are the direct product of government policy. The One-child policy dramatically
reduced average family size (Cheung & Yeung, 2015; Hesketh et al., 2005), while economic
transition and urbanization, not only produced rapid economic growth, but also lifted hundreds of
millions of people out of poverty, thus significantly improving health outcomes (Cook, 2002;
Ravallion, Chen & Sangraula, 2007; World Bank and State Council, 2014).
The evolving age structure of Chinese society towards an increasingly top-heavy
population pyramid suggests that the age dependence ratio, that is the share of children and elderly
people versus the working-age population, will continue to rise (Liu & Sun, 2016; Zhang et al.,
2012). In 2018, for the first time, the number of people aged 15 and under was surpassed by those
who aged 60 and above (Xinhua, 2019b). With a shrinking labor force on the not so distant horizon,
how will China take care of its senior citizens?
China, for a long time, has been able to count on kin support in lieu of a government-
sponsored system of social support. Familism is deeply rooted in Chinese culture and Confucian

75
ideals such as “filial piety” ( xiao) continue to provide a strong normative imperative for
intergenerational ties (Fairbank & Goldman, 2006; Guo, Aranda & Silverstein, 2009; Zhu, 2016).
But relying on the familial support alone is increasingly growing unfeasible. For one, China
has one of the earliest retirement ages in the world: most of the formally employed stop working
between the ages of 50 and 60 (Feng et al., 2019; Zhang et al., 2012). What’s more, the compressed
demographic transition, brought about by the One-child Policy, has majorly unsettled the balance
of traditional intergenerational reciprocity. At the individual family level, the inverted pyramid
translates into the “4-2-1 syndrome.” Originally, the term described the experience of the first
generation of couples born after the One-child Policy. Singletons themselves, upon starting their
own family they would be responsible for the care of one child and four parents.
58
Nowadays, the
“4-2-1 syndrome” describes how in the contemporary Chinese nuclear family, one child shoulders
the investment, hopes, and expectations of four grandparents and two parents (Hesketh et al. 2005;
Fong, 2004).
Recognizing trouble in the making, the government has responded, but the actions taken
are still insufficient when measured against the magnitude of the problem. Starting in the early
2000s, state campaigns have intensified efforts to increase the number and capacity of nursing
homes (Chu & Chi, 2008). So far, old age care institutions have found only cautious approval
among older people and their families, but nursing home placements of older adults are expected
to increase substantially within the next ten years (Chu & Chi, 2008). The more difficult task,
however, has been the reform of the pension system – a major undertaking aimed at universal
coverage. Until 2005, China only had an earning-related urban pension program, which meant that

58
Also known as “ ” (shang you lao, xia lou xiao), translating to “above there’s the old people, below
there’s the young people.”

76
almost all of the rural population and all non-continuous and part-time urban workers were
excluded. Gradually, the government came up with pension arrangements for these population, but
coverage remained highly fragmented (Cai & Cheng, 2014; Feng, He & Sato, 2011; Liu & Sun,
2016). In 2014, a new universal pension plan was launched that now combines all previous system
under one umbrella. But with a country average of 81 yuan (13 USD) base pension per month, the
benefit level is devastatingly low (ILO, 2016; Liu & Sun, 2016). Yet, given the shrinking
workforce, recent studies are raising alarms that the system could run into liquidity problems as
early as 2035. Meanwhile, private pension schemes are still in their early stages and especially for
rural residents more complicated medical care often requires expensive trips to the next biggest
city and must be paid out of pocket (Hesketh et al., 2005; Li, Xiao & Xiao, 2009; Strauss et al.,
2012; Tang, 2019).
Hence, children have been and will be –for the foreseeable future– their parents’ safety net
for old age. From this vantage point, investing in children’s education becomes about more than
giving them every opportunity to succeed in an increasingly competitive race to the top; it becomes
about ensuring the entire family has a long term survival strategy. Lin (2019) calls parental
investment in children’s education – especially in families of limited means – “purchasing hope”
(Lin, 2019: 1). These families were “purchasing hope” not only for their children, but for the entire
family.
Because of the “4-2-1 Syndrome”, today’s only children are under immense pressure to
meet these great expectations. Before the One-child Policy, Chinese families traditionally were
large, and responsibility to care for the elderly could be distributed among several children. Now,
however, the system of intergenerational transfers, which functioned as a micro-level welfare
system, has narrowed its focus: not only parents, but also grandparents pool resources to invest in

77
the only offspring – in order to maximize the child’s chances at securing a high-paying job that
could support the family later in life (Fong, 2004; Zhu, Whalley & Zhao, 2014; Zhu, 2016; Ying,
2003).
To this end, parents and grandparents will often severely reduce other types of spending
and forego retirement saving (Ying, 2003). Ninghong, the young woman who set aside her dreams
of pursuing a Ph.D. to earn money so she could start supporting her rural family, had watched her
parents through years of skimping so that she could have a shot at getting into a good university.
Ninghong’s parents both were farmers originally but took jobs in a close by city to be able to put
her in a government run boarding school that prepared students for the national university entrance
exam. According to data from 2010, college expenses could be up to 4.3 times of rural families’
annual income; and this was not counting the years of expenses leading up to high school
graduation and college admission (Qin et al., 2018).
59
The intergenerational exchange flows heavy
with sacrifice, hope, and worry about the future.
Although my friend Lele’s story was set in completely different circumstances – she comes
from a comfortable middle class background in Chengdu – like Ninghong, she wanted to show her
parents that their investments in her education had been worth it. When we met up in the summer
of 2018, she had just graduated with a Master’s degree in English from Shanghai International
Studies University ( Shanghai waiguoyu daxue), one of China’s top ten foreign
studies universities. She was determined to make a career in journalism, but had mostly been

59
Recall also that students from poorer and rural backgrounds are also more likely to attend second tier colleges.
Because of the way higher education funding works in China, lower quality colleges and polytechnicals will charge
much higher tuitions to make up for the lack of government funding that elite public institutions enjoy (Mok, 2009).
What’s more, need-based scholarships and loans are mostly available for students at top universities with little
financial aid is available for everyone else (Loyalka, Son & Wei, 2012; Yang, 2010).

78
getting interviews from private tutoring schools and consulting companies specializing in helping
Chinese children and their families prepare to apply for college abroad. On our first get-together
that summer, in June 2018, she told me how she had been turning down numerous interviews and
job offers. With the eyes on the prize (a job in journalism), she had convinced one of her friends
to let her stay in an empty bed in that friend’s shared dorm room. This had allowed her to keep
living expenses low so that she could stay picky on the job front. But by July, both her friend, who
was risking trouble with the university, and her parents, who kept sending her money, were getting
impatient. After a much dreaded visit by her parents at the end of July, she finally took a job as a
sales person for a higher education consulting group in August. Still, she continued interviewing
for journalism jobs and succeeded: In September she first got and accepted a position as a staff
writer for an online newspaper in Shanghai. Later, in October, she got a prestigious position writing
for China Central Television (CCTV), one of China’s “big three” media outlets (along with the
People's Daily and Xinhua) in Beijing.
Adult children giving back to aging parents thus went beyond supporting them financially
and included making them proud and giving them “face.” In China, the concept of “face” is an
integral part of social interaction culture. While “ ” (mianzi, face) corresponds to the social
self and represents respect and reputation, “ ” (lian, face) relates more to the personal self and
has morality as a distinguishing component (Fairbanks & Goldman, 2006; Zhou & Zhang, 2017).
Both Meifeng and Yang, another recent graduate in his first job since college, talked about
landing a job in Shanghai as “bringing out [their] full potential” ( fahuiqianli) and, with
that, giving their parents “face” ( rang fumu you mianzi). When Yiran told me about
his decision to leave the South and come to Shanghai, there was obvious pride in the way he talked:

79
“All my friends back home in Dongguan, nobody left. They are still in Guangdong, (dagong,
doing manual, temporary work for hire
60
). I am the only one who left. I wanted to have different
experiences, I wanted to have opportunities. I came to make money. I saw this job on 58.com,
applied, got it, and just came here.” He was proud, not only for forging ahead, cutting out his own
path, but also for what he now could offer to his parents. In sending his father money every month,
he could “give [his] parents face” ( gei fumu zhang lian) because it helped them take
care of their parents, his grandparents who lived with them. He was certain he would return home
one day, but not before he had made “enough money.”
61

The pressure and fear to disappoint were a constant silent companion for some. Yang told
me about how just thinking of phone calls with his parents made him cringe. While he thought
they probably meant to cheer him on with their optimistic talk about his future, he just felt the
responsibility loom. Hong, the former real estate broker who lived in a group rental for his first
three years in Shanghai, told me how he had stayed much longer in the group rental than he had
planned because his family was pressuring him to keep his expenses low: “I come from a small
village, my family does not understand the prices in Shanghai.” He even skipped out on going
home for Chinese New Year his second year away from home. Chinese New Year celebrates the
beginning of the new year according to the lunar calendar and is arguably the biggest family
holiday in China. People usually return home from work or study to reunite with their families in
their native place ( laojia). It is customary to give out red envelopes ( hongbao) filled
with money. When children are small they are the recipients of red envelopes, but adult children

60
(dagongzhe), literally “those who do manual work for hire”, is how migrant workers are typically called.
61
Traditionally, the other-related “ ” (mianzi, face) and the self-oriented “ ” (lian, face) are used in relation to
other- versus self-judgement.

80
are expected to give their parents red envelopes well-filled with money, to show gratitude and
respect. That year he hadn’t been able to save up as much as he had wanted to. So he passed up on
the trip home with the excuse of work to not make his parents worry about how he was doing and
to not “lose face” ( diu mianzi).
A young Tsinghua University ( qinghua daxue) graduate student from a mid-
sized city in Henan, who I met after a conference presentation, was rather blunt about it: “It’s as
simple as that, if you manage to go to college in the city, you can’t just go back home afterwards.
People will think you are a loser. People will look down on your family. They supported you and
you have nothing to show for it? “ After a pause she added: “And what about looking your parents
into their eyes and having to tell them: I couldn’t do it. How are they supposed to understand?”
She, too, will stay in Beijing after graduation and work hard, she said, for herself, and for her
family’s sake.
3.5.2 Class Matters
While some felt the pressure to make money, even send money back home as soon as they landed
their first jobs, others continued to receive support from their families. Lele, for instance, was
supported by her parents not only for months after graduation without a job, but continued to get
a monthly allowance from them to subsidize her modest salary. In her first journalism job she was
earning 4,500 RMB (ca. 690 USD) during probation, a 20% cut from the regular salary, but her
parents’ assistance allowed her to take the job and still rent a small room in a shared apartment for
2,500 RMB (ca. 380 USD) a month. Bo, another recent graduate from an elite university, told me
his parents were paying his rent. Several others reported getting help from their families to pay for
the deposits and other upfront costs to rent a place to stay. The two interviewees from Shanghai

81
who had lived in university dormitories during their studies moved back home when they started
working.
62

When I talked to the head of market research for the architectural arm of one of Shanghai’s
biggest state-owned construction firms about the housing struggles of recent graduates, he said he
was well aware of the situation: “If the starting salary is around 5000, which is common, you can
be sure that the family helps out.” To him, parents continuing the support of their children even
after they entered the job market was a generational phenomenon: “Those born in the 80s and 90s
are the first generations to grow up as only children and during a period of economic boom in
China. Families now can and want to do that[supporting their children].” The job in the big city
then becomes just another extension to their children’s journey of human capital accumulation:
“Look, Shanghai is and will remain China’s center of opportunities. That is why families are
willing to support their children during the first few years working here. They want them to gain
this experience. Even if salaries are not enough to live on, they think it is still worth it.”
My informants’ stories, however different, all have one thing in common: decisions on
education, migration, and even housing were not at all their decisions alone. Rather, those
decisions were deeply entwined with their respective larger familial contexts. Research on
international migration has long documented migration as family strategies, with the whole family
pooling resources to maximize the chances of success of the emigrant (Boyd, 1989; De Haas, 2010;
Palloni et al., 2001; Waters, 2006). In China, where disparities between regions can be as wide as
between countries, a similar dynamic seems to be unfolding.

62
Traditionally in China, unmarried adult children live with their parents until marriage or, more recently, until home
purchase.

82
And, as the disparate trajectories of Ninghong and Lele, show, where their families stood
in the social order mattered a great deal. Socio-economic background mattered for the
opportunities and the risks they could take and how comfortable – or not so comfortable – they
could be doing so. Filial piety as a social institution confined some more than others. Being able
to push back the moment when the direction of intergenerational transfers reverses is a huge
privilege.
3.6. The Promise of Upward Mobility
Sections 3.4 and 3.5 unpacked the motivation driving young people to seek jobs in China’s
increasingly exclusive big cities, as well as the structural socio-cultural forces co-producing the
group rental housing market in the process. Sections 3.6 and 3.7 address the question of sense-
making: How do young people handle the group renting experience emotionally? What processes
of individual and social cognition are involved in coming to terms with their struggles?
3.6.1 Expectations and Reality: Cognitive Dissonance Among the Elite
For most, moving to and arriving in Shanghai was bound up with great hopes and anticipation. As
Lihua put it: “I came to Shanghai because it’s a big city, it’s modern, it’s cosmopolitan (
wufang zachu). Shanghai is where things are happening.” Shanghai is China’s largest city and a
global financial and logistic hub with a large international and domestic non-local population.
Within China, people from Shanghai have a reputation for being fancy and fashionable, but also
arrogant and unwelcoming. Shanghai, in the popular imaginary, is the embodiment of China’s
progress. A broker told me he had seen lots of young people coming to him wanting to live
centrally in order to “feel like they are really here. Of course, they also want to show off to the
people back home, even if all they can afford is a bed and their actual life is far from fancy.”

83
While the image of Shanghai as China’s claim to the future appeared powerful, if somewhat
elusive, the promise of economic opportunity was very real. When I asked the question “Why did
you come to Shanghai?”, the answer was always “for better opportunities”, be it educational or
work wise. Meifeng, for instance, told me that what she loved most about her job here in Shanghai
was that she worked at an international company and spoke in English to foreign clients, mostly
in Australia: “It’s nice. I can feel like I am a talent.”
The word “talent” ( rencai) was actually present in a lot of my conversations. Since
the late 1990s, the government has discursively constructed “talents” as a new social category and
policy target (Biao, 2011). In the official discourse, talents are those with specialized skills and
high levels of education – with foreign-trained returnees at the top of the list (Zweig, 2006). As a
group, they have been painted as an invaluable resource, a driving force for progress and prosperity,
made out to be the country’s future (Central Government, 2010; Xinhua, 2018b). The
government’s promotion of human capital as a growth strategy comes to bear in a range of policy
initiatives as well. “Attracting talent” ( yinjin rencai), for instance, is at the heart of the
Shanghai’s hukou reform and has a municipal government department dedicated to overseeing
talent attraction strategies ( Shanghaishi rencai fuwu zhongxin, Shanghai
City Talent Service Center).
63
Hoffman (2010) calls the strategic merger of individual professional
development and state-building “patriotic professionalism.”
The clash of expectation versus reality was perceived most harshly by the elite graduates.
It was almost exclusively the graduates from top universities that talked about starting work after
leaving school with disbelief and disappointment. As Fong (2004) wrote, they were the ones who

63
It is only one of several such offices across the country (Hoffman, 2010).

84
had been “socialized to become part of the elite” (Fong, 2004: 28). Now they were finding out that
the struggle continued after university. Xianliang, a recent graduate from New York University’s
newly established Shanghai campus and a Shanghai native, told me: “I couldn’t believe it when I
started looking for a job. I graduated from NYU Shanghai, how are starting salaries so low?” Lele,
too, was shocked at how little her salary could buy her. “You know, we already knew we couldn’t
afford to buy in Shanghai, but now we can’t even afford to rent properly?”
The group renters among the elite graduates especially felt like life after graduation was
like jumping into ice cold water. Ninghong told me how she had felt completely ill-prepared for
dealing with figuring out life in Shanghai outside of campus: “I had no clue how to do this, looking
for a place to live. I feel cheated, I didn’t know what I would be able to afford, I didn’t know what
questions to ask when I decided. I didn’t even really understand what the role of (zhongjie,
a broker and property management company) was. I wasn’t prepared.” For elite graduates from
modest, rural backgrounds, their lack of knowledge or social capital often only becomes apparent
after graduation. Typically, their social networks in the city will consist of only their (equally
clueless) peers and their parents’ lifeworlds are so vastly different from what they are facing that
they cannot offer much guidance either (Suda, 2016; Wang et al., 2017).
When Yang, another recent graduate and group renter, heard about a raid of a group rental
unit in his building, he was shocked. Seeing the large posters with pictures of “ ” (chengguan,
an arm of local law enforcement) officers cleaning out the trashed apartment hung up all over the
entrance area and elevator of his building – presumably in part to scare off any other code violators
and in part to appease whose complaint had prompted the raid – he told me it was the first time he
was really forced to face the fact that where he lived, what he was doing was illegal.

85
By contrast, everyone in my group rental was well aware that what they usually just called
“ ” (sushi, dormitory) was illegal. Yet they were rather blasé about it. Meifang asked me about
my dissertation one morning and after I told her a little bit more about my research she asked “So
you know this here is illegal, right?” When I replied that I did, she added “Well, what else can we
do?” and moved on to the next question.
This matter-of-fact attitude also showed when I asked the women in the group rental where
I stayed how they felt about living there. The unanimous answer :“it’s so inconvenient ( bu
fangbian)!” I remember my surprise that they wouldn’t use a stronger word, but they simply
continued with listing things that were inconvenient: no place to store your belongings, lining up
to use the bathroom, no proper kitchen to keep or prepare food. I heard the word “inconvenient” a
couple more times in other descriptions of group rental living. Longwei told me: “You have to
know what you’re getting into. It is definitely a matter of attitude. If you are easily bothered, this
is not for you. Sure, I don’t like the snoring, standing in line to use the bathroom, but it’s fine. It
is temporary anyways.”
The non-elite college graduates and young migrant workers owed their pragmatic,
unemotional frame of mind to the fact that there was no chasm between what they imagined living
and working in Shanghai after graduation would be like and what it turned out to be in reality.
They felt less betrayed by their expectations; they had always anticipated that working in Shanghai
they would start from the bottom and working their way up was going to be hard. As one of the
brokers
64
I got to know quite well told me: “I see these young graduates arriving in Shanghai all
the time. But making it in Shanghai is quite hard nowadays. Do you know the saying “thousands

64
He is a Shanghai local and has been in this business for more than eight years.

86
of troops crossing a single-log bridge” ( qian wan maguo dumuqiao)?” The
Chinese saying is used to describe something that many will try, but few will succeed at.
How people respond to crammed living conditions depends not only on objective levels of
crowding, but also on expectations (Dunn, 2002; Dunn and Hayes, 2000; Hu & Coulter, 2017).
While research from the Global North has long linked lack of adequate living space to adverse
effects on mental health, research on housing conditions and psychological well-being in pre-
reform China found no such relationship. Traditions of multigenerational co-residence, close-knit
kinship ties, as well as cultural normalization of density were believed to cushion the negative
consequences of overcrowding for urban residents in urban China before transition (Booth, 1976;
Evans & Lepore, 1992; Evans et al., 2003; Gove & Hughes, 1983; Huang, 2003; Wang, 2004).
More recent work, however, now finds significant association between residential crowding and
poor psychological well-being; especially for those on the higher end of the income distribution.
Radical housing reforms, together with fundamental socio-economic, and cultural changes have
raised people’s space expectations (Hu & Coulter, 2017; Huang, 2003; Yi &, 2014). The
psychological distress, also referred to as cognitive dissonance, is caused by a mismatch between
expected and actually accessible living spaces (Dunn, 2002; Dunn and Hayes, 2000).
How findings differ over space and time indicate that expectations for and psychological
responses to housing conditions are both subjective and context dependent. The group renters from
more modest backgrounds and with no or less prestigious higher education were less affected by
their living conditions because their expectations were more in line with their limited means. The
graduates from elite universities had aspired to more and had come up short.

87
3.6.2 Internalizing and Reproducing “Sacrificing” as a Social Institution
Notwithstanding differences in educational attainment, all the young people I talked to were united
in their strong belief in the transitory nature of group renting. Recall, for instance, how Longwei
concluded the evaluation of his living situation with “It is temporary anyways.” The women in the
room where I rented frequently discussed the idea of moving out. Although they each had only
arrived weeks before I did, they were already making plans for their next move. Notably, after I
had asked them how they liked living in the group rental and they had named all the ways in which
it was inconvenient, Jing started making a real job of looking for a new place to stay. When I met
Meifeng a year later, she told me how Jing and her found a different, less crowded group rental
just weeks after I left.
Knowing and articulating why they rented beds, having a reason, certainly also helped.
Group renters talked to me about saving money, gaining work experience, making a career change,
or buying time while working their way up. Importantly, however, these short-term reasons were
in service of long-term goals.
Xianliang, the NYU Shanghai graduate, said he was trying to see his “disappointingly low”
starting salary and the sacrifices he was making
65
to take a low-paying research and writing
position for an architectural magazine as another form of investment in his future. “Okay maybe I
am not paid much money right now, but I am gaining experience, I am building a network. All of
that is social capital. And at least I do have the potential to earn much more in the long run.” Lihua,
too, talked about having to pay her dues, but expecting it to pay off in the long run. “In this industry
[hospitality] experience counts a lot. Everyone has to work their way up, prove themselves,

65
He moved back in with his parents with whom he has a strained relationship. In order to be closer to financial
independence, he has a side job as an English speaking tour guide for Shanghai’s architectural history.

88
regardless of their education. I have to say I was surprised, but it seems like most of my colleagues
and roommates went to university too. We all have to follow this process. Of course, if you can
get into a high-quality place, like an international hotel chain, that experience will count a lot more.”
Even Hong, who rented a bed for more than three years because of his family’s pressure to keep
expenses low, was eager to find a silver lining to his experience: “You see, what I have now is the
ability to adjust to all kinds of circumstances. I have done it. I know I can do it. It’s something I
have, something the spoiled Shanghai locals definitely do not have.”
Believing in temporariness and long-term payoffs helped the people I talked to cope with
current disappointments and the stresses of group renting. This kind of optimism for the future
despite present day precarity is actually wide-spread among less privileged educated migrants (He
& Mai, 2015; Liang, 2010; Wang et al., 2017). Their emotional investment in this narrative allowed
them to live on future dreams.
It also worked, because the notion of a better tomorrow lying ahead was constantly repeated
and reinforced by everyone and everything around them: Parents, same-age peers, government
institutions and media all held up and normalized the struggle for rewards over the long haul.
The parent generation of the young people I met grew up witnessing China change from a
poverty-stricken country under socialist rule to a rapidly urbanizing economy power house. They
had likely experienced or at least seen prosperity spring from deprivation, and themselves gone
through hardship. “Eating bitterness” ( chi ku), meaning enduring hardship, is commonly
heralded as character-forming by many Chinese parents. The well-known legend of a fish
swimming against the current and jumping over a gate to become a dragon ( liyu tiao

89
longmen) is told from young age and teaches that overcoming great difficulties will be
recompensed in the end.
The narrative of suffering for eventual achievement was also pushed by the new party line
built around the “Chinese dream” ( zhongguo meng). First proposed by Xi Jinping after he
took office in 2012, the China dream emphasized the interconnectedness of “the people’s” dream
of a “great revival” ( zhonghua minzu weida fuxing) and individual efforts,
sacrifices, and accomplishments (Wang & Feng, 2016). Slogans such as “With a dream in your
heart, there is power under your feet” ( xinzhong you mengxiang,
jiaoxia jiu you liliang), were frequently found on banners and propaganda posters around Shanghai.
Xi Jinping and government media regularly called upon the young generation in particular to work
hard and make sacrifices in order to reach individual and collective success (CCP News Net, 2018;
Xiang, 2019b; Xinhua, 2018c; Wang & Feng, 2016).
What’s more, many of the group renters mostly knew other group renters. As migrants they
formed large parts of their social networks after their arrival in the city. With similar stories all
around them, the relatively homogeneous peer interactions further helped normalize the young
people’s adverse experiences.
The fact that group rentals were able to hide quite effectively in regular apartment buildings
also speaks to a level of general buy-in and sympathy for the narrative of young people struggling.
Among other things, landlords spent considerable time, money, and energy negotiating their illegal
businesses with compound security guards, neighborhood committees ( juweihu),
66
and

66
Neighborhood committees are the lowest administrative level of the government and in charge of policy
enforcement and registering residential movement.

90
neighboring residents. They frequently brought small “gifts” such as cigarettes, fruit, or liquor.
But landlords and brokers also relied on the trope of young people struggling to make a living and
build a career in order to get all those who knew to turn a blind eye. As one landlord tried to
convince me: “Many young people, college graduates, they come to Shanghai for work, but they
can barely pay for rent and food. I am helping these people.”
67
Recall also how the descriptive
texts of the listings advertising bed space online overwhelmingly referred to “students”, “recent
graduates”, “white-collar workers”, and “newcomers.”
While I was in Shanghai during summer 2018, a 29-year old Henan man attacked a group
of young students in front of their elite private primary school and killed two young boys who had
been waiting to be picked up. The incident happened in Shanghai’s Xuhui district, where I lived
at the time, and everyone was talking about it. Reportedly, the attacker had wanted to “takes
revenge on society” after he had graduated from a decent university but failed to secure proper
employment for years (Kuo, 2018; Ramzy, 2018; Zuo, 2018). For a couple of weeks afterwards,
everyone I talked to about my research connected it to the attack. While most people I spoke to
condemned the act, they also had some sympathy for the attacker. As one of my neighbors put it:
“The pressure on young people these days is just too high!”
Cognitive anthropologists have proposed that human motivation develops from cultural
models. Cultural models are held broadly among members of a community and create narratives,
expectations, and goals to make sense of the chaos of human existence (Bennardo & De Munck
2014; D’Andrade, 1981; 1992; 1995; Quinn et al., 1992). The narratives normalizing personal

67
Landlords and brokers likely also made payments as additional “incentives”. But none of them wanted to speak
about any bribes besides the “small gifts”. Longwei told me that the landlord of his group rental managed three rentals
in the same building, 30 people each. Each one of the tenants had gotten a chip card, needed to open the electronic
entrance gate to the compound. That is 90 cards. He tells me: “There you can see how good his relationships are…”

91
struggle, working one’s own way up, and believing in long-term rewards were so powerful because
they were constantly socially reproduced.
3.6.3 Questioning the Viability of the Social Contract
Yet, there were signs of contesting this narrative, the discursive construction of a counter culture
of sorts. Xiaofei, for instance, vocally complained and wondered whether all the efforts she put
into her job were really going to be worth it. She had graduated from the well-known Northeast
Normal University ( dongbei shifan daxue) in 2016 and was now working in the
Human Resources department of a large company, her second job since arriving in Shanghai. She
started work at 8:30 am and although she was supposed to be able to leave by 5:30 pm, she never
left before 7:00 pm. Most days she stayed until 10:00 pm or even midnight, working weekends if
needed – all without overtime compensation. She kept telling me about her “crazy eye boss” who
had money, a car, and an apartment in Shanghai,
68
but never went home to be with his family.
Throughout our conversion, Xiaofei kept coming back to anecdotes about her workaholic boss as
if saying “is this really what I want for my future?”. At some point she added, almost as a cynical
postscript: “Have you heard of the terms “overtime dog” ( jiaban gou) or “single dog” ( dansheng gou)? No? Well, that’s who I am: Working all the time and no boyfriend in sight.”
“Overtime dog” and “single dog” is internet slang ( wangluo liyu) playing with
the traditionally negative perception of dogs
69
in Chinese culture for self-ridicule. “Overtime dog”

68
In 2019, it cost around 15,000 USD to buy a license plate in Shanghai (Zhou, 2019). Shanghai’s average property
price was around 715 USD per square foot in 2017 (Ren, 2020).
69
Traditionally, describing people as “dogs” would be seen as insulting or derogatory. Examples include “dog officer”
( gou guan), outdated term used to insult corrupt and lazy officials, or “dog man and woman” ( gou nan
nü), used to describe a cheating couple.

92
is used by white-collar workers to make fun of their perpetual state of working overtime ( jia
ban). In 2016, a Shanghai choir dedicated a tongue-in-cheek song to white-collar workers’ daily
misery. The lyrics to the song titled “My Body Feels so Drained” ( ganjue shenti
bei tao kong) are pointedly descriptive. Lines such as “No time to remove my makeup for 18 days.
I’ve worn these monthly lenses for two and a half years. My life is a mess” make the day-to-day
struggles come to life and resonated with millions as it went viral shortly after having been shared
on WeChat.
70

Figure 4: The Shanghai Rainbow Chamber Singers Perform “My Body Feels so Drained”

Sources: Screenshot Taken from Youku (https://v.youku.com/v_show/id_XMTY2Njg5NjA0MA==.html)

China’s online communities are full of witty self-deprecation. For instance, there is the
term (fangnü), literally meaning “house slave” used by those who have tied their lives to
large mortgages for (urban) apartments and constantly struggle financially to pay off monthly
installments (Chiang, 2018). Or, the widely popular (diaosi), literally meaning “male pubic

70
The song by the Shanghai Rainbow Chamber Singers ( ) is currently available on youtube.com:
https://www.youtube.com/watch?v=XFaQwZyPOZQ.

93
hair”, but actually meaning “loser” as opposed to the idealized (gao fu shuai), a man who
is tall, rich, and handsome. Originally a derogatory term, it has become a way to relate for single,
less financially successful men who have claimed the label as their own (Cao, 2017).

In September 2018, the sudden death of a middle-aged man who had rented a room from
the popular online rental platform Ziroom ( ziru) in Hangzhou caught the public’s attention.
The man had moved to Hangzhou for work and had rented a flat through Ziroom. His death was
blamed on high levels of formaldehyde in the paint of his rented home (Zhang, 2018; Xinhua,
2018d). Immediately following the public uproar, memes started popping up on WeChat. The one
below was shared with the title “You see, you work yourself to death ( lei si lei huo) to
save some money in order to rent a place, and then you have to worry about the harm of
formaldehyde.” The meme and text describe a “talent” ( cai) being swallowed whole by
formaldehyde.
Figure 5: Meme Satirizing the Housing Struggles of Young Adults in Urban China

Sources: Created and shared by popular WeChat public account “Bad Reviewer” ( chaping jun)

94
Also, just days before the new iPhone XS was released, a Shanghai real estate consulting
company put together their calculations of how much rent the new iPhone could buy you in
different locations in Shanghai. Meant as a joke and put together to advertise their services, the
article seemingly hit the mark and was shared widely on WeChat (Urban Surveyors, 2018).
These are just a few examples of a very lively discourse happening online. Making fun of
the absurdity of their situations, releasing building cynicism, and finding comfort in a sense of
shared commiseration all have psychological benefits (Liu, 2018; Michel, 2015; Moor, Bindler &
Pandich, 2010). The active online exchange indicates a need to process these experiences. But is
the collective self-mockery also indicative of seething resentment in the face of too much pressure
and too few opportunities? Is class consciousness rising in contemporary China?
3.7. Class Consciousness?
The previous section discusses the individual internalization and societal reproduction of
“sacrificing” as a social institution, as well as nascent cognitive dissonance and the humorous
expression of discontent online. It addresses how group renters make sense of their experiences as
part of their individual biographies and also shines a light on the societal context within which this
sensemaking takes place. This sections asks whether individual experiences accumulate to a rising
consciousness about the contemporary social hierarchy in China. Did group renters identify class
relations? If so, how did they position themselves in relation to them? And what was the societal
discourse around class?
3.7.1 Shanghai Divided
Throughout my field work, the lack of awareness among the more privileged young people I spoke
with stayed with me. Group renting was an open secret among a certain group of people in their

95
20s – as well as landlords, brokers, immediate neighbors, security guards, and local law
enforcement. Yet, most of my friends and the majority of elite graduates I met were shocked and
often even wary of my stories initially.
In fact, incredulity became the expected constant whenever I first told anyone about my
research. Shanghai locals in particular would insist that group rentals were “a problem of the past”.
For others, their imagination immediately filled the beds for rent with migrant laborers (
dagongzhe). Bo, a Fudan University ( fudan daxue)
71
graduate, for instance told me: “I
know about group rentals from my chats with didi drivers,
72
they talk about it because that’s where
they live. College graduates will be supported by their parents if they can’t make ends meet.”
Telling people about my research soon became a predictable back-and-forth. I would
describe the research and look at faces scrunched up in disbelief. The first question would always
be about the tenants, the next about density. Most could not believe that I had seen 30, even 40
people sharing otherwise ordinary three-bedroom apartments. Sometimes, I even had to show
pictures to get people to consider that what I was telling them was indeed happening.
Some of the skepticism might very well be owed to the fact that it was me, an outsider, a
foreigner, talking about something they didn’t know was going on in “their city.” But another part,
I believe, stemmed from a resistance to acknowledge that precarity was actually quite close, rather
than happening far away to “other” people. Was there a deliberate element to the ignorance?

71
Fudan University is Shanghai’s most prestigious university and widely considered one of the most prestigious and
selective universities in China.
72
Didi ( ) is China’s biggest ride hailing app-based service, akin to Uber or Lyft in the United States.

96
A few days into the 2016 fieldwork, I watched the research assistants’ initial bewilderment
morph into a rumination of sorts. Initially, they had been rattled by the at times gruesome living
conditions we saw. But something seemed to shift as they realized that most of the tenants were
recent college graduates. Riding down the elevator after seeing a group rental apartment one
afternoon, the research assistant I was paired up with that day turned to me and said with concern
in her voice and on her face: “I am starting to ask myself, is this my future? Is this what is next for
me, too?”
The urge to draw a line of distinction, to paint the group renters as a different “other”
echoed in a conversation I had with two recent Tongji University graduates. When I just met the
couple, both working in real estate, and told them about this project, they said they did not know
any group renters. After we had been talking for a while, however, Tingting said she remembered
some of her colleagues talking about how they had rented a bed when they just started working at
their company. She said she had been surprised to learn this about her colleagues and added
empathetically: “At the time they could have afforded more, they told me, but not a whole
apartment by themselves. So they were stuck, they had no choice but to rent in a group rental and
to save up to move out.” I asked her whether she would ever consider group renting. The couple
looked at each other and, after a moment of pause, Tingting answered as her boyfriend nodded in
support: “It’s hard to say…the environment and the people…they are just not our crowd. Yes, I
think that is how you could say it. They are just not our crowd.”
In a conversation with a senior housing market and policy analyst at one of Shanghai’s
most prestigious research institutes, prejudice was more unapologetic. Yes, he told me, he knew
about group rentals: “Group rentals are breeding grounds for crime, drugs, prostitution, all sorts of
things. We know about this. That’s why we need to control this kind of low quality [ suzhi

97
di]
73
people and their ugly [ chou] behavior.” Throughout the rest of our exchange, he dismissed
any of my attempts to convince him otherwise.
Interestingly, many of the group renters, too, were eager to draw lines of distinction; only
they were concerned with putting distance between them and their past. Tingting, for instance, told
me that she heard about her colleagues group renting only much later, after the they had long
moved out; and even then they did not really like talking about it. I almost did not hear about
Haotian’s group rental experience. The electrical engineer, whom I met more than eight years after
he had rented a bed when he was just out of college, only told me about it because I insisted on
going back in time chronologically to get a better understanding of his employment and housing
choices. We had been talking for more than two hours when he came out with it; and even then he
was eager to change topics. Mingzhu, the former street vendor, too, was reluctant to talk about her
group renting experience. Although willing to sit down and talk to me to help me with my research,
she was extremely vague on many of the details. I met her at her home for lunch with her husband
and their infant daughter. Mingzhu is the wife of the cousin of a friend of mine, who was also
present during lunch. When I tried to press her on how long she stayed in the group rental, for
instance, her husband, a Shanghai local, jumped in – almost as if to protect her from having to dig
out the memory: “Look, I went there once, only once to visit. It was not nice. Believe me, I am
telling you, not nice at all.” Neither of them wanted to dwell on memories of that time, it seemed.
None of the young people I met, not as current but as former group renters, were
particularly keen on talking about their experience. They had all “made it” to what might be

73
The concept of suzhi is widely used in official and popular discourse to talk about the relative “quality” of
individuals or groups of people. It has many dimensions and, as discussed elsewhere (Anagnost, 2004; Jacka, 2009;
Qian, 2018) is intimately linked to notions of civility and modernity, often used to justify the differential life conditions
and prospects of people deemed having different levels of suzhi.

98
described as an urban middle-class standard of living and it seemed like they did not care for
looking back. As they had moved away from group renting and towards a middle-class lifestyle
they must have been exposed to the ignorance and prejudice of the privileged. Perhaps treating
group rentals as a faint memory of something they briefly did in the past was a way to cope and
create a cohesive internal narrative of their trajectories.
3.7.2 Facing the Elephant in the Room: Class Relations in Contemporary China
The lives of group renters and their more privileged peers constantly intersected – they were each
other’s co-workers, neighbors, fellow transit riders, and shoppers. Yet, the group renters’ private
social world remained hidden as if their struggles were a societal taboo, a secret kept by shame
and othering.
What could explain the emotional response? Generally, embarrassment is felt when failure
to meet an expectation or a desired standard is attributed to the self (Fiske & Taylor, 1991). But
how come cognitive processing seemed to center on the individual rather than conceiting the role
of structural forces and increasingly rigid class relations?
Just as “sacrificing”, “personal responsibility” as well has been actively fostered as a
societal narrative in post-reform China. Prior to reforms, class consciousness had been central to
the CCP’s ideology – governance and policy making spent decades on flattening social hierarchies
and promoting the revolution of the peasantry. With the regime change, however, emphasizing
class was growing incompatible with the party’s post-reform growth agenda and the new CCP-
leadership needed to change course in order to address the newly emerging social and economic
hierarchies (Guo, 2012; Whyte, 2010). The notion of “class” had become politically fraught.

99
In an effort to depoliticize class relations, a stylized middle class was singled out as the
model for others to emulate. Soon, expanding the middle class became a political project, the
CCP’s antidote to looming social instability in the face of widening inequality (Crossley, 2012;
Tomba, 2004). The fight against any nascent social antagonism was accompanied by a new
language intended to repackage the national discourse around inequality: In any official social
analysis the word “class” ( jieji) has been replaced with “social stratum” ( shehui
jieceng), Hu Jintao spent much of his tenure promoting a “harmonious society” ( hexie
shehui), and Xi Jinping’s “China dream” ( zhongguo meng) emphasizes the interrelation of
individual and national, collective aspirations (Anagnost, 2008).
Perhaps most influential, however, Deng Xiaoping’s “let some people get rich first” ( rang yi bufen ren xian fu qilai) restated inequality as a matter of developmental
time. By framing those who prospered first as harbingers of more wide-ranging prosperity to come,
they were elevated to role models rather than becoming the target of class resentment (Anagnost,
2008; Tomba, 2004).
As the economy grew and the ripple effects were indeed felt widely – even if unequally so
– the collective experience of growth and progress further aided the popularization of the narratives
around personal responsibility and working one’s own way up the socio-economic ladder (Whyte,
2010; Woronov, 2012). In this new narrative, individual life trajectories are now seen as the sum
of individual choices, obscuring any structural elements to inequality.
The experience of rapid and sustained economic growth also matters as a frame of reference
for reflecting on class relations. Class consciousness requires awareness of one’s place in the social

100
hierarchy. But current generations have seen the social hierarchy overthrown more than once, have
witnessed more class disruption than class continuity. Even if social mobility is noticeably
becoming harder to achieve, group renters still have so many more opportunities than their parents
did. My sense is that although the path to upward mobility is narrowing, the changing dynamics –
intensifying competitiveness in labor markets and the materialization of pronounced class privilege
– are instead perceived as a lengthening of the same path. From the many hours spent in
conversation with group renters, it appears they still believe in upward mobility and eventual
middle-class belonging through education, migration, hard work, and sacrifice.

101
CHAPTER 4. IS SHANGHAI A SPECIAL CASE?
4.1. Introduction
Chapter 3 draws on my ethnographic research to delve into the social meaning of group rental
housing beyond its function as a niche informal housing market. In particular, the previous chapter
explored questions about motivation and sensemaking.
The chapter started by describing the overcrowded living situation and the opportunities
group renting affords. In the short term, economizing on rent by minimizing space enabled my
interviewees to save money – either to send back home to their families or because a new dynamic
of structural forces is producing lower starting wages for most educated migrants. Either way, in-
depth interviewing revealed group renting as a stepping stone, a means to buying space and time
while group rental tenants set up better situations for themselves. Looked at as one entry in each
interviewee’s personal timeline, group rentals turned out to be part of a long-term mobility strategy.
Group renters’ bid for upward mobility and financial security is driven by social norms, in
particular filial piety, that dictate moral indebtedness to their families. In a rapidly aging society
with a tenuous pension system and public safety net, achieving higher social and financial status
has become a high-stakes goals as it means the ability to fulfill familial obligations by paying
respect and financial support.
Filial piety as a social institution, however, trapped some more than others; Socio economic
background and hukou status significantly influence the job market entrants’ experience. As has
been discussed in the international migration literature, the migrant’s actions are deeply

102
intertwined with the larger family: decisions on education, migration, and even housing were taken
not individually, but as part of a family strategy.
The chapter also investigates how the young people I talked to processed their experiences.
Asking about cognitive processes gets at the cultural model, the societal narrative underneath the
social phenomenon. The individual and collective sensemaking connects it to the larger question
about changing mechanics of social mobility and class relations. Group renters and other
struggling job market entrants process their experience by focusing on temporariness. The coping
happens in a societal contexts in which narratives about sacrificing and individual responsibility
to achieve social mobility are constantly reproduced. Having witnessed more class disruption than
continuity, my interviewees did not voice class-based sentiments. They perceive their
opportunities as still many, and many more compared to their parents’ generation.
The goal in this chapter is to contextualize the findings from Shanghai and triangulate some
basic claims generated based on the ethnographic fieldwork. In particular, I address research
question five, which asks whether the group rental phenomenon captured in Shanghai is a regional
anomaly or the local instance of a larger phenomenon. In other words, how relevant is this in-depth
study of group renting in Shanghai for the Chinese society as a whole? Is Shanghai a special case?
Several of my interviewees reported group rental experiences outside of Shanghai, in
Beijing, Dalian, Guangzhou, and Wuhan. I also read newspaper articles and government
communication regarding group rentals in several other cities besides Shanghai. When I first
started web scraping online advertisement data, I also monitored daily ads for the largest cities:
Table 6 shows bed space rental listings across all major Chinese cities.
74
In sum, I had reason to

74
Note that volume speaks more to activity, not necessarily market size.

103
believe that group renting was not confined to Shanghai. Still, how the case of group renting in
Shanghai fits in with group renting throughout China was an open question.
Table 6: Average Daily Online Advertisements for Low Rent Shared Housing in China’s
Most Populous Cities

City Average Daily Advertisements Urban Population in Million

Shanghai 228,738 24.5
Beijing 216,388 21.5
Guangzhou 10,356 20.8
Tianjin 21,891 15.5
Shenzhen 22,545 12.4
Wuhan 53,328 10.7
Hangzhou 23,737 9.0
Nanjing 54,027 8.2
Source: Population data from the National Bureau of Statistic of China (2010); Online advertisements
retrieved from gangji.com in June 2017.

In relation to research question five – Is Shanghai a special case? – I hypothesize that
Shanghai status as a Tier 1 city matters, meaning that I do not expect to find any meaningful
difference with other Tier 1 cities, but do expect Tier 1 cities to differ substantially from non-Tier
1 cities. Recall that Tier 1 cities (Beijing, Shanghai, Guangzhou, and Shenzhen) have traditionally
been the most densely populated, with the biggest economic, cultural, and political influence.
To address the question of how to place the findings about Shanghai in a larger context, I
draw on survey data collected via a cell phone social media survey conducted in July 2019. The
survey approach leverages the widely popular Chinese app “WeChat” to reach over 25,000 survey
respondents working in cities across all over China. The survey design was informed by the
ethnographic research. In particular, the survey instrument asks about general demographics such

104
as age, gender, hukou, as well as educational background, employment and housing trajectories,
family financial support, and current employment status.
75

Drawing on descriptive statistics, I compare responses between Shanghai and other Tier 1
cities, as well as between Tier 1 cities and non-Tier 1 cities. More broadly, the data also allows me
to better understand the prevalence of group renting as a housing strategy among job market
entrants, as well as to triangulate qualitative findings about group renter demographics. In
particular, I am also able to probe whether those without Tier 1 city hukou and without degrees
from elite higher education institutions indeed make up for the majority of group renters.
4.2. Research Design and Data Collection
4.2.1 Survey Design
The survey instrument repackages a life-history approach to fit into a survey that can easily be
disseminated widely and answered quickly. Depending on the answers of respondents, it includes
22 to 33 question.
76
The survey asks about general demographics such as age, gender, hukou, as
well as educational background, employment and housing trajectories, family financial support,
and current employment status. The full survey instrument can be found in Appendix B.
I used the insights from the ethnographic research to inform the design of this survey.
Relevant cut-off points for salary or rent buckets in closed-ended questions, for instance, were
informed by the conversations with recent graduates in 2018. These conversations also guided the

75
Findings presented in this dissertation are partial and preliminary. The rich survey data was collected with the
intention to develop a larger argument about the relationship between skills, wages, and housing in the context of
urbanization and (uneven) economic development.
76
For instance, the survey asks about a second job after graduation. If the respondent answers that they have not
changed jobs, the survey tool will automatically skip ahead to the next relevant question. For more details see
Appendix B.

105
kind of questions included. For instance, I wanted to get at the quality difference of degrees from
elite versus non-elite institutions. These would have be averaged out, had I asked about levels of
educational attainment only. In the survey I hence I added a simple yes-no question about whether
the degree-granting institution was part of the 211 or 985 projects (Question 5, Appendix B).
Recall that “Project 211” and “Project 985” are government designated institutions of excellence
that receive the majority of government funding, need-based scholarships, and are also the most
selective (People’s Daily Online, 2008; Xinhua, 2017a). I learned that “211” and “985” was
commonly known terminology through interviews and was able to exploit this knowledge for
question design.
In general, I kept questions short and formulated them with the short attention span of
survey respondents in mind. Out of the 33 question, approximately half are closed-ended multiple-
choice questions and half are structured open-ended, meaning that they simply required the input
of numbers of names (of cities, universities etc.). I estimated that the survey could be answered
within six minutes if all 33 questions were applicable.
77
The mean fill-in time for actual
respondents was 301 seconds, roughly 5 minutes, with 80% of surveys returned after 400 seconds
(6.7 minutes).
The actual wording of questions is critical in expressing the meaning and intent of the
question and in ensuring that it is interpreted in the same way across survey respondents. Here,
too, local knowledge I developed through my long-term investment in the Chinese context, as well
as Chinese language fluency helped me in using culturally appropriate terminology.

77
The time estimate is based on the survey length prediction tool provided by Versta Research, a marketing research
and public opinion polling firm (https://verstaresearch.com/newsletters/how-to-estimate-the-length-of-a-survey/).

106
The question about hukou registration is a case in point. Recall that hukou is tied to access
and also the source of wide-ranging discrimination that can touch almost every aspect of life,
including housing, employment, and education (Chan, 1996; 2010; 2013). Consequently, it is
conceivable that some respondents may not feel comfortable giving out information that feels too
personal, or they may take offensive in a question perceived as intrusive, potentially compromising
their willingness to take the survey. I thus decided to pose the question about hukou as an
unstructured open-ended question to give respondents the option to respond with the level of detail
of their choice (province, city, district, etc.).
78

Most importantly, however, I did not solely rely on my own judgement in designing the
survey. In a first step, I solicited the help of a Chinese native speaker, who is also in the target age
group, in designing the preliminary working version of the survey instrument. Next, I asked a
group of friends and former interviewees to test the survey and give me feedback for improvement
and clarification. Through this iterative pilot testing over multiple rounds, I was able to fine-tune
wording to avoid confusion and ensure precise measurement of variables of interest. Piloting the
survey prior to launching it also allowed me to check whether closed-ended question choices were
exhaustive and mutually exclusive, and if needed, which answer choices needed to be added or
modified. Finally, the online tool I used to set up and distribute the survey (Wenjuan Xing ) is widely used by Chinese higher education institutions and some of them make their past
survey instruments freely available online. Wenjuan Xing’s website has a large repository of
sample surveys, which I also used to check the format and appropriateness of questions.
79

78
Note that the level of detail in this answers may be a function of the respondent’s social vulnerability.
79
For more information see company website: https://www.wjx.cn/newwjx/mysojump/newselecttemplete.aspx.

107
4.2.2 Survey Distribution
The survey was conducted using the WeChat mini-app Wenjuan Xing ( ). Wenjuan Xing is
a professional online survey, evaluation, and voting platform that includes free questionnaire
design and data collection. The real value, however, lies in the platform’s integration with WeChat
( weixin). WeChat is the most widely used social media and private messaging app in China.
In 2018, its mother company Tencent announced over one billion active registered users (Jao,
2018). Beyond messaging and social media, in urban areas – where smart phone and mobile
internet use is widespread and affordable – WeChat has become integral to most people’s routines.
It now features the ability to send and receive money, pay in stores and for public transit, as well
as booking planes, trains, and hotels, to name just a few CNNIC, 2018).
80
Given prevalence of
usage, the survey tool is well suited to reach the target population while keeping the risk of bias
due to structurally unequal access low (Mossberger et al., 2012).
After setting up the survey on Wenjuan Xing, the mini-app automatically generates a QR
code, which can easily be shared and when scanned immediately starts the survey. To promote
survey participation, I created a range of promotional material with different designs incorporating
the QR code (Appendix B) and attached a small monetary award for completed surveys.
81

To distribute the survey, I used a combination of online and offline promotion strategies.
First, I drew on my own social network, sending individual messages to all of my 223 WeChat
contacts, asking them to participate in the survey and/ or to promote it in their respective social
networks. To counter any bias introduced by the limitations of my list of contacts, I also hired local

80
Being central to so many people’s daily lives, WeChat has recently be harnessed for research in a wide range of
areas (Harwit, 2017; Montag, Becker & Gan, 2018; Tong, 2013; Zhang et al., 2017).
81
Wenjuan Xing allows for the automatic transfer of “red envelops” after successful survey completion. More on the
determination of “successful” later on.

108
research assistants to leverage their social networks for survey distribution. I recruited 12 local
undergraduate students from different institutions for this project and trained them in how to use
personalized messages to promote the survey. I specifically instructed them to reach out to those
among their social networks, who had started working without attending an institution of higher
education. Additionally, I also divided them up into groups of two and had them do random
intercept surveys in busy pedestrian areas during the after-work rush hours, using name card-sized
flyers with the QR code and printed survey sheets.
Presumably because of the financial award, the survey turned out to circulate quickly. After
the pilot run on July 1
st
yielded a total of 4,200 responses within less than 4 hours, I decided on an
iterative process of releasing and stopping the survey for limited time windows. I repeated this
process again on July 5
th
, 8
th
, and 10
th
, each time letting the survey run for a few hours in the
evening.
The raw data count of completed surveys was 25,640. To increase data reliability, Wenjuan
Xing offers a number of “safety measures.” These include requiring respondents to log on via
WeChat and restricting the submission of responses to one survey per account, as well as setting a
minimum time to complete the survey for it to be counted as valid.
82
Applying these safety
measures, Wenjuan Xing flagged 9,731 responses, which I subsequently excluded. Further data
cleaning yields a final data set of 15,345 valid responses with respondents from all of China’s 26
provinces.

82
I set this time to 3.5 minutes.

109
4.3. Findings
4.3.1 Likely Oversampling of High-Skilled Workers
Table 7 shows descriptive statistics for a selection of the WeChat survey data. With a mean age of
25 and 78% of respondents between the ages of 19 and 30, the survey reached the target population
of (recent) job market entrants. Note that slightly more women (54%) than men answered the
survey. Given that roughly two thirds of respondents have a bachelor’s degree or above, this does
not necessarily mean oversampling of women since recent government statistics report the share
of female college students at 53% (Xinhua, 2017b).
Table 7: Descriptive Statistics for All Observations (N=15,345)
Observations Mean SD Min Max
Demographics
Age 14,967 25.1 5.56 15 59
Gender (F=1, M=0) 15,345 0.54 0.5 0 1
Tier 1 City Hukou 15,345 0.17 0.37 0 1
Tier 1 City First Job 15,345 0.27 0.45 0 1
Education
Educational Attainment
0= No Higher Education
1= Vocational School
2= Bachelor’s Degree
3= Master’s Degree
15,345 1.8 1.04 0 3
Bachelor or above 15,345 0.67 0.47 0 1
Elite Education 12,832 0.54 0.49 0 1
Salary
Salary (<5,000 RMB) 15,334 0.82 0.38 0 1
Salary (<3,000 RMB) 15,334 0.40 0.49 0 1
Trial Period 15,339 0.91 0.28 0 1
Housing
Rent (<1,000 RMB) 14,805 0.71 0.45 0 1
Rent (<600 RMB) 14,805 0.43 0.5 0 1
Group Renting 10,421 0.21 0.41 0 1
Sharing a Room 10,421 0.79 0.41 0 1
Family Support
Family Provided Housing 15,345 0.12 0.32 0 1
Family Financial Support 12,676 0.64 0.48 0 1

110
17% of respondents have a Tier 1 city hukou and 27% of respondents got their first job in
a Tier 1 city. The total population of Tier 1 city residents makes up for about 10% of the country’s
urban population of roughly 780 million people, but the share of working age population may be
higher (Ding, 2019; National Bureau of Statistics, 2010; Roxburgh, 2018). To the best of my
knowledge there are no comparable data for job market entrants across the educational attainment
spectrum. For college graduates, recent survey data puts the share of job market entrants that take
their first job in Tier 1 cities at a roughly comparable 33% (Zhaopin, 2017).
While oversampling in terms of gender of geography seem less concerning, oversampling
at the high-end of the skill-spectrum is likely. In 2016, higher education enrollment was reported
at 43% (Sun, 2017). Among survey respondents, two thirds reported having a bachelor’s degree
or above. Additionally, only 16% of respondents started working without any higher education.
Of those with post-secondary education, 54% reported graduating from an elite institution.
The overwhelming majority of respondents (82%) earned an after tax monthly salary of
less than 5,000 RMB (approx.770 USD), and out of these, about half started working earning less
than 3,000 RMB (approx. 462 USD). Finding other data sources to put these findings into context
is difficult. For instance, survey data from Shanghai found that the average monthly income of
migrant workers across the skill and age spectrum was 5,328 RMB in 2016 (Chen, 2017); but
Shanghai is on the high end of China’s urban wage distribution and the reported mean averages
across workers with different levels of experience (Chen, 2017; Yu, 2019). Another survey reports
the average income of migrant workers across all Chinese cities at 4,107 RMB in 2018 (National
Bureau of Statistics, 2019). Again, this number averages across migrants of all ages and levels of
experience. Survey data among college graduates across China in the same year, found an average
starting salary of 4,317 RMB (Xinhua, 2018e), but I was unable to find data for entry level salaries

111
among those without postsecondary education. Against the backdrop of my interviews in 2018,
however, the numbers seem reasonable, as only those with Master’s degrees from elite universities
and/ or in technical fields reported starting salaries above 5,000 RMB. Note also that in line with
my interviews from 2018, 91% of respondents reported going through a trial period during their
first job.
Most survey respondents (71%) paid less than 1,000 RMB (approx. 154 USD), almost half
(43%) less than 600 RMB (approx. 92 USD) in rent every month. Here, too, it is difficult to find
data for contextualization. A government report from 2013 puts the average rent paid by migrant
workers in Shanghai in 2012 at 538 RMB, but Shanghai is known to have some of the highest rents
in the country and the average was taken across different household sizes (Ministry of Agriculture,
2013; Shen, 2015). Notably, the majority reported sharing their room (79%), but only 21% shared
with more than one other person.
83
Only 12% relied on family provided housing, but almost two
thirds (64%) received financial support from their families, even after they had started working.
In sum, the survey data seems to be by and large representative, barring potential bias from
likely oversampling on the top end of the education distribution. Given the generally positive linear
association between educational attainment and salaries (Appleton et al., 2014; Gao & Smyth,
2015; Hu, A., & Hibel, 2014; 2015; Zhong, 2011), the survey data may show a positive bias with
regards to salaries earned and rents paid. By the same logic, the survey data could understate the
prevalence of room sharing and group renting.

83
In line with some major city’s guidelines, I define group renting as sharing the room with more people than is legally
allowed. A maximum room occupancy of two people was precedented by ordinances in Shanghai and Beijing (Beijing
Municipal Commission of Housing and Urban-Rural Development 2011; Shanghai Municipal Government, 2011).
Room sharing includes all who reported living in one room with one or more other persons.

112
4.3.2. Tier 1 Cities Are Different
To begin exploring Shanghai in comparison to other Chinese cities, I disaggregate the survey data
by city type (Table 8). I created additional dummy variables indicating whether respondents
worked in Shanghai, another Tier 1 city, any Tier 1 city, or a non-Tier city for their first job and
used these variables to separate the dataset. This allows me to first assess the situation of those
with their first job in Shanghai versus other Tier 1 cities, and then compare the data on Tier 1
versus non-Tier 1 city respondents. The ethnographic research identified four key variables of
interest: hukou, education, income, and housing. The comparison focuses on key data points within
these categories.
Among those working in Shanghai for their first job, more than half (53%) hold a Tier 1
city hukou in comparison to slightly less than half (49%) in other Tier 1 cities. Note also that the
share of those with degrees from elite institutions is slightly lower in Shanghai.
84
Salaries earned
are slightly higher with roughly one third making more than 5,000 RMB per month, versus roughly
one fourth in other Tier 1 cities. Rents paid, too, are slightly higher in Shanghai. In Shanghai, the
majority (55%) paid more than 1,000 RMB per month, versus 42% paying rent in this price range
working in other Tier 1 cities. All differences are significant at least at the 5% level.
Comparing housing arrangements in Shanghai to other Tier 1 cities, the small difference
in shares of group renters and room sharers is not statistically significant. Note, however, the share
of those with family provided housing, which is almost twice as high in Shanghai as compared to
other Tier 1 cities. This difference is statistically significant at the 1% level.
85

84
This could be a sampling effect since we were able to supplement online distribution of the survey with offline
surveying in Shanghai only.
85
The complete reporting of all difference in means tests can be found in Appendix C, Section 1.

Table 8: Descriptive Statistics by City Type
All Shanghai Other Tier 1 Tier 1 City Non-Tier 1 City
Demographics
Age 25.1 26.07 24.79 25.07 25.11
Gender (F=0, M=1) 0.42 0.54 0.54 0.54 0.59
Tier 1 City Hukou 0.17 0.53 0.49 0.49 0.05
Education
Educational Attainment
0= No Higher Education
1= Vocational School
2= Bachelor’s Degree
3= Master’s Degree
1.8 1.88 2.00 1.98 1.73
Bachelor Or Above 0.67 0.74 0.74 0.74 0.64
Elite Education 0.54 0.59 0.65 0.64 0.51
Salary
Salary (<5,000 RMB) 0.82 0.68 0.74 0.72 0.86
Salary (<3,000 RMB) 0.40 0.24 0.23 0.23 0.46
Trial Period 0.91 0.93 0.94 0.94 0.90
Housing
Rent (<1,000 RMB) 0.71 0.45 0.58 0.56 0.77
Rent (<600 RMB) 0.43 0.25 0.32 0.30 0.48
Group Renting 0.21 0.21 0.23 0.23 0.21
Sharing A Room 0.79 0.76 0.78 0.78 0.80
Family Support
Family Provided Housing 0.12 0.12 0.07 0.08 0.14
Family Financial Support 0.64 0.65 0.66 0.66 0.64

113

114
In sum, those working in Shanghai for their first job seem to earn higher wages, but also
pay higher rents. Although a lower share of respondents reported group renting and room sharing
in Shanghai, this difference is not statistically significant. Shanghai appears to be pointedly
different from other Tier 1 cities only with regard to the high share of those relying on their families
to provide housing.
Table 9: Cross-Tabulation of First Job Location and Local Hukou

Shanghai Hukou

Beijing Hukou
First Job SH 0 1 Total First Job BJ 0 1 Total
0 14,334 115 14,449 0 12,693 367 13,060
1 463 432 895 1 1,031 1,253 2,284
Total 14,797 547 15,344 Total 13,724 1,620 15,344

Guangzhou Hukou

Shenzhen Hukou
First Job GZ 0 1 Total First Job SZ 0 1 Total
0 14,673 79 14,752 0 14,891 59 14,950
1 423 169 592 1 306 88 394
Total 15,096 248 15,344 Total 15,197 147 15,344

To investigate this anomaly, I pairwise cross-tabulated the location of first jobs with local
hukou for all Tier 1 cities. Shanghai’s difference in housing arrangements might be owed to a
higher share of respondents with local hukou, who can rely on family for housing. As Table 9
shows, the share of local hukou holders among those who first started working in Shanghai (48.4%)
is perhaps roughly comparable to Beijing (54.8%), but much higher than in Guangzhou (28.4%)
and Shenzhen (22.3%). Yet, the share of respondents with family provided housing was between
6% and 7% for Beijing, Guangzhou, and Shenzhen. More analysis is needed to make sense of the
higher share of those with family provided housing in Shanghai. Importantly, for the purpose of
this analysis, the difference appears not to be driven by share of local hukou holders.

115
Next, I discuss how respondents in Tier 1 cities compare to those in non-Tier 1 cities. Note
first the difference in hukou status. While in Tier 1 cities, almost half (49%) of respondents held a
Tier 1 city hukou, only 5% of respondents working their first job in non-Tier 1 cities held a Tier 1
city hukou (Table 8). The disaggregate data displayed in Table 9 further shows that in each of the
Tier 1 cities, those with local hukou, taking their first job locally outstrip those migrating for their
first job. In fact, only one fifth of those with Tier 1 hukou take their first job in a non-Tier 1 city
(Table 10).
86
Conversely, only 16% of respondents with non-Tier 1 city hukou migrate to a Tier 1
city for their first job. More analysis is needed but this pattern appears (alarmingly structural). It
could be indicative of two mechanisms at work: (1) localized social capital disincentives Tier 1
city hukou holder from migrating, and/ or (2) internal migration is stratifying in China.
Table 10: Cross-Tabulation of First Job Location and Hukou
First Job: Tier 1 City First Job: Non-Tier 1 City
Hukou: Tier 1 City 2,058 504
Hukou: Non- Tier 1 City 2,107 10,675

Respondents working in Tier 1 cities on average have higher levels of education. Compared
to respondents working in non-Tier 1 cities, a higher share of those taking their first job in Tier 1
cities reported having a bachelor’s degree or above (74% vs. 64%), and among those with higher
education there were more graduates from elite institutions (64% vs. 51%). This pattern could
indicate skill-based sorting, which has been identified as a trend in the dynamic distribution of
workers in the U.S. (Diamond, 2016; Moretti, 2004; 2012).

86
The share is lowest among those with Shanghai hukou (14%), and highest among those with Shenzhen hukou (27%).

116
In terms of salaries, those working in Tier 1 cities generally earned higher salaries, with
higher shares on the top end and lower shares on the low end of the monthly income distribution.
The share of those earning 5,000 RMB and more was twice as high in Tier 1 cities (28%) versus
non-Tier cities (14%). The opposite is true for those earning less than 3,000 RMB: the stare was
23% in Tier 1 cities versus 46% in non-Tier 1 cities. Note also the 4% difference in respondents
reporting having a trial period during their first job. All differences are statistically significant at
the 1% level. This finding is in line with a large urban economics literature, which identifies a
positive linear relationship between individual earnings and city size; with positive human capital
externalities as one of the density-dependent driving forces in higher productivity, which in turn
are reflected in higher earnings (Combes, Duranton & Gobillon, 2008; De la Roca & Puga, 2017;
Glaeser & Mare, 2001; Glaeser, 2011; Moretti, 2012; Rauch, 1993).
Together with earning higher salaries, respondents also paid higher rents in Tier 1 cities:
45% reported paying 1,000 RMB and more on rent in Tier 1 cities – in comparison to only one
third in non-Tier 1 cities. Those working in Tier 1 cities also display a slightly higher share of
group renters (23% vs. 21%), but a lower share of room sharers (78% vs. 80%). A substantially
higher share relied on their families for supplying housing in non-Tier 1 cities versus Tier 1 cities.
All differences are statistically significant, at least at the 5% level.
87
Further analysis is needed to
uncover the role of local hukou in non-Tier 1 cities as a potential force for this difference.
In sum, the disaggregated data for Tier 1 versus non-Tier cities suggests that respondents
in Tier 1 cities face more competitive environments: they had higher levels of education, higher
salaries, but also trial periods were more common and housing conditions worse – higher rents,

87
The complete reporting of all difference in means tests can be found in Appendix C, Section 2.

117
more group renting, and fewer relied on family provided housing. Furthermore, the data indicate
how hukou is influencing migration patterns: those with Tier 1 city hukou are disproportionally
working in a Tier 1 city for their first job and most do not migrate for their first job, while only a
fraction of those without Tier 1 city hukou migrate to a Tier 1 city for their first job.
4.3.3 Group Renters Are Mostly Non-Elite Graduates
Recall that 21.2% of survey respondents reported living in group rentals during their first job after
graduation. This share is slightly higher in Tier 1 cities (23.3%) and the difference is statistically
significant, suggesting that group renting could be more common in Tier 1 cities (Table 8).
Table 11: Number of Group Renters by First Job Location, Education, and Hukou
Shanghai Elite
Education
Non-Elite
Education
No Higher
Education
Total
Hukou: Tier 1 City 23 16 10 49
Hukou: Non-Tier 1 City 25 39 20 84
Total 48 55 30 133

Tier 1 City Elite
Education
Non-Elite
Education
No Higher
Education
Total
Hukou: Tier 1 City 123 103 35 261
Hukou: Non-Tier 1 City 153 161 87 401
Total 276 264 122 662

Non-Tier 1 City Elite
Education
Non-Elite
Education
No Higher
Education
Total
Hukou: Tier 1 City 27 31 4 62
Hukou: Non-Tier 1 City 502 645 345 1,492
Total 529 676 349 1,554

Table 11 shows cross-tabulations – for group renters only – of educational attainment and
Tier 1 hukou status by first job location. The results are in line with those reported in chapters 2
and 3: No matter where respondents worked for their first job, those without Tier 1 city hukou and

118
without degrees from elite higher education institutions represent the largest share among group
renters. Note also that the share of elite graduates is the second largest share in all subgroups.
Among respondents in Tier 1 cities, elite graduates actually make up for the largest group (42%)
among bed space renters, when hukou is not considered. Recall, however, the likely oversampling
on the top end of education distribution, which could be a source of bias.
In summary, this chapter draws on survey data collected via a cell phone social media
survey in 2019 to probe whether group renting in Shanghai is merely a regional anomaly or in fact
a case study of a larger phenomenon. I find regardless of city type, about one fifth of survey
respondents reported living in group rentals while working in their first job. Furthermore, job
market entrants in Shanghai report living and working environments comparable to those in other
Tier 1 cities. Those working in Tier 1 cities, however, face more competitive environments than
those in non-Tier 1 cities, including higher salaries, greater prevalence of trial periods and worse
housing conditions.

119
CHAPTER 5. SUMMARY AND REFLECTIONS
5.1. Summary
In this dissertation, I investigate the hidden, informal group rental housing situation in Shanghai,
China. Based on remotely collected online advertisement data, three summers of fieldwork, a cell
phone social media survey, and a total of 29 months spent over several years living, studying, and
working in China, I employ a mix of methods to piece together an in-depth understanding of group
renting not just as a housing, but as a social phenomenon. Towards this end, I ask five research
questions in service of one overarching question: What does the emergence of group renting tell
us about the changing mechanics of social mobility and contemporary inequality in urban China?
First, I approach group renting with the tools of housing market analysis to understand the
target tenant demographic and the renters’ revealed preferences. Using groundtruthed online data,
I find that at its core this informal housing sub-market is driven by the fundamental housing market
trade-off: space versus access to jobs and opportunities. Here, this calculus is pushed to the extreme
as tenants minimize the space consumed to the renting of merely a bed. Evidence also strongly
suggests a sellers’ market, including landlords’ very detailed communication – by means of rental
listings – with a clearly delimited target renter group of recent (educated) migrants. This is a
demographic of migrants understudied in the literature because it has recently emerged after
national pushes in higher education and urbanization. College education of varying quality has
increased dramatically but hukou still constrains both educational and labor opportunities. I find
that unlike less educated migrants of the previous decade who lived in urban villages and
dormitories on the periphery, these recent migrant, entry-level white collar workers make
sacrifices to be in the city center.

120
I also draw on participant observation and interviews in order to delve deeper into the
socio-cultural context around group renting. In particular, this second, vastly different part
analyzes motives and cognitive processing. I find that beyond providing affordable access to
downtown employment centers, group rentals function as a time-space strategy for newcomers to
Shanghai. These recent migrants need to keep expenses low, either to be able to send money back
to their families or as a stepping stone until they achieve a more secure footing in Shanghai’s
highly competitive labor market. Whether because of remittances or gaining a foothold, group
rentals as incubators for city starters are part of a longer-term, multi-generational social mobility
strategy. In China, social norms require adult children to take care of their aging family members.
The responsibility weighs heavily because of the country’s rapid demographic transition,
accelerated by the One-Child policy, in the absence of sufficient government support for old age.
Moving to Shanghai is these migrants’ bid for upward mobility and financial security. Group
renting is what affords them a shot at a high-paying career.
Importantly, hukou and socio-economic background emerge as structuring variables across
the various life trajectories of my interviewees. Because family resources, but mostly hukou
determine access to quality education, those from rural and/ or more modest backgrounds
disproportionally earn degrees from second-tier institutions. The quality difference is reflected in
labor market opportunities and compensation, which also maps onto housing market outcomes.
The same factors – hukou and essentially class – also mediate the weight of the unspoken
intergenerational contract. Parents’ coverage for old age is a function of class and hukou status
and effectively determines when children are expected to start remitting money and their degree
of dependence. This responsibility in turn influences risk aversion in the labor market and
consumption choices, including housing.

121
Group renters process their experience by drawing on socially constructed narratives
around “sacrificing” and “personal responsibility.” Focusing on purpose and temporariness, they
are able to make sense of group renting also because it is normalized through the constant social
reproduction of these narratives. Aside from government and media communication, their family’s
experience is a critical frame of reference: the lived experiences of growth and relative
permeability of social hierarchies continues to bolster this generation’s belief in social mobility
and, for the moment, thwarts class-based resentment even in the face of structural disadvantage.
Finally, I turn to the question of whether investigating group rentals in Shanghai makes for
a legitimate case study of a broader phenomenon across urban China. To this end, I draw on a
large-scale cell phone social media survey which collected data on demographics as well as the
employment and housing trajectories of (recent) job market entrants throughout urban China. I
find that Shanghai is not a regional anomaly but a specific instance of a broader occurrence:
regardless of where respondents took their first job, one out of five rented a bed after graduation
and those with post-secondary degrees from non-elite institutions and non-Tier 1 city hukou
consistently made up for the largest share among group renters.
Furthermore, there is suggestive evidence of sorting. Job market entrants in Shanghai
appear roughly comparable to those in other Tier 1 cities, but descriptive statistics suggest a
statistically significant difference between the reported characteristics of those starting their jobs
in Tier 1 versus non-Tier 1 cities: higher salaries, greater prevalence of trial periods, and worse
housing conditions. Importantly, hukou appears to be structuring this sorting process. This adds
another layer to the way in which hukou and class are creating structural disadvantage in
contemporary urban China (Figure 6).

122
Figure 6: Hukou and Class - Structural Disadvantage for Educated Migrants

5.2. Reflections and Open Questions
In many ways, this dissertation has opened up more questions than it has addressed. In this final
section, I reflect on some of these areas for further deliberation with particular attention paid to
implication for planning in China and beyond.
5.2.1 Hidden Informality and Planning
This study adds to the discourse on contemporary informality by adding an in-depth study on one
instance of informal use, rather than informal land occupation or informal structures.
Informal use amidst formal spaces is possible because it happens behind closed doors and
is thus visually hidden. Informal uses can stay hidden as the result of strategic choices and ongoing
negotiations on the suppliers’ side. As touched upon in chapter 3.6, landlords engage in stealth
strategies but also bribe local stakeholders (low-level administrators, neighbors, compound
security). Group rental landlords also target compounds with large vacancy rates to evade social

123
enforcement and to take advantage of economies of scale on their investments. While this
dissertation focuses on group renters and their experiences, looking at the supply side of group
renting as a housing phenomenon reveals that necessity of invisibility involves a social element.
The case study of group rentals thus invites a broader inquiry into the role of social relations in
sustaining housing informality: When are informal uses tolerated? What are the necessary
conditions for social buy-in?
Another question this dissertation only begins to address is the question of why informality
manifests in this hidden form in contemporary urban China. I conjecture that the hidden nature is
the product of our current moment of urban development. On the one hand, urbanization amidst a
turn to a post-industrial system of production means that density-dependent knowledge and service
economies continue to drive more and more people cityward. At the same time, the intensifying
competition for urban space is fueled by concurrent increases in financialization of urban space
and clamping down in terms of property rights. As urban space is less freely accessible,
undercapitalized newcomers are left to find interstitial spaces – in built up urban centers this means
increasing the intensity of use of existing, formal spaces.
Importantly, this current moment in urban development is seemingly shared among
growing cities globally, and so is the response. I believe that group renting is part of broader
housing trend. In growing cities around the world, people have been negotiating new forms of
sharing living space. Higher density not only in units per land area but in people per unit no longer
is the exclusive realm of rapidly urbanizing cities of the Global South or the urban poor. Recent
scholarship reports people sharing housing among wider demographics and over longer periods of
time. In Mumbai but also in New York, San Francisco, and Los Angeles, property developers are
starting to experiment with “high-end co-living” and “dorms for grow-ups” but, for the most part,

124
house sharing is organized outside of formal markets and planning solutions. The case of bed space
rentals, although an extreme example uniquely shaped by China’s political economy, may be part
of a global shift in how urban space is consumed, where, and by whom.
Given these broad-scale developments that appear to transcend socio-cultural contexts and
historical development trajectories, what are the planning implications? Where can (or should)
planning engage and how?
Planning cannot solve the many challenges brought about by the historic shift.
Demographic transition and urbanization in the age of post-industrialization require
comprehensive policy measures beyond what spatial planning can offer. For the case of group
rentals in China, for instance, my dissertation points towards supporting economic development
planning outside of Tier 1 cities, and reforming access to education and the pension system as
obvious, broad-stroke areas of policy intervention that would help the group renters’ situation.
Generally, what planning can do is to strive to better negotiate the intensifying struggle for
space and advocate for more equitable solutions. Instead of thinking about informality as
regulatory violations that planners ought to correct, planning could take a learning approach:
studying informal solutions and using this knowledge about unmet housing needs could help in
devising more effective planning responses.
In practice, this could mean a fresh approach to small space subsidized rental housing. The
dissertation highlights the trading of personal space for locational access as the central calculus
behind group renting. As private developers are already devising smaller unit size projects in prime
locations catering to high-end customers, city planners could think creatively about ways to ensure
that access to opportunity does not become merely a function of purchasing power. What could a

125
public effort in the realm of micro units and formal house sharing, that goes beyond the socially
stigmatized single room occupancy housing, look like?
Recall also that I find that group renting is affording access to opportunity – but specifically
in service of the longer-term goal of upward mobility. In a country with vast regional income and
wage differences, group renters move to Shanghai and take jobs that barely cover their financial
needs in order to get a shot at a high-paying white collar career. Group rentals informally afford
them a temporary foothold on their social mobility strategies. But what could formal spatial
planning for mobility pathways look like? How should planning engage with cities becoming
“learning centers”? And what is the education-migration nexus telling us about the possibility of
inclusive urbanization? What if “the right to the city” is increasingly linked to education?
Not a focal point in this dissertation, but a more subtle common thread is the role of gender.
Most of my friends and the majority of my more in-depth conversations were with women. Also,
my own experience of group renting was in an all-female shared apartment. Through interviewing
and participant observation I find that women commonly experience additional constraints in
accessing group renting as an affordable housing strategy. Landlords voiced strong preferences for
young, male tenants and only a fraction of online advertisements listed beds in all female
apartments, the preferred option among most women. I also found indication that the perception
of risk structures the set of acceptable and attainable housing solutions for women. If sharing
housing is about accessing economic opportunity and this access is gendered, then the absence of
planning engagement helps reproduce gendered disparities. Given that informal house sharing is a
global trend, my dissertation findings invite more research to understand the operation of gender
in shared housing markets across different geographic and cultural contexts, including the less
crowded shared arrangements that have been documented in the Global North. More broadly, the

126
intersection of gender and informality, extensively studied for the realm of work, has received little
scholarly attention in the realm of housing. As planner turn to the shared housing challenge, what
could a gender-conscious planning approach look like?
Finally, the politics of visibility deserve closer planning consideration. In this dissertation,
online data made the hidden group rental visible for me as a researcher. The availability of new
data sources has sparked excitement in the planning community as new data approaches promise
to reveal more of what is going on in our cities and the ability to stay in touch with fast-changing
urban conditions. This is especially true for the urbanism of the marginalized who seldomly show
up in officially collected data but for whom new data mean the potential for representation.
Similarly, rental housing has been perennially understudied, in part because even formal rental
agreements are difficult to trace. As this dissertation demonstrates, drawing on new data sources
such as online advertisements can be crucial in identifying hidden urban phenomena, but the
picture is often incomplete. Social inequity and politics in the material world also shape the
geography of the digital world. Critical engagement and an awareness of blind spots, biases, and
the dangers of visibility will be crucial as planning practice and research leverage online data
sources to address planning challenges.
5.2.2 Informality and Inequality
The overarching research question of this dissertation asked what the group rental housing might
be telling us about changing mechanisms of social mobility and contemporary inequality in urban
China. A related question that I have addressed only partially is why there is an informal response
to this housing need in the first place. It has been argued that informality is often just the visible
symptom of deeper-rooted social inequalities and my dissertation research offers up evidence that

127
this is indeed the case for group rentals. Here, I want to briefly throw a light on two other potential
explanations that have been discussed in the literature: market failure and deliberate government
inaction.
During my dissertation research I found some indication that group rentals may be a
transitory institutional response as formal markets and policy making have not yet caught up to
the housing needs of educated migrants. Indeed, both educated migrants and the supply surge in
skilled workers more broadly are relatively new phenomena in urban China.
Over the course of my dissertation research I also witnessed the group rental market in
Shanghai evolving. When I did the market survey in 2016, group rentals could be found in virtually
any compound in central Shanghai and I saw a variety of actors eager to enter a market clearly
governed by extreme demand overhang. By 2018, when I returned, it was already much more
difficult to find group rentals. The category “bed space for rent” had been banned from online
advertising platforms and in the much fewer listings that remained landlords were communicating
in a decisively more veiled manner. Starting in late 2017, Shanghai had started several policy
campaigns under the umbrella of “population control” that included the crackdown of illegal
dwellings, mostly affecting migrants. At the same time, the municipal government had begun to
aggressively promote and incentivize the development of a formal high-density rental housing
market; Professionally managed subdivided units and single occupancy private rental housing has
since started to pop up all the city. But even with the city investing in the development of this new
kind of rental housing, who is going to be able to rent these? The dissertation findings compel the
question of whether there are real social mobility possibilities for those who don’t have hukou and
who are from non-elite schools. Does this new rental housing work to “make space” for these less
privileged job market entrants? What could be done to help them be successful?

128
Can group renting be understood as the symptom of deliberate government inaction? In
light of skill-based sorting, municipal governments all over the world are faced with the same
dilemma: they want to attract key workers, but, at the same time, they know that wages and house
prices are so out of balance that there is no formal market solution for them. The “solution” is then
often turning a blind eye and leaving workers to find housing arrangements informally.
In China, institutional differentiation between locals and migrants in the form of hukou
adds another layer to this issue. It has been argued that hukou can be understood as an internal
passport system. And indeed, the way urban governments are treating internal migrants is akin to
the handling of foreign workers: municipal government use their power strategically to pick and
choose who is able to live and work in the contemporary Chinese city. As I demonstrate in my
dissertation, for the group rental demographic hukou has both direct and indirect, but no less
powerful effects; I find that hukou affects their education opportunities, migration destinations,
relative familial responsibility burden – and as a result their employment and housing trajectories.
Importantly, education does not override the institutional disadvantage. If it is still hukou
that ultimately matters most, what does this mean for the China’s growth project and present-day
social contract?
Finally, my dissertation also evidences the crucial role of cities in striving for social
mobility and therefore the centrality of housing in cities. But I also find that access to cities is
structured by hukou and education. What are the implications of this evolving migration-education
nexus for the future of urban China? What can we learn more broadly about “a new right to the
city”?

129
REFERENCES

Ainsworth, J. W. (2002). Why does it take a village? The mediation of neighborhood effects on
educational achievement. Social forces, 81(1), 117-152.

Alonso, W. (1960). A theory of the urban land market. Papers in Regional Science, 6(1), 149-157.

Anagnost, A. (2004). The corporeal politics of quality (suzhi). Public culture, 16(2), 189-208.
Anagnost, A. (2008). From ‘Class’ to ‘Social Strata': grasping the social totality in reform-era
China. Third World Quarterly, 29(3), 497-519.
Anderson, C. (2008). The end of theory: The data deluge makes the scientific method obsolete.
Wired magazine, 16(7), 16-07.
AlSayyad, N., & Roy, A. (Eds.). (2003). Urban Informality: Transnational Perspectives from the
Middle East, Latin America, and South Asia. Lexington Books.
Appleton, S., Song, L., & Xia, Q. (2014). Understanding urban wage inequality in China 1988–
2008: Evidence from quantile analysis. World Development, 62, 1-13.
Bacolod, M., Blum, B.& Strange, W. (2009) Skills in the city, Journal of Urban Economics, 65,
pp. 136–153.

Bai, L. (2006). Graduate unemployment: Dilemmas and challenges in China's move to mass higher
education. The China Quarterly, 185, 128-144.
Banister, J., Bloom, D. E., & Rosenberg, L. (2012). Population aging and economic growth in
China. In The Chinese economy, pp. 114-149. Palgrave Macmillan, London.
Bayer, P., Ferreira, F., & McMillan, R. (2007). A unified framework for measuring preferences
for schools and neighborhoods. Journal of political economy, 115(4), 588-638.
Beijing Municipal Commission of Housing and Urban-Rural Development (2011). . Available at:
http://bjjs.zjw.beijing.gov.cn/bjjs/xxgk/gfxwj/zfcxjswwj/315934/index.shtml. (last
accessed: 05/08/2020).
Bennardo, G., & Munck, V. C. D. (2014). Cultural models: Genesis, methods, and experiences.
Oxford: Oxford University Press.
Beracha, E., & Wintoki, M. B. (2013). Forecasting residential real estate price changes from online
search activity. Journal of Real Estate Research, 35(3), 283-312.
Bian, Y. (2002). Chinese social stratification and social mobility. Annual review of
sociology, 28(1), 91-116.

130
Bian, Y., & Logan, J. R. (1996). Market transition and the persistence of power: The changing
stratification system in urban China. American Sociological Review, 739-758.
Biao, X. (2011). A ritual economy of ‘talent’: China and overseas Chinese professionals. Journal
of Ethnic and Migration Studies, 37(5), 821-838.
Blake, K. S., Kellerson, R. L., & Simic, A. (2007). Measuring overcrowding in
housing. Washington, DC: Department of Housing and Urban Development, Office of
Policy Development and Research. Available at:
https://www.huduser.gov/publications/pdf/measuringovercrowding_in_hsg.pdf. (last
accessed: 04/10/2020).
Boeing, G. (2020). Online rental housing market representation and the digital reproduction of
urban inequality. Environment and Planning A: Economy and Space, 52(2), 449-468.
Boeing, G., & Waddell, P. (2017). New insights into rental housing markets across the United
States: Web scraping and analyzing craigslist rental listings. Journal of Planning
Education and Research, 37(4), 457-476.
Booth, A. (1976). Urban crowding and its consequences. Praeger Publishers.
Boudon, R. (1974). Education, opportunity, and social inequality: Changing prospects in western
society.
Bourdieu, P. (1977). Outline of a Theory of Practice. Cambridge University Press.
Bourdieu, P. (1986). The forms of capital. In Richardson, J.G. (Eds.) Handbook of theory and
research for the sociology of education. Greenwood Press: New York, 241– 58.
Boyd, M. (1989). Family and personal networks in international migration: recent developments
and new agendas. International migration review, 23(3), 638-670.
Boyd, D., & Crawford, K. (2012). Critical questions for big data: Provocations for a cultural,
technological, and scholarly phenomenon. Information, communication & society, 15(5),
662-679.
Brandt, L., & Rawski, T. G. (Eds.). (2008). China's great economic transformation. Cambridge
University Press.
Cai, Q. (2003). Migrant remittances and family ties: a case study in China. International Journal
of Population Geography, 9(6), 471-483.

Cai, Y., & Cheng, Y. (2014). Pension reform in China: challenges and opportunities. Journal of
Economic Surveys, 28(4), 636-651.
Campagna, G. (2016). Linking crowding, housing inadequacy, and perceived housing stress.
Journal of Environmental Psychology, 45, 252–266.

131
Cao, S. (2017). Beyond the Celebration of Losers: The Construction of diaosi Masculinity in
Contemporary Chinese Youth Culture. In East Asian Men (pp. 219-235). Palgrave
Macmillan, London.
Cebolla-Boado, H., Hu, Y., & Soysal, Y. N. (2018). Why study abroad? Sorting of Chinese
students across British universities. British Journal of Sociology of Education, 39(3), 365-
380.
Central Government of the People’s Republic of China (2010). (2010 2020 ) . 2010-06-06. Available at: http://www.gov.cn/jrzg/2010-
06/06/content_1621777.htm. (last accessed: 05/03/2020).
Chan, K.W. (1996) Post‐Mao China: a two‐class urban society in the making. International
Journal of Urban and Regional Research 20, 134– 50.
Chan, K. W. (2010). Fundamentals of China's urbanization and policy. China Review, 63-93
Chan, K. W. (2013). China: internal migration. The encyclopedia of global human migration.
Chan, W. K. (2015). Higher education and graduate employment in China: Challenges for
sustainable development. Higher education policy, 28(1), 35-53.
Chen, A.P. (2017). : 2016 5000 . [Shanghai: The average
monthly income of migrant workers in 2016 exceeded 5,000 yuan]. Xinhua. 2017-05-27.
Available at: http:// www.xinhuanet.com//politics/2017-05/27/c_11 21049739.htm (last
accessed: 09/09/2019).
Chen, Y. (2011). Occupational attainment of migrants and local workers: Findings from a survey
in Shanghai’s manufacturing sector. Urban Studies, 48(1), 3-21.

Chen, Y., & Feng, S. (2013). Access to public schools and the education of migrant children in
China. China Economic Review, 26, 75-88.
Chan, W. K., & Ngok, K. (2011). Accumulating human capital while increasing educational
inequality: a study on higher education policy in China. Asia Pacific Journal of Education,
31(3), 293-310.
Chen, J., & Hao, Q. (2010). Submarket, heterogeneity and hedonic prediction accuracy of real
estate prices: Evidence from Shanghai. International Real Estate Review, 13(2), 190-217.
Chen, J., Hao, Q., & Stephens, M. (2010). Assessing housing affordability in post-reform China:
A case study of Shanghai. Housing Studies, 25(6), 877-901.
Chen, S., Jefferson, G. H., & Zhang, J. (2011). Structural change, productivity growth and
industrial transformation in China. China Economic Review, 22(1), 133-150.

132
Chetty, R., Hendren, N., Kline, P., & Saez, E. (2014). Where is the land of opportunity? The
geography of intergenerational mobility in the United States. The Quarterly Journal of
Economics, 129(4), 1553-1623.
Cheung, A. K. L., & Yeung, W. J. J. (2015). Temporal-spatial patterns of one-person households
in China, 1982–2005. Demographic Research, 32, 1209-1238.
Chiang, S. Y. (2018). The Narrative Construction of Fang-Nu ( ): An Urban Identity in Post-
Modern China. In Dynamics of Community Formation (pp. 119-133). Palgrave Macmillan,
New York.
China Internet Network Information Center, CNNIC (2018). Statistical Report on Internet
Development in China. Available at:
https://cnnic.com.cn/IDR/ReportDownloads/201807/P020180711391069195909.pdf.
(last accessed: 01/24/2019).
China News (2014). . China News. 2014-
07-18. Available at: http://www.chinanews.com/gn/2014/07-28/6429768.shtml. (last
accessed: 05/12/2020).
Chinese Communist Party (CCP) News Net (2018). . 2018-5-11. Available at:
http://cpc.people.com.cn/xuexi/n1/2018/0511/c385474-29979061.html. (last accessed:
04/20/2020).
Choi, S. Y. (2016). Children and migration in China. Asian Population Studies, 12(2), 131-134.
Chu, L. W., & Chi, I. (2008). Nursing homes in China. Journal of the American Medical Directors
Association, 9(4), 237-243.
Churchill, S.A. & Mishra, V. (2018) Returns to education in China: a meta-analysis, Applied
Economics, 50(54), 5903-5919.

Clapham, D., Mackie, P., Orford, S., Thomas, I., & Buckley, K. (2014). The housing pathways of
young people in the UK. Environment and Planning A, 46(8), 2016-2031.
Clark, V., Tuffin, K., Frewin, K., & Bowker, N. (2017). Shared housing among young adults:
avoiding complications in domestic relationships. Journal of youth studies, 20(9), 1191-
1207.
Cole, E. R., & Omari, S. R. (2003). Race, class and the dilemmas of upward mobility for African
Americans. Journal of Social issues, 59(4), 785-802.
Combes, P. P., Duranton, G., & Gobillon, L. (2008). Spatial wage disparities: Sorting matters!.
Journal of urban economics, 63(2), 723-742.

133
Combes, P. P., Démurger, S., Li, S., & Wang, J. (2020). Unequal migration and urbanisation gains
in China. Journal of Development Economics, 142, 102328.
Conley, D. (2001). A room with a view or a room of one's own? Housing and social stratification.
Sociological Forum, 16(2), 263-280.
Cook, S. (2002). From rice bowl to safety net: Insecurity and social protection during China’s
transition. Development Policy Review, 20(5), 615-635.
Crossley, N. (2012). Social class. In Grenfell, M. (Ed.s). Pierre Bourdieu: Key Concepts, Second
Edition. Durham: Acumen, 85-97.
Cui, C., Geertman, S., & Hooimeijer, P. (2014). The intra-urban distribution of skilled migrants:
Case studies of Shanghai and Nanjing. Habitat International, 44, 1-10.
Cui, C., Geertman, S., & Hooimeijer, P. (2016). Access to homeownership in urban China: A
comparison between skilled migrants and skilled locals in Nanjing. Cities, 50, 188-196.
Cui, C., Deng, W., & Lu, T. (2019). Pathways to homeownership in urban China: Transitions and
generational fractures. Journal of Housing and the Built Environment, 1-19.
D'Andrade, R. G. (1981). The cultural part of cognition. Cognitive science, 5(3), 179-195.
D’Andrade, R. G. (1992). Schemas and motivation. In: D’andrade & Strauss (ed.). Human Motives
and Cultural Models, pp. 23-44.
D'Andrade, R. G. (1995). The development of cognitive anthropology. Cambridge University Press.
Das, P., Ziobrowski, A., & Coulson, N. E. (2015). Online information search, market fundamentals
and apartment real estate. The journal of real estate finance and economics, 51(4), 480-
502.
Davin, D. (1998). Internal migration in contemporary China. Springer.
De, H. H. (2010). Migration and Development: A Theoretical Perspective. The International
migration review, 44(1), 227-264.
De la Roca, J. & Puga, D. (2017). Learning by working in big cities. The Review of Economic
Studies, 84(1), 106-142.
De Soto, H. (1989). The Other Path: The Invisible Revolution in the Third World. London IB:
Tauris.
De Soto, H. (2000). The mystery of capital: Why capitalism triumphs in the West and fails
everywhere else. New York: Basic Books.

134
Department for Communities and Local Government (2014). English Housing Survey: Headline
Report. Available at:
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/284648/En
glish_Housing_Survey_Headline_Report_2012-13.pdf. (last accessed: 03/15/2017).
Diamond, R. (2016). The determinants and welfare implications of US workers' diverging location
choices by skill: 1980-2000. American Economic Review, 106(3), 479-524.
Ding, L. (2002). Rural–urban income disparity: impact of growth, allocative efficiency, and local
growth welfare. China Economic Review, 13(4), 419-429.
Ding, X. (2019). Ding, X. (2019). College Education and Internal Migration in China. University
of Kentucky Working Paper. Available at:
http://gatton.uky.edu/sites/default/files/EconomicsWorkshop/Ding_College%26Migratio
n.pdf. (last accessed: 05/08/2020).
Du, Y., Park, A., & Wang, S. (2005). Migration and rural poverty in China. Journal of comparative
economics, 33(4), 688-709.

Dunn, J. R. (2002). Housing and inequalities in health: a study of socioeconomic dimensions of
housing and self reported health from a survey of Vancouver residents. Journal of
Epidemiology & Community Health, 56(9), 671-681.
Dunn, J. R., & Hayes, M. V. (2000). Social inequality, population health, and housing: a study of
two Vancouver neighborhoods. Social science & medicine, 51(4), 563-587.
Durst, N. J., & Wegmann, J. (2017). Informal housing in the United States. International Journal
of Urban and Regional Research, 41(2), 282-297.
Edelman, B. G., & Luca, M. (2014). Digital discrimination: The case of Airbnb. com. Harvard
Business School NOM Unit Working Paper, (14-054).
Eeckhout, J., Pinheiro, R., & Schmidheiny, K. (2014). Spatial sorting. Journal of Political
Economy, 122(3), 554-620.

Ellis, R. A., & Lane, W. C. (1963). Structural supports for upward mobility. American Sociological
Review, 743-756.
Eurostat (2014). Housing Conditions. Available at: http://ec.europa.eu/eurostat/statistics-
explained/index.php/Housing_conditions#Housing_quality_-_overcrowding. (last
accessed: 03/15/2017).
Evans, G. W. (2003). The built environment and mental health. Journal of Urban Health, 80(4),
536-555.
Evans, G. W., & Lepore, S. J. (1992). Conceptual and analytic issues in crowding research. Journal
of Environmental Psychology, 12(2), 163-173.

135
Evans, J. A., & Aceves, P. (2016). Machine translation: mining text for social theory. Annual
Review of Sociology, 42.
Evans, G. W., Wells, N. M., & Moch, A. (2003). Housing and mental health: A review of the
evidence and a methodological and conceptual critique. Journal of Social Issues, 59, 475–
500.
Fairbank, J. K., & Goldman, M. (2006). China: A new history. Harvard University Press.
Fan, C. C. (2002). The elite, the natives, and the outsiders: Migration and labor market
segmentation in urban China. Annals of the Association of American Geographers, 92(1),
103-124.
Fan, C. C. (2007). China on the Move: Migration, the State, and the Household. Routledge.
Feler, L., & Henderson, J. V. (2011). Exclusionary policies in urban development: Under-servicing
migrant households in Brazilian cities. Journal of urban economics, 69(3), 253-272.
Feng, J., He, L., & Sato, H. (2011). Public pension and household saving: Evidence from urban
China. Journal of Comparative Economics, 39(4), 470-485.
Feng, Q., Yeung, W. J. J., Wang, Z., & Zeng, Y. (2019). Age of retirement and human capital in
an aging China, 2015–2050. European Journal of Population, 35(1), 29-62.
Fiske, S. T., & Taylor, S. E. (1991). Social cognition (2nd ed.). New York: McGraw-Hill.
Folch, D. C., Spielman, S. E., & Manduca, R. (2018). Fast food data: Where user‐generated content
works and where it does not. Geographical Analysis, 50(2), 125-140.
Fong, V. L. (2002). China's one‐child policy and the empowerment of urban daughters. American
Anthropologist, 104(4), 1098-1109.
Fong, V. L. (2004). Only hope: Coming of age under China's one-child policy. Stanford University
Press.
Fraser, D. (2000). Inventing oasis: luxury housing advertisements and reconfiguring domestic
space in Shanghai. In: Davis DS (ed.). The consumer revolution in urban China. Berkeley,
CA and Los Angeles, CA: University of California Press, 25-53.
Fu, Q., & Ren, Q. (2010). Educational inequality under China's rural–urban divide: The hukou
system and return to education. Environment and Planning A, 42(3), 592-610.
Gao, W., & Smyth, R. (2015). Education expansion and returns to schooling in urban China, 2001–
2010: Evidence from three waves of the China Urban Labor Survey. Journal of the Asia
Pacific Economy, 20(2), 178-201.
Garner, C. L., & Raudenbush, S. W. (1991). Neighborhood effects on educational attainment: A
multilevel analysis. Sociology of education, 251-262.

136

Ge, W. (1999). Special economic zones and the opening of the Chinese economy: Some lessons
for economic liberalization. World Development, 27(7), 1267-1285.
Geertz, C. (1973). Thick Description: Toward an interpretive theory of culture. In: The
Interpretation of Cultures: Selected Essays. New York, Basic Books, pp. 3-32.
Gibbon, G., & Bell, R.W. (1939). History of the London County Council 1889–1939. New York:
Macmillan Company.
Glaeser, E.L. (2011). Triumph of the City: How Our Greatest Invention Makes Us Richer, Smarter,
Greener, Healthier, and Happier. London: MacMillan.
Glaser, B., & Strauss, A. (1967). The discovery of grounded theory. Weidenfield & Nicolson,
London.
Glaeser, E. and Maré, D. (2001) Cities and skills, Journal of Labor Economics, 19, pp. 316–342.

Goldin, C., & Katz, L. F. (2007). Long-run changes in the US wage structure: narrowing, widening,
polarizing (No. w13568). National Bureau of Economic Research.
Goodman, L. S., & Mayer, C. (2018). Homeownership and the American dream. Journal of
Economic Perspectives, 32(1), 31-58.
Gover, W. R., & Hughes, M. D. (1983). Overcrowding in the household: An analysis of
determinants and effects. New York: Academic Press.
Guihaire, V., & Hao, J. K. (2008). Transit network design and scheduling: A global
review. Transportation Research Part A: Policy and Practice, 42(10), 1251-1273.
Guo, Y. (2012). Classes without class consciousness and class consciousness without classes: the
meaning of class in the People's Republic of China. Journal of Contemporary China,
21(77), 723-739.
Guo, M., Aranda, M. P., & Silverstein, M. (2009). The impact of out-migration on the inter-
generational support and psychological wellbeing of older adults in rural China. Ageing &
Society, 29(7), 1085-1104
Hannum, E., & Wang, M. (2006). Geography and educational inequality in China. China economic
review, 17(3), 253-265.
Hannum, E., Wang, M., & Adams, J. (2008). Urban-rural disparities in access to primary and
secondary education under market reform. One country, two societies, 125-146.
Hanser, A. (2005). The gendered rice bowl: The sexual politics of service work in urban China.
Gender & Society, 19(5), 581-600.

137
Harris, R. (2018). Modes of informal urban development: A global phenomenon. Journal of
Planning Literature, 33(3), 267-286.
Harris, J.R., & Todaro, M. P. (1970). Migration, unemployment and development: a two-sector
analysis. The American economic review, 60(1), 126-142.
Harten, J.G., Kim A.M., & Brazier, J.C. (2020). Real and fake data in Shanghai’s informal rental
housing market: Groundtruthing data scraped from the internet. Urban Studies (online first)
Hartig, T., Johansson, G., & Kylin, C. (2003). Residence in the social ecology of stress and
restoration. Journal of Social Issues, 59(3), 611-636.
Harwit, E. (2017). WeChat: Social and political development of China’s dominant messaging app.
Chinese Journal of Communication, 10(3), 312-327.
Haveman, R., & Smeeding, T. (2006). The role of higher education in social mobility. The Future
of children, 125-150.
He, C., & Zhu, S. (2007). Economic Transition and Industrial Restructuring in China: Structural
Convergence or Divergence?. Post-Communist Economies, 19(3): 317-342.
He, Y., & Mai, Y. (2015). Higher education expansion in China and the ‘ant tribe’ problem. Higher
Education Policy, 28(3), 333-352.
He, C., Chen, T., Mao, X., & Zhou, Y. (2016). Economic transition, urbanization and population
redistribution in China. Habitat International, 51(2), 39-47.
Heath, S., & Kenyon, L. (2001). Single young professionals and shared household living. Journal
of Youth Studies, 4(1), 83-100.
Hesketh, T., & Zhu, W. X. (1997). Health in China: the one child family policy: the good, the bad,
and the ugly. Bmj, 314(7095), 1685.
Hesketh, T., Li, L., & Zhu, X. W. (2005). The effect of China's one-child family policy after 25
years. The New England Journal of Medicine, 353(11), 1171- 1176.
Higginbotham, E., & Weber, L. (1992). Moving up with kin and community: Upward social
mobility for black and white women. Gender & Society, 6(3), 416-440.
Hochstenbach, C., & Boterman, W. R. (2017). Intergenerational support shaping residential
trajectories: Young people leaving home in a gentrifying city. Urban Studies, 54(2), 399-
420.
Hoffman, L. M. (2010). Patriotic professionalism in urban China: Fostering talent. Temple
University Press.
Hogan, B., & Berry, B. (2011). Racial and ethnic biases in rental housing: An audit study of online
apartment listings. City & Community, 10(4), 351-372.

138
Home Affairs Department (HAD), The Government of the Hong Kong Special Administrative
Region (2017). Bedspace apartments: Code of practice. Available at:
https://www.hadla.gov.hk/filemanager/en/docs/code_practicebed-space.pdf (last accessed:
01/08/2019).
Hsieh, H. F., & Shannon, S. E. (2005). Three approaches to qualitative content
analysis. Qualitative health research, 15(9), 1277-1288.
Hu, Y. (2014). 2 90 . New People’s News.
2014-05-01. Available at: http://shanghai.xinmin.cn/tfbd/2014/05/01/24197848.html. (last
accessed: 05/12/2020).
Hu, A. (2017). Providing more but receiving less: Daughters in intergenerational exchange in
mainland China. Journal of Marriage and Family, 79(3), 739-757.
Hu, A., & Hibel, J. (2014). Changes in college attainment and the economic returns to a college
degree in urban China, 2003–2010: Implications for social equality. Social science
research, 44, 173-186.
Hu, A., & Hibel, J. (2015). Increasing heterogeneity in the economic returns to higher education
in urban China. The Social Science Journal, 52(3), 322-330.
Hu, A., & Vargas, N. (2015). Economic consequences of horizontal stratification in postsecondary
education: evidence from urban China. Higher Education, 70(3), 337-358.
Hu, Y., & Coulter, R. (2017). Living space and psychological well-being in urban China:
Differentiated relationships across socio-economic gradients. Environment and Planning
A, 49(4), 911-929.
Hu, L., He, S., Han, Z., Xiao, H., Su, S., Weng, M., & Cai, Z. (2019). Monitoring housing rental
prices based on social media: An integrated approach of machine-learning algorithms and
hedonic modeling to inform equitable housing policies. Land use policy, 82, 657-673.
Huang, Y. (2003). A room of one's own: housing consumption and residential crowding in
transitional urban China. Environment and Planning A, 35(4), 591-614.
Huang, Y., & Clark, W. A. (2002). Housing tenure choice in transitional urban China: a multilevel
analysis. Urban studies, 39(1), 7-32.
Huang, Y., & Tao, R. (2015). Housing migrants in Chinese cities: Current status and policy
design. Environment and Planning C: Government and Policy, 33(3), 640-660.
Huang, Y., & Yi, C. (2015). Invisible migrant enclaves in Chinese cities: Underground living in
Beijing, China. Urban Studies, 52(15), 2948-2973.
Huang, S., Xin, Z., & Hou, J. (2014). The relationship between family background and the chances
to attend higher education and its tendency
China Youth Study , 26(3), 46–52.

139
International Labour Organization, ILO (2016). People’s Republic of China: Universal pension
coverage Available at: https://www.social-
protection.org/gimi/gess/RessourcePDF.action?ressource.ressourceId=51765. (last
accessed: 04/10/2020).
International Labour Organization, ILO (2016). Jobs and skills for youth: review of policies for
youth employment of China. 2016-07-21. Available at:
https://www.ilo.org/employment/Whatwedo/Publications/WCMS_502338/lang--
en/index.htm. (last accessed: 04/13/2020).
Jacka, T. (2009). Cultivating citizens: Suzhi (quality) discourse in the PRC. Positions: 17(3), 523-
535.
Jao, N. (2018). WeChat now has over 1 billion active monthly users worldwide. Technode. 2018-
03-05. Available at: https://technode.com/2018/03/05/wechat-1-billion-users/ (last
accessed: 05/08/2020).
Katz, L.F. & Autor, D.H. (1999). Changes in the wage structure and earnings inequality. In
Handbook of labor economics (Vol. 3, pp. 1463-1555). Elsevier.
Kim, A. M. (2004). A market without the ‘right’ property rights. Economics of Transition, 12(2), 275-
305.
Kim, A. M. (2007). North versus South: the impact of social norms in the market pricing of private
property rights in Vietnam. World Development, 35(12), 2079-2095.
Kim, A. M. (2016). The extreme primacy of location: Beijing's underground rental housing
market. Cities, 52, 148-158.
Kim, A.M. (2018). Smarter than Smart Cities: GIS and spatial analysis for socio-economic
applications that recover humanistic media and visualization. In: Cao, K. & Silva, E. A.
(eds) Comprehensive Geographic Information Systems. Vol. 3. Waltham, MA: Elsevier,
pp. 360–370.
Kim, A. M. (2019). The New Geographies and Spatialities of Informality. In The New Companion
to Urban Design (pp. 152-162). Routledge, pp. 152-162.
Kitchin, R. (2013). Big data and human geography: Opportunities, challenges and risks. Dialogues
in human geography, 3(3), 262-267.
Kitchin, R. (2014). The real-time city? Big data and smart urbanism. GeoJournal, 79(1), 1-14.
Knight, J., Deng, Q., & Li, S. (2017). China’s expansion of higher education: The labour market
consequences of a supply shock. China Economic Review, 43, 127-141.
Kuo, L. (2018). Two boys fatally stabbed near school in Shanghai. The Guardian. 2018-06-28.
Available at: https://www.theguardian.com/world/2018/jun/28/two-boys-fatally-stabbed-
near-school-shanghai. last accessed: 04/21/2020).

140
Lee, C. K. (1998). Gender and the South China miracle: Two worlds of factory women. Univ of
California Press.
Li, C. (2010). Introduction: The rise of the middle class in the Middle Kingdom. In Li, C. (Eds.)
China’s Emerging Middle Class: Beyond Economic Transformation. Washington, D.C.:
Brookings Institution Press, 3-31.
Li, Z., & Wu, F. (2008). Tenure based residential segregation in post reform Chinese cities: a
case study of Shanghai. Transactions of the Institute of British Geographers, 33(3), 404-
419.
Li, B. & Duda, M. (2010). Employers as landlords for rural-to-urban migrants in Chinese
cities. Environment and Urbanization 22(1): 13-31.
Li, S., & Sicular, T. (2014). The distribution of household income in China: Inequality, poverty
and policies. The China Quarterly, 217, 1-41.
Li, S. M., & Liu, Y. (2016). The jobs-housing relationship and commuting in Guangzhou, China:
Hukou and dual structure. Journal of Transport Geography, 54, 286-294.
Li, J., & Liu, Z. (2018). Housing stress and mental health of migrant populations in urban China.
Cities, 81, 172-179.
Li, X., Xiao, Z., & Xiao, S. (2009). Suicide among the elderly in mainland
China. Psychogeriatrics, 9(2), 62-66.
Li, L., Li, S. M., & Chen, Y. (2010). Better city, better life, but for whom?: The hukou and resident
card system and the consequential citizenship stratification in Shanghai. City, Culture and
Society, 1(3), 145-154.
Li, H., Meng, L., Shi, X., & Wu, B. (2012). Does attending elite colleges pay in China?. Journal
of Comparative Economics, 40(1), 78-88.
Li, S., Wang, L., & Chang, K. L. (2018). Do internal migrants suffer from housing extreme
overcrowding in urban China?. Housing Studies, 33(5), 708-733.
Li, H., Wei, Y. D., Wu, Y., & Tian, G. (2019). Analyzing housing prices in Shanghai with open
data: Amenity, accessibility and urban structure. Cities, 91, 165-179.
Lian, S. (2010) II: ?. (Ant tribe II: Whose Era is it?) DynoMedia Inc.
Liao, B. & Wong, D.W. (2015). Changing urban residential patterns of Chinese migrants:
Shanghai, 2000–2010. Urban Geography 36(1): 109-126.
Lin, X. (2019). “Purchasing hope”: the consumption of children’s education in urban China. The
Journal of Chinese Sociology, 6(1): 8.

141
Lin, J. Y., Cai, F., & Li, Z. (2003). The China miracle: Development strategy and economic reform.
Chinese University Press.
Lin, J. P., & Yi, C. C. (2011). Filial norms and intergenerational support to aging parents in China
and Taiwan. International Journal of Social Welfare, 20, S109-S120.
Ling, M. (2015). “Bad students go to vocational schools!”: Education, social reproduction and
migrant youth in urban China. The China Journal, (73), 108-131.
Liu, Z. (2005). Institution and inequality: the hukou system in China. Journal of comparative
economics, 33(1), 133-157.
Liu, Z. (2007). The external returns to education: Evidence from Chinese cities. Journal of Urban
Economics, 61(3), 542-564.
Liu, J. (2018). Subversive Writing: Li Xiaoguai's Newly Coined Chinese Characters and His
Comic Blogging. Chinese Literature, Essays, Articles, Reviews, 40, 199-220.
Liu, T., & Sun, L. (2016). Pension reform in China. Journal of aging & social policy, 28(1), 15-
28.
Liu, Z., Wang, Y., & Tao, R. (2013). Social capital and migrant housing experiences in urban
China: A structural equation modeling analysis. Housing Studies, 28(8), 1155-1174.
Liu, X., Song, Y., Wu, K., Wang, J., Li, D., & Long, Y. (2015a). Understanding urban China with
open data. Cities, 47, 53-61.
Liu, T., Qi, Y., Cao, G., & Liu, H. (2015b). Spatial patterns, driving forces, and urbanization effects
of China’s internal migration: County-level analysis based on the 2000 and 2010
censuses. Journal of Geographical Sciences, 25(2), 236-256.
Logan, J. R., Fang, Y., & Zhang, Z. (2009). Access to housing in urban China. International
Journal of Urban and Regional Research, 33(4), 914-935.
Logan, J. R., Fang, Y., & Zhang, Z. (2010). The winners in China's urban housing reform. Housing
studies, 25(1), 101-117.
Loyalka, P., Song, Y., & Wei, J. (2012). The distribution of financial aid in China: Is aid reaching
poor students?. China economic review, 23(4), 898-917.
Lubove, R. (1963). The Progressives and the Slums: Tenement House Reform in New York City,
1890– 1917. Pittsburgh, PA: University of Pittsburgh Press.
Lu, M., & Xia, Y. (2016). Migration in the People's Republic of China. ADBI Working Paper
Series, No. 593.

Lucas, R. E. (1989). On the mechanics of economic development. NBER Working Paper, (R1176).

142
Ludwig, J., Duncan, G. J., Gennetian, L. A., Katz, L. F., Kessler, R. C., Kling, J. R., &
Sanbonmatsu, L. (2013). Long-term neighborhood effects on low-income families:
Evidence from Moving to Opportunity. American economic review, 103(3), 226-31

Luo, D., & Xing, C. (2016). Population adjustments in response to local demand shifts in China.
Journal of Housing Economics, 33, 101-114.

Luo, Y., Guo, F., & Shi, J. (2018). Expansion and inequality of higher education in China: How
likely would Chinese poor students get to success?. Higher Education Research &
Development, 37(5), 1015-1034.
Ma, Y., & Yang, D. (2015). Development track and route analysis of higher education opportunity
inequality between rural and urban students ()
. Tsinghua Journal of Education () ,), 36(2), 7–14.
Maalsen, S. (2020). ‘Generation Share’: digitalized geographies of shared housing. Social &
Cultural Geography, 21(1), 105-113.
Machin, S., Salvanes, K. G., & Pelkonen, P. (2012). Education and mobility. Journal of the
European Economic Association, 10(2), 417-450.

MacQueen, K. M., McLellan, E., Kay, K., & Milstein, B. (1998). Codebook development for team-
based qualitative analysis. Cam Journal, 10(2), 31-36.
Mak, S. W., Choy, L. H., & Ho, W. K. (2007). Privatization, housing conditions and affordability
in the People's Republic of China. Habitat International, 31(2), 177-192.
Markham, A. N. (2005). The methods, politics, and ethics of representation in online ethnography.
In The Sage Handbook of Qualitative Research, SAGE, Thousand Oaks, CA.
Malamud, O., & Wozniak, A. (2012). The impact of college on migration: evidence from the
Vietnam generation. Journal of Human resources, 47(4), 913-950.

Meng, X., Shen, K., & Xue, S. (2013). Economic reform, education expansion, and earnings
inequality for urban males in China, 1988–2009. Journal of Comparative
Economics, 41(1), 227-244.
Meng, X., & Zhang, J. (2001). The two-tier labor market in urban China: occupational segregation
and wage differentials between urban residents and rural migrants in Shanghai. Journal of
comparative Economics, 29(3), 485-504.

Michel, V. (2015). China online: Netspeak and wordplay used by over 700 million Chinese
internet users. Tuttle Publishing.
Mills, E. S. (1967). An aggregative model of resource allocation in a metropolitan area. The
American Economic Review, 57(2), 197-210.

143

Ministry of Agriculture of the People’s Republic of China (2013). 960
7 . Available at:
http://jiuban.moa.gov.cn/fwllm/qgxxlb/sh/201307/t20130702_3514783.htm. (last
accessed: 05/08/2020).
Ministry of Education of the People’s Republic of China (2017a) . Gāoděng jiàoyù
xuéxiào shù. Number of Higher Education Institutions.
http://en.moe.gov.cn/Resources/Statistics/edu_stat2017/rgion/201808/t20180808_3447
50.html
Ministry of Education of the People’s Republic of China (2017b) . Gāoděng jiàoyù pǔtōng běn, zhuānkē xuéshēng shù. The number of ordinary and junior
college students in higher education. http://en.moe.gov.cn/Re-
sources/Statistics/edu_stat2017/region/201808/t20180808_344744.html.
Miraftab, F., & Kudva, N. (2014). Introduction. In: Miraftab, F., & Kudva, N. (Eds.). Cities of the
global South reader. Pp. 1-6. Routledge.
Mitchell, R. (2018). Web scraping with Python: Collecting more data from the modern web.
O'Reilly Media, Inc.
Mok, K. H. (2009). The growing importance of the privateness in education: Challenges for higher
education governance in China. Compare, 39(1), 35-49.
Mok, K. H., & Qian, J. (2018). Massification of higher education and youth transition: skills
mismatch, informal sector jobs and implications for China. Journal of Education and Work,
31(4), 339-352.
Mok, K. H., & Wu, A. M. (2016). Higher education, changing labour market and social mobility
in the era of massification in China. Journal of Education and Work, 29(1), 77-97.
Molloy, R., Smith, C. L., & Wozniak, A. (2011). Internal migration in the United States. Journal
of Economic perspectives, 25(3), 173-96.

Montag, C., Becker, B., & Gan, C. (2018). The multipurpose application WeChat: a review on
recent research. Frontiers in psychology, 9, 2247.
Montgomery, J. L. (2012). The inheritance of inequality: hukou and related barriers to compulsory
education for China's migrant children. Pac. Rim L. & Pol'y J., 21, 591.
Moore, R. L., Bindler, E., & Pandich, D. (2010). RESEARCH NOTE: Language with attitude:
American slang and Chinese lǐyǔ1. Journal of sociolinguistics, 14(4), 524-538.
Moretti, E. (2004). Human capital externalities in cities. In Handbook of regional and urban
economics (Vol. 4, pp. 2243-2291). Elsevier.

144
Moretti, E. (2011). Local Labor Markets, Volume 4. Elsevier B.V.

Moretti, E. (2012). The new geography of jobs. Houghton Mifflin Harcourt.
Mostafa, A., Wong, F. K., & Hui, C. M. (2006). Relationship between housing affordability and
economic development in mainland China—case of Shanghai. Journal of Urban Planning
and Development, 132(1), 62-70.
Mossberger, K., Tolbert, C. J., Bowen, D., & Jimenez, B. (2012). Unraveling different barriers to
Internet use: Urban residents and neighborhood effects. Urban Affairs Review, 48(6), 771-
810.
Mühlhahn, K. (2019). Making China Modern: From the Great Qing to Xi Jinping. Harvard
University Press.
Murphy, R. (2002). How migrant labor is changing rural China. Cambridge University Press.

Muth, R. (1969). Cities and Housing. The University of Chicago Press. Chicago, Il.
Mykyta, L. (2012). Economic downturns and the failure to launch: The living arrangements of
young adults in the US 1995-2011. US Census Bureau Social, Economic and Housing
Statistics Division (SEHSD) Working Paper, 24.
Mykyta, L., & Macartney, S. (2011). The effects of recession on household composition: ‘doubling
up’ and economic well-being. US Census Bureau. Social, Economic and Household
Statistics Division Working Paper, 4.
National Bureau of Statistics (2019). 2018 . Available at:
http://www.stats.gov.cn/tjsj/zxfb/201904/t20190429_1662268.html. (last accessed:
05/08/2020).
Naughton, B. J. (2006). The Chinese economy: Transitions and growth. MIT press.
Nee, V. (1996). The emergence of a market society: Changing mechanisms of stratification in
China. American journal of sociology, 101(4), 908-949.
Noble, S. U. (2018). Algorithms of oppression: How search engines reinforce racism. NYU Press.
Office of the Deputy Prime Minister of the UK (2004). The Impact of Overcrowding on Health
and Education: A Review of the Evidence and Literature. Available at:
http://dera.ioe.ac.uk/5073/1/138631.pdf. (last accessed 03/02/2017).
Padgett, D. K. (2012). Qualitative and mixed methods in public health. Sage publications.
Palloni, A., Massey, D. S., Ceballos, M., Espinosa, K., & Spittel, M. (2001). Social capital and
international migration: A test using information on family networks. American Journal of
Sociology, 106(5), 1262-1298.

145
People’s Daily Online (2008). Over 10 billion yuan to be invested in "211 Project". 2008-03-26.
Available at: http://english.people.com.cn/90001/6381319.html (last accessed 04/08/2020).
People’s Daily (2019). 996 . 2019-04-16.
Available at: http://sc.people.com.cn/n2/2019/0416/c345459-32846731.html. (last
accessed: 04/13/2020).
Postiglione, G. A. (2015). Schooling and Inequality in China. In: Postiglione, G. (eds.). Education
and Social Change in China: Inequality in a Market Economy. Pp. 3-24, London and New
York: Routlege.
Perlman, J. E. (1979). The myth of marginality: Urban poverty and politics in Rio de Janeiro. Univ
of California Press.
Pun, N. (2005). Made in China: Women Factory Workers in a Global Workplace. Durham, London,
Hong Kong: Duke University Press and Hong Kong University Press.
Pun, N. & Smith, C. (2007). Putting transnational labour process in its place: the dormitory labour
regime in post-socialist China. Work, employment and society, 21(1), 27-45.
Qian, L. (2018). The ‘inferior’ talk back: Suzhi (human quality), social mobility, and E-commerce
economy in China. Journal of Contemporary China, 27(114), 887-901.
Qin, L., Wang, W., & Lu, Y. (2018). The working and living conditions of college-educated rural
migrants in China. Asian Population Studies, 14(2), 172-193.
Quinn, N., Fessler, D., Johnson, A. W., Lebra, T. S., Lucy, J., & Whitehead, H. (1992). Human
motives and cultural models. Cambridge University Press.
Quinn, K., Kaufman, J. S., Siddiqi, A., & Yeatts, K. B. (2010). Stress and the city: housing
stressors are associated with respiratory health among low socioeconomic status Chicago
children. Journal of Urban Health, 87(4), 688-702.
Rae, A. (2015). Online housing search and the geography of submarkets. Housing Studies, 30(3),
453-472.
Ramzy, A. (2018). Shanghai Knife Attack Kills Two Schoolchildren. The New York Times. 2018-
06-28. Available at: https://www.nytimes.com/2018/06/28/world/asia/shanghai-school-
knife-attack-.html. (last accessed: 04/21/2020).
Rauch, J.E. (1993). Productivity Gains from Geo- graphic Concentration of Human Capital:
Evidence from the Cities. Journal of Urban Economics 34 (3): 380–400.
Ravallion, M., Chen, S., & Sangraula, P. (2007). New evidence on the urbanization of global
poverty. Population and Development Review, 33(4), 667-701.

146
Redfearn, C. L. (2009). How informative are average effects? Hedonic regression and amenity
capitalization in complex urban housing markets. Regional Science and Urban
Economics, 39(3), 297-306.
Ren, D. (2020). Shanghai homeowners left to rue falling prices, as government stays committed
to keeping property sector in check. South China Morning Post. 2020-01-26. Available at:
https://www.scmp.com/business/china-business/article/3047704/shanghai-homeowners-
left-rue-falling-prices-government. (last accessed: 04/21/2020).
Rhodes, D. (2002). Students and housing: a testing time?. In Young people, housing and social
policy (pp. 81-94). Routledge.
Rofel, L. (1999). Other modernities: Gendered yearnings in China after socialism. University of
California Press. Berkeley.
Roxburgh, H. (2018). China's radical plan to limit the populations of Beijing and Shanghai. The
Guardian. 03/19/2018. Available at:
https://www.theguardian.com/cities/2018/mar/19/plan-big-city-disease-populations-fall-
beijing-shanghai. (last accessed 04/10/2020).
Roy, A. (2004). Transnational Trespassings: The Geopolitics of Urban Informality. In: Alzayad,
N. & Roy, A. (Eds.) (2004). Urban Informality: Transnational Perspectives from the
Middle East, Latin America, and South Asia. Lanham, MD: Lexington Books, 2004.
Roy, A. (2005). Urban informality: toward an epistemology of planning. Journal of the american
planning association, 71(2), 147-158.
Roy, A. (2009). Why India cannot plan its cities: Informality, insurgence and the idiom of
urbanization. Planning theory, 8(1), 76-87.
Rozelle, S., Taylor, J. E., & DeBrauw, A. (1999). Migration, remittances, and agricultural
productivity in China. American Economic Review, 89(2), 287-291.

Sandel, M., & Wright, R. J. (2006). When home is where the stress is: expanding the dimensions
of housing that influence asthma morbidity. Archives of disease in childhood, 91(11), 942-
948.
Sandelowski, M. (1995). Qualitative analysis: What it is and how to begin. Research in nursing &
health, 18(4), 371-375.
Sargeson, S. (2002). Subduing “the rural house-building craze”: attitudes towards housing
construction and land use controls in four Zhejiang villages. The China Quarterly, 172,
927-955.

Schweitzer, L. (2014). Planning and social media: a case study of public transit and stigma on
Twitter. Journal of the American Planning Association, 80(3), 218-238.

147
Seltzer, J. A., Lau, C. Q., & Bianchi, S. M. (2012). Doubling up when times are tough: A study of
obligations to share a home in response to economic hardship. Social science research,
41(5), 1307-1319.
Scrapy Community (2016). Scrapy Documentation. Available at: https://scrapy.org. (last accessed:
05/03/2020).
Shanghai Municipal Government. (2011). The Procedures of Shanghai Municipality on the
Administration of Residential Tenancy. Decree of Shanghai Municipal People’s
Government, No. 68. 07/07/2011. Available at:
http://www.shanghai.gov.cn/shanghai/node27118/node27386/node27408/node27621/nod
e28697/u26ai30039.html. (last accessed 03/02/2017).
Shanghai Municipal Government (2017). Living Conditions of Migrant Workers in Shanghai ( ). Available at:
http://www.shanghai.gov.cn/nw2/nw2314/nw24651/nw42131/nw42178/u21aw1232783.h
tml. (last accessed: 01/31/2020).
Shelton, T., Poorthuis, A., & Zook, M. (2015). Social media and the city: Rethinking urban socio-
spatial inequality using user-generated geographic information. Landscape and urban
planning, 142, 198-211.
Shen, Y. (2015). Why Does the Government Fail to Improve the Living Conditions of Migrant
Workers in Shanghai? Reflections on the Policies and the Implementations of Public Rental
Housing under Neoliberalism. Asia & the Pacific Policy Studies, 2(1), 58-74.
Shi, W., Chen, J., & Wang, H. (2016). Affordable housing policy in China: New developments
and new challenges. Habitat International, 54, 224-233.
Solinger, D. J. (1999). Contesting citizenship in urban China: Peasant migrants, the state, and the
logic of the market. University of California Press.
Song, Y., Zenou, Y., & Ding, C. (2008). Let's not throw the baby out with the bath water: the role
of urban villages in housing rural migrants in China. Urban Studies, 45(2), 313-330.
Strauss, A., & Corbin, J. (1994). Grounded theory methodology. In: Handbook of qualitative
research, 17, 273-85.
Strauss, J., Hong, H., Lei, X., Li, L., Park, A., Yang, L., & Zhao, Y. (2012). Healthcare and
insurance among the elderly in China: evidence from the CHARLS pilot. In Aging in Asia:
Findings from New and Emerging Data Initiatives. National Academies Press (US).
Suda, K. (2016). A Room of One's Own: Highly Educated Migrants' Strategies for Creating a
Home in Guangzhou. Population, Space and Place, 22(2), 146-157.
Sun, Y. (2017). China’s Higher Education Enrollment Rate Reaches 42.7 per cent in 2016.
People’s Daily Online. 2017-11-07. http://en.people.cn/n3/2017/0711/c90000-
9240278.html (last accessed 04/08/2020).

148
Sun. Z. (2018). High-Speed Urbanization and the “Over-sized Cities” in Contemporary China.
05/04/2018. Public Talk at The New School, Starr Foundation Hall, 63 Fifth Ave, NY.
Swider, S. (2016). Building China: Informal work and the new precariat. Cornell University Press.
Tanasescu, A., Wing-Tak, E. C., & Smart, A. (2010). Tops and bottoms: State tolerance of illegal
housing in Hong Kong and Calgary. Habitat International, 34(4), 478-484.
Tang, F. (2019). China’s state pension fund to run dry by 2035 as workforce shrinks due to effects
of one-child policy, says study. 2019-04-12. South China Morning Post. Available at:
https://www.scmp.com/economy/china-economy/article/3005759/chinas-state-pension-
fund-run-dry-2035-workforce-shrinks-due. (last accessed: 04/04/2020).
Tang, M., & Coulson, N. E. (2017). The impact of China's housing provident fund on
homeownership, housing consumption and housing investment. Regional Science and
Urban Economics, 63, 25-37.
Tesch, R. (2013). Qualitative research: Analysis types and software. Routledge.
Timberlake, M., Wei, Y. D., Ma, X., & Hao, J. (2014). Global cities with Chinese
characteristics. Cities, 41, 162-170.
Ting, K. F., & Chiu, S. W. (2002). Leaving the parental home: Chinese culture in an urban
context. Journal of Marriage and Family, 64(3), 614-626.
Tomba, L. (2004). Creating an urban middle class: Social engineering in Beijing. The China
Journal, (51), 1-26.
Tong, H. (2013). Research on WeChat according to communication study and its
influence. Chongqing Social Sciences, 9(11), 61-66.
Unger, J. (2002). Transformation of Rural China. ME Sharpe.
UN-Habitat (2003). The Challenge of Slums. Global Report on Human Settlements 2003. London:
Earthscan.
UN-Habitat (2009). Better Information, Better Cities: Monitoring the Habitat Agenda and the
Millennium Development Goals- Slums Target. Available at :
/www.ine.pt/ngt_server/attachfileu.jsp?look_parentBoui=143216309&att_display=n&att
_download=y. (last accessed 02/02/2017).
UN-Habitat (2010). Housing as a strategy for poverty reduction in Ghana United Nations human
settlements programme. Nairobi, Kenya: UN-HABITAT.
Urban Surveyors, (2018). | iPhone Available at: http://www.surea.com/Home/News/?ID=5238&typeID=3.
(last accessed: 04/20/2020).

149
Wallis, C. (2013). Technomobility in China: Young migrant women and mobile phones. NYU
Press.
Wan, Y. (2006). Expansion of Chinese higher education since 1998: Its causes and outcomes. Asia
Pacific Education Review, 7(1), 19-32.
Wang, Y. P. (2004). Urban poverty, housing and social change in China. Routledge.
Wang, Y.P., & Murie, A. (1996). The process of commercialisation of urban housing in China.
Urban Studies, 33(6), 971-989.
Wang, L. & Feng, D.P. (2016). . Xinhua Net.
2016-11-29. Available at: http://www.xinhuanet.com//politics/2016-
11/29/c_1120016588.htm. (last accessed: 04/20/2020).
Wang, H., Li, W., & Deng, Y. (2017). Precarity among highly educated migrants: college
graduates in Beijing, China. Urban Geography, 38(10), 1497-1516.
Ward, P. M. (2019). Self‐Help Housing. The Wiley Blackwell Encyclopedia of Urban and
Regional Studies, 1-6.
Waters, J.L. (2006). Geographies of cultural capital: education, international migration and family
strategies between Hong Kong and Canada. Transactions of the Institute of British
Geographers, 31(2), 179-192.
Wei, Y.H. (2000). Regional Development in China: States, Globalization, and Inequality. (London,
Routledge).
Whyte, M.K. (2010). Myth of the Social Volcano: Perceptions of Inequality and Distributive
Injustice in Contemporary China. Redwood City, CA: Stanford University Press.
Wiemers, E. E. (2014). The effect of unemployment on household composition and doubling up.
Demography, 51(6), 2155-2178.
Willis, P. E. (1981). Learning to labor: How working class kids get working class jobs. Columbia
University Press.
Wong, E. (2013). A Chinese Virtue Is Now the Law. 2013-07-02. The New York Times. Available
at: https://www.nytimes.com/2013/07/03/world/asia/filial-piety-once-a-virtue-in-china-is-
now-the-law.html. (last accessed: 04/13/2020).
World Bank and Development Research Center of the State Council (2014). Urban China: Toward
Efficient, Inclusive, and Sustainable Urbanization. Washington, DC: World Bank.
World Health Organization, WHO. (2000). Overcrowding. Available at:
http://www.who.int/ceh/indicators/overcrowding.pdf. (last accessed 02/02/2017).

150
Woronov, T. E. (2012). Class Consciousness, Service Work: youth and class in Nanjing vocational
secondary schools. Journal of Contemporary China, 21(77), 779-791.
Wu, W. (2002). Migrant housing in urban China: choices and constraints. Urban Affairs Review,
38(1), 90-119.
Wu, W. (2004). Sources of migrant housing disadvantage in urban China. Environment and
planning A, 36(7), 1285-1304.
Wu, W. (2008). Migrant settlement and spatial distribution in metropolitan Shanghai. The
Professional Geographer, 60(1), 101-120.
Wu, F. (2016a). Housing in Chinese urban villages: the dwellers, conditions and tenancy
informality. Housing Studies, 31(7), 852-870.
Wu, J. (2016b). . The first
Shanghai sublessor was put on a blacklist for not stopping to engage in group rental
activities. The Paper. 2016-03-03. Available at:
https://www.thepaper.cn/newsDetail_forward_1438600 (last accessed: 05/12/2020).
Wu, J.Y. (2019) Beijing releases a new version of the rental contract model text: Explicitly
prohibiting illegal group rentals ( :
). Beijing News. 2019-07-08. Available at:
http://www.bjnews.com.cn/news/2019/07/08/600569.html (last accessed: 05/05/2020).
Wu, F., Zhang, F., & Webster, C. (2013). Informality and the development and demolition of urban
villages in the Chinese peri-urban area. Urban Studies, 50(10), 1919-1934.
Wu, Q., Zhang, X., & Waley, P. (2016). Jiaoyufication: When gentrification goes to school in the
Chinese inner city. Urban studies, 53(16), 3510-3526
Xiang, B. (2019). Xi urges patriotism among youth, striving for brighter China. Xinhua. 2019-04-
30. Available at: http://www.xinhuanet.com/english/2019-04/30/c_138024983.htm. (last
accessed: 04/20/2020).
Xiao, Y., & Bian, Y. (2018). The influence of hukou and college education in China’s labour
market. Urban Studies, 55(7), 1504-1524.

Xing, C., & Zhang, J. (2017). The preference for larger cities in China: Evidence from rural-urban
migrants. China Economic Review, 43, 72-90.
Xinhua (2017a). 211 985 . 2017-06-21. Available at:
http://www.xinhuanet.com/forum/2017-06/21/c_129637701.htm (last accessed
04/08/2020).
Xinhua (2017b). Women dominate higher education in China. Xinhua. 2017-10-28. Available at:
http://en.people.cn/n3/2017/1028/c90000-9285962.html. (last accessed: 05/08/2020).

151
Xinhua (2018a). 2017 28652 . 2018-04-27.
http://www.xinhuanet.com/finance/2018-04/27/c_129860718.htm (last accessed:
04/08/2020).
Xinhua (2018b). . Available at:
http://www.xinhuanet.com/politics/2018lh/2018-03/07/c_1122502719.htm. (last accessed:
04/20/2020).
Xinhua (2018c). . 2018-11-
28. Available at: http://www.xinhuanet.com/2018-11/28/c_1123781233.htm. (last
accessed: 04/20/2020).
Xinhua (2018d). . 2018-09-12.
Available at: http://www.xinhuanet.com/fortune/2018-09/12/c_129952080.htm. (last
accessed: 04/20/2020).
Xinhua (2018e). China sees steady graduate employment rate in 2017: survey. Xinhua. 2018-06-
11. Available at: http://www.xinhuanet.com/english/2018-06/11/c_137246978.htm. (last
accessed: 05/08/2020).
Xinhua (2019a). Chinese university graduates rise exponentially, have diverse career options.
2019-06-24. Available at: http://www.xinhuanet.com/english/2019-
06/24/c_138169311.htm (last accessed: 04/08/2020).
Xinhua (2019b). China's senior population may surge from 2025. Xinhuanet. 2018-02-09.
Available at: http://www.xinhuanet.com/english/2019-02/09/c_137808376.htm. (last
accessed: 04/04/2020).
Yang, P. (2010). Who gets more financial aid in China? A multilevel analysis. International
Journal of Educational Development, 30(6), 560-569.
Yang, Z., & Chen, J. (2014). Housing Reform and the Housing Market in Urban China. In Z. Yang,
& J. Chen (Eds.). Housing affordability and housing policy in urban China. Berlin:
Springer
Yang, J., Huang, X., & Liu, X. (2014). An analysis of education inequality in China. International
Journal of Educational Development, 37, 2-10.
Yang, S., Wang, M. Y., & Wang, C. (2015). Socio-spatial restructuring in Shanghai: Sorting out
where you live by affordability and social status. Cities, 47, 23-34.
Yeung, W. J. J. (2013). Higher education expansion and social stratification in China. Chinese
Sociological Review, 45(4), 54-80.
Yi, C., & Huang, Y. (2014). Housing consumption and housing inequality in Chinese cities during
the first decade of the twenty-first century. Housing Studies, 29(2), 291-311.

152
Ying, G. (2003). Consumption patterns of Chinese children. Journal of Family and Economic
Issues, 24(4), 373-379.
Yu, Z. (2006). Heterogeneity and dynamics in China's emerging urban housing market: two sides
of a success story from the late 1990s. Habitat International, 30(2), 277-304.
Yu, X. (2019). Average monthly salary in major cities nears 8,500 yuan. China Daily. 2019- 07-
04. Available at:
https://www.chinadaily.com.cn/a/201907/04/WS5d1d63cfa3105895c2e7ba4d.html. (last
accessed: 04/03/2020).
Zhan, H. J. (2004). Willingness and expectations: Intergenerational differences in attitudes toward
filial responsibility in China. Marriage & Family Review, 36(1-2), 175-200.
Zhang, P. (2018). Poisoned paint fears addressed after renter dies in eastern China. South China
Morning Post. 2018-09-03.Available at:
https://www.scmp.com/news/china/society/article/2162564/poisoned-paint-fears-
addressed-after-renter-dies-eastern-china. (last accessed: 04/20/2020).
Zhang, X., & Kanbur, R. (2009). Spatial inequality in education and health care in China. In
Regional Inequality in China (pp. 92-110). Routledge.
Zhang, Y., & Chen, J. (2014). The changing prevalence of housing overcrowding in post-reform
China: The case of Shanghai, 2000–2010. Habitat International, 42, 214-223.
Zhang, N. J., Guo, M., & Zheng, X. (2012). China: awakening giant developing solutions to
population aging. The Gerontologist, 52(5), 589-596.
Zhang, D., Li, X., & Xue, J. (2015). Education inequality between rural and urban areas of the
People's Republic of China, migrants’ children education, and some implications. Asian
Development Review, 32(1), 196-224.
Zhang, X., Wen, D., Liang, J., & Lei, J. (2017). How the public uses social media wechat to obtain
health information in china: a survey study. BMC medical informatics and decision
making, 17(2), 66.
Zhang, J., Wang, R., & Lu, C. (2019). A quantitative analysis of Hukou reform in Chinese cities:
2000–2016. Growth and Change, 50(1), 201-221.
Zhao, L., Liu, S., & Zhang, W. (2018). New trends in internal migration in China: Profiles of the
New‐generation migrants. China & World Economy, 26(1), 18-41.
Zhaopin (2014). 2014 : . 2014 Graduate
Employability Report: College Graduates’ Expectations and Reality.
Zhaopin (2015). 2015 . 2015 Graduate Employability Report.
Zhaopin (2017). 2017 . 2017 Graduate Employability Report.

153
Zhong, H. (2011). Returns to higher education in China: What is the role of college quality?. China
Economic Review, 22(2), 260-275.
Zhong, W., Wang, D., Xie, D., & Yan, L. (2017). Dynamic characteristics of Shanghai’s
population distribution using cell phone signaling data. (Geographical
Research), 36, 972-984.
Zhou, W. (2019). Bidders in Shanghai's car license auctions declining. China Daily. 2019-04-03.
Available at: http://www.chinadaily.com.cn/a/201904/03/WS5ca40d7da3104842260b41
df.html. (last accessed: 04/03/2020).
Zhou, X., Moen, P., & Tuma, N. B. (1998). Educational stratification in urban China: 1949–94.
Sociology of Education, 71, 199–222.
Zhou, L., & Zhang, S. J. (2017). How face as a system of value-constructs operates through the
interplay of mianzi and lian in Chinese: a corpus-based study. Language Sciences, 64, 152-
166.
Zhu, X. (2012). Understanding China's growth: Past, present, and future. Journal of Economic
Perspectives, 26(4), 103-24.
Zhu, H. (2016). Adult children’s characteristics and intergenerational financial transfers in urban
China. Chinese Journal of Sociology, 2(1), 75-94.
Zhu, X., Whalley, J., & Zhao, X. (2014). Intergenerational transfer, human capital and long-term
growth in China under the one child policy. Economic Modelling, 40, 275-283.
Zuo, M. (2018). Two children stabbed to death outside Shanghai primary school. South China
Morning Post. 2018-06-28. Available at:
https://www.scmp.com/news/china/society/article/2152879/two-children-stabbed-death-
outside-shanghai-primary-school. (last accessed: 04/21/2020).
Zweig, D. (2006). Competing for talent: China's strategies to reverse the brain drain. International
labour review, 145(1 2), 65-90.

154
APPENDIX

Appendix A: The Veracity of Online Advertisement Data for Housing and
Informality Research

The study of Shanghai’s group rental housing market also provides an opportunity to explore the
opportunities and limitations of scraped data. Through groundtruthing fieldwork I discovered that
the data I had scraped was systematically misrepresenting actual market conditions. As a result, I
also invested in collecting “real” market data. Through comparative analyses between the field
data and the scraped data it is possible to identify the differences between the two and to find more
general lessons about how to more responsibly utilize scraped data for research.
Recall that because our research team located and chose sites for field data collection based
on initial online advertisements, I am able to match actual and advertised rentals for a one-to-one
comparison (Table 12).

Table 12: Comparative Descriptive Statistics: Mean Values of Online Data versus Field Data

Field Survey Scraped Data Gap
Dist. to CBD (km) 4.41 4.34 1.59% farther
Dist. to Closest
Subway Stop (km)
0.43 0.38 11.63% farther
Number of Subway Stops
(within 800m)

1.67 1.67 0%
Number of Subway Lines
(within 800m)
2.27 2.35 3.52% less lines
Rent (RMB) 850.16 688.03 19.07% costlier
People Per Room 6.6 6.54 0.91% more people
Number of Bedrooms 3.24 2.95 8.95% more bedrooms
Number of Bathrooms 1.61 1.98 22.98% less bathrooms

In 94% of the cases, the online data was inaccurate. These inaccuracies included the
information about the key variables, location (86%) and rents (81%). Perhaps not surprisingly, real
rents were higher in 81% of the cases and on average 20% more expensive. Since the bias appears
one-directional, the differential should not endanger the integrity of multivariate regression
analysis. Similarly, most real apartments (79%) were slightly farther from the city center but still
within walking distance from the fake address posted online. The median distance between real
and fake addresses was 332 meters. In other words, the ads present a more attractive situation in
terms of location and rent than what is actually offered but still within a range that might be
acceptable to prospective renters.
The inaccurate information may have two root causes. On the one hand, listing inaccurate,
but proximate addresses could be a means to disguise and shield the informal rental activity. As
mentioned earlier, during fieldwork brokers invariably asked to meet the research teams in public

155
places and then escorted us to the actual places. On the other hand, advertised beds were often
rented within hours and brokers/landlords frequently managed more than one group rental property.
Given the volume and turnover, they may simply re-use outdated ads out of convenience. Either
way, I find that the internet ads are used to locate the potential renters more than to give very
accurate information about the housing situation.
I also estimate hedonic models using the larger online advertisement data (Table 13). The
overall goodness of fit is reasonably high, even though important price structuring variables I
found with the field survey data (lower bunk dummy variable, cooking dummy variable) are not
available in the ads. Compare coefficient estimates across specifications, I find similarities and
differences in the revealed price structure.

Table 13: Hedonic Regression Results,
ß
Web Scraped Online Advertisement Data
Model 1 Model 2 Model 3 Model 4 Model 5
Location
Distance to CBD (km) -0.164***
(-24.64)
-0.115***
(-22.20)
-0.106***
(-21.17)
-0.107***
(-21.17)
-0.102***
(-20.25)
Distance to closest
subway stop (km)
0.018
(0.43)
0.116***
(3.64)
0.015
(0.49)
0.016
(0.50)
0.023
(0.76)
Number of subway lines
(within 800m radius)
-0.008
(-1.04)
-0.006
(-0.97)
-0.008
(-1.39)
-0.008
(-1.39)
-0.010*
(-1.76)
Apartment
People per room

-0.148***
(-47.47)
-0.145***
(-48.34)
-0.146***
(-48.10)
-0.140***
(-45.45)
Number of bathrooms

-0.251***
(-16.59)
-0.246***
(-15.04)
-0.312***
(-18.10)
Number of bedrooms

-0.009
(-0.88)

Apartment Size (sqm)

0.001***
(6.30)
N 3146 3146 3135 3135 3118
Adj. R-squared 0.173 0.518 0.559 0.559 0.571
F-statistic 220.0 846.8 794.8 662.4 693.4
ß
In all model specifications the logarithm of the monthly rent price per bed is the dependent variable.

The coefficient estimates on access to the central business district are more important in
the regression with web scraped data (cf. Table 5). The estimates are roughly 5 times larger: being
1 km farther away from the center decreases the rent price by 10-16% (compare to 2-4% for the
real market data). Coefficients on the people per room crowding variable are also at least twice as
large. Adding one more person to a room now decreases the rent by roughly 15%.
Curiously, the coefficient estimates on the number of bathrooms are still significant and
economically sizable. But now adding a bathroom is associated with a decrease in rent by 24-31 %
The discrepancies could be the result of advertisers attempting to make places look more attractive.
Also, access to transit variables are not stable in these models of the internet data. Omitted variable
bias is likely since the cooking dummy and lower bunk dummies are not available for these
estimations and were highly significant and increased the R-squares considerably in field data
model estimates.

156
What did groundtruthing the online data reveal about the veracity of this new data streams
for the study of urban phenomenon?
Contrasting the picture painted online with the experience in the field, I learned that the
online market is not data simply reflecting the real housing market itself. Rather, the online
advertisements are communication data about the market, revealing who the ad listers are seeking
to find in a seller’s market, more than data about the property itself. Groundtruthing also revealed
how these online listings are used strategically by the seller. They obfuscate details in order to
remain hidden in this informal market that the state seeks to shut down. They also advertise slightly
more attractive features in order to entice the initial contact with potential renters. They then show
the actual rentals which are systematically in farther locations, more crowded conditions, and
higher prices. But, these are not outrageously different as to deter a rental agreement and indeed
these bed rentals sell out very quickly. So, while the hedonic price models for both fake and real
datasets conform to urban economic theory, the market prices and locations they represent differ.
These finding have more general implications for the use of scraped data. Scraping online
data provides an innovative way to quickly amass data and has been adopted for a variety of
research agendas (Boeing & Waddell, 2017; Evans & Aceves, 2016; Edelman & Luca, 2014; Folch,
Spielman & Manduca, 2018; Schweitzer, 2014). Often, however, researchers use social media and
internet communications as proxies for the object of study itself, in this case, the real estate market.
This study suggests that instead we need to utilize these data as related but distinct
communication objects. In any society, social stratification, strategy, and cultural norms shape this
online communication. This finding should also have implications for non-communicative big data
generated by sensors. Even with positivist data such as trip patterns, it is impossible to understand
the factors shaping the trips without an understanding of social stratification, in particular
institutional contexts (Schweitzer, 2014). Instead of assuming big data represents a unitary and
homogenous public, scholars need to further develop how social stratification shapes the patterns
in the new data streams. Greater inter-disciplinary integration between the social sciences and
computer science is needed to more accurately understand our new data. Big data is not always
better data but rather different data.

157
Appendix B: Online Social Media (WeChat) Survey Instrument

158

159

160

161

162

Figure 7: Selected Promotional Material to Distribute the Survey Offline

163
Appendix C: Supplemental Material Chapter 4
Difference in Mean Tests: Shanghai vs. Other Tier 1 Cities (First Job Location)

Table 14: Difference in Mean Test, Share of Tier 1 City Hukou Holders, Shanghai vs. Other
Their 1 Cities

*This test was performed on the subgroup of Tier 1 cities.

164
Table 15: Difference in Mean Test, Share of Elite Graduates, Shanghai vs. Other Their 1
Cities

*This test was performed on the subgroup of Tier 1 cities.

Table 16: Difference in Mean Test, Share of Those Earning Less Than 5,000 RMB/ Month,
Shanghai vs. Other Their 1 Cities

*This test was performed on the subgroup of Tier 1 cities

165
Table 17: Difference in Mean Test, Share of Those Spending Less Than 1,000 RMB/ Month
on Rent, Shanghai vs. Other Their 1 Cities

*This test was performed on the subgroup of Tier 1 cities

Table 18: Difference in Mean Test, Share of Group Renters, Shanghai vs. other Their 1 Cities

*This test was performed on the subgroup of Tier 1 cities.

166
Table 19: Difference in Mean Test, Share of Room Shares, Shanghai vs. other Their 1 Cities

*This test was performed on the subgroup of Tier 1 cities.

Table 20: Difference in Mean Test, Share of Those with Family-Provided Housing, Shanghai
vs. Other Their 1 Cities

*This test was performed on the subgroup of Tier 1 cities.

167
Difference in Mean Tests: Tier 1 Cities vs. Non-Tier 1 Cities (First Job Location)

Table 21: Difference in Mean Test, Share of Those with a Bachelor’s Degree or above, Tier
1 vs. Non-Tier 1 Cities

Table 22: Difference in Mean Test, Share of Elite Graduates, Tier 1 vs. Non-Tier 1 Cities

168
Table 23: Difference in Mean Test, Share of Those Earning Less Than 5,000 RMB/ Month,
Tier 1 vs. Non-Tier 1 Cities

Table 24: Difference in Mean Test, Share of Those Earning Less Than 3,000 RMB/ Month,
Tier 1 vs. Non-Tier 1 Cities

169
Table 25: Difference in Mean Test, Share of Spending Less Than 1,000 RMB/ Month on Rent,
Tier 1 vs. Non-Tier 1 Cities

Table 26: Difference in Mean Test, Share of Group Renters, Tier 1 vs. Non-Tier 1 Cities

170
Table 27: Difference in Mean Test, Share of Room Sharers, Tier 1 vs. Non-Tier 1 Cities

Table 28: Difference in Mean Test, Share of Those with Family-Provided Housing, Tier 1 vs.
Non-Tier 1 Cities

171
Table 29: Difference in Mean Test, Share of Those with First Job Trial Periods, Tier 1 vs.
Non-Tier 1 Cities

Abstract (if available)

Abstract In China’s cities, some educated migrants rent individual beds in overcrowded shared apartments in the city center. This dissertation studies the hidden, informal housing phenomenon known as “group rentals” in Shanghai, China and asks what its emergence tells us about changing dynamics of social mobility and contemporary inequality. Based on remotely collected online advertisement data, three summers of fieldwork, a cell phone social media survey, and several years living, studying, and working in China, I employ a mix of methods to generate an in-depth investigation into group renting, not just as a housing, but as a social phenomenon. I find that group renters are mostly recent migrants with degrees from second-tier colleges who trade-off space for access to jobs and opportunities. Beyond providing affordable access, group rentals function as a time-space strategy for Shanghai newcomers who need to keep expenses low

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Household mobility and neighborhood impacts

PDF

Three essays on housing demographics: depressed housing access amid crisis of housing shortage

PDF

The geography of Black commerce and culture: Los Angeles, California, and beyond

PDF

Productive frictions and urbanism in transition: planning lessons from traffic flows and urban street life in Ho Chi Minh City, Vietnam

PDF

Property and labor formalization in the age of the sharing economy: Airbnb, housing affordability, and entrepreneurship in Havana

PDF

Affordable south Los Angeles: survival, support, and different futures

PDF

Environmental justice in real estate, public services, and policy

PDF

The interplay between green space and transit: a case study of revitalization within Atlanta’s Adair Park and its potential to improve future policy outcomes

PDF

The impact of mobility and government rental subsidies on the welfare of households and affordability of markets

PDF

Civic expression in Little Tokyo: how art and culture empowers communities and transforms public participation

PDF

Insights into residential mobility and pricing of rental housing: the role of gentrification, home-ownership barriers, and market concentrations in low-income household welfare

PDF

Evaluating art for social change: the Anti-Eviction Mapping Project and the Los Angeles Poverty Department in relationship to Tania Bruguera’s concept of Arte Útil; Evaluating art for social cha...

PDF

The development implications of China’s Belt and Road Initiative for Russia, Kazakhstan and Belarus

PDF

Goal setting and performance in federal agencies

PDF

Innovation and good intentions: evaluations of three cross-sectoral programs for at-risk populations in Southern California

PDF

Reproducing inequity in organizations: gendered and racialized emotional labor in pubic organizations

PDF

Diverging pathways to citizenship and immigrant integration in the U.S.

PDF

Healthy mobility: untangling the relationships between the built environment, travel behavior, and environmental health

PDF

Resilient and equitable urbanism by design: insights from the collaborative process to reimagine the SF Bay Area

PDF

More knowledge, better plans: a study of heterogeneity in dementia prevalence and mid-life cognitive changes

Asset Metadata

Creator Harten, Julia Gabriele (author)

Core Title Crowded with potential: housing and social mobility strategies among China's educated migrants

School School of Policy, Planning and Development

Degree Doctor of Philosophy

Degree Program Public Policy and Management

Publication Date 07/27/2020

Defense Date 05/27/2020

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag crowding,educated migrants,international planning,migrant housing,OAI-PMH Harvest,rental housing,social mobility,urban China

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Kim, Annette M. (committee chair), Schweitzer, Lisa A. (committee member), Webster, Christopher J. (committee member)

Creator Email jharten@usc.edu,julia.harten@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-341796

Unique identifier UC11664180

Identifier etd-HartenJuli-8783.pdf (filename),usctheses-c89-341796 (legacy record id)

Legacy Identifier etd-HartenJuli-8783.pdf

Dmrecord 341796

Document Type Dissertation

Rights Harten, Julia Gabriele

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