Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Refugee protection and empowerment joint venture: understanding how and why public and private actors cooperate to tackle the global refugee crisis
(USC Thesis Other)
Refugee protection and empowerment joint venture: understanding how and why public and private actors cooperate to tackle the global refugee crisis
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
REFUGEEPROTECTIONANDEMPOWERMENTJOINT
VENTURE:UNDERSTANDINGHOWANDWHYPUBLIC
ANDPRIVATEACTORSCOOPERATETOTACKLETHE
GLOBALREFUGEECRISIS
by
Stefanie Anna Neumeier
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
(POLITICAL SCIENCE AND INTERNATIONAL RELATIONS)
May 2023
Copyright 2023 Stefanie Anna Neumeier
I dedicate this thesis to my mother Brigitte,
who has supported me unconditionally throughout
my entire undergraduate and graduate career.
ii
Acknowledgements
First and foremost, I would like to thank my dissertation committee, Wayne Sandholtz,
PatrickJames,andHannahGarry,fortheircontinuoussupportandencouragementthrough-
out my PhD journey. I am especially grateful to my advisor, Prof. Wayne Sandholtz, who
has taken me under his wing from the first day of my PhD. His kindness, patience, and sup-
port throughout my PhD career is unparalleled; I’m so grateful for having had the chance
to work with him on various research papers and projects.
I would also like to thank Patrick James for his mentorship during and even before the
start of my PhD. I’m especially grateful for the numerous research and co-authorship oppor-
tunities that Pat has offered me and for the many times he has written recommendations
for my grant and fellowship applications. I also thank Hannah Garry for offering such great
feedback on my dissertation proposal and for supporting me in grant/fellowship applications
- I am grateful to have had such an expert in asylum and refugee issues on my dissertation
committee!
A PhD can be solitary experience, however, I have never felt alone or lonely. A huge
thanks goes out to the broader POIR and academic community who have made this PhD
one of the best times of my life. I’d like to thank Jonathan Markowitz, James Lo, and
Pablo Barbera who have taught me everything I know about research design, methods, and
coding. I’d also like to thank Veridiana Chavarin who has been the best and most caring
academic/graduate program advisor in POIR. A special thanks goes out to my cohort: Na
Young, PK, Kyle, Edward, Ayana, Jarred, Joey, and Anne. Without you, this experience
would have not been the same. Thanks for the countless hours we spent together inside and
outside of classrooms!
The most special thanks goes out to Joseph Saraceno and Anne van Wijk in my cohort.
While we started as PhD peers we have become so much more over the past six years. I’m
so grateful to have them in my life and look forward to spending even more quality time
in our “joint” post-PhD lives. I also want to thank Michael Koop for always helping me,
especially with all things related finance, taxes, car, work, and visa.
I also want to thank Maria Perez who has not only been an amazing and supportive peer
and colleague in the PhD program but who has become one of my closest friends. Thanks
for always having an open ear personally and professionally. I am also grateful for Therese
Anders,JenniferRogla,andVictoriaChonChinwhohavebeenfantasticmentorsandfriends
throughout and beyond my PhD.
I’d also like to thank Thomas Valente and the Center for Applied Network Analysis
(CANA) at USC. Without Tom and the CANA group I would have never been able to learn
as much about social network analysis as I did. I’m grateful for their continuous support
iii
and help, for answering methods questions, and accepting me into their amazing community
of peer network scientists.
My overall PhD and PhD research would have not been possible without the generous
financial support of the USC Center of International Studies (CIS), the Political Science
and International Relations (POIR) program, the USC Graduate School, the USC Bartling
Research Fund. Through grants, stipends, and fellowships, I was able to fund my research
and fully focus on my research activities which included field research, summer school and
training,conferencepresentations,fundingforresearchassistantsandmysurveyexperiment,
and the provision of software/programs.
OnthetopicofPhDresearch,I’dalsoliketothankmyamazingundergraduateRAsOlivia
Griffith, Megan Bennett, and Amy Pang who have assisted me with the data collection and
verification of my original network dataset. This was a huge, arduous task, and I could have
not done it without them.
Another big thanks goes out to Nicole Smith, who has been an academic peer, refugee
and NGO expert, and long-term friend. My research has benefited from her personal and
professionalinsightsworkingatoneofthebiggestandmostimportantrefugee-focusedNGOs
- I especially thank her for connecting me with interviewees and pointing me towards the
right documents and reports.
I would also like to express gratitude to Dr. Gabriel Hubbard - my therapist of over five
years. A PhD can be a physical and mental roller coaster, and Dr. H has supported me
during this wild ride. I thank him for always being understanding and flexible, for teaching
me important coping strategies, and for being part of my self-development journey.
I also want to thank my former and current roommates: Ghazaleh and Laura. Ghazaleh
and I met and lived together during our first year in the PhD program - we learned to
navigate LA, USC, and the PhD life together. To my current roommate Laura, I want to
give thanks because I could not imagine living with anybody else. You two have made my
PhD journey easier by making coming home something I look forward to.
Abigpartofmylifehasalwaysbeenmusic. Ihavespenthoursworkingonprojectswhile
listening to music. Though it would be impossible to thank every single artist and band I
have listened to throughout these years, I want to thank my favorite artist of all time: David
Bowie. His music has followed me like a red string throughout my academic career.
Finally, I would also like to thankmy family and friends inGermany. The biggest thanks
goes to my mom Brigitte and my dad Peter who have supported my academic endeavors
throughout the years. Being so far away from home was not always easy for my parents, but
I have always felt their unconditional support, patience, and love. I could have not done it
without their continuous encouragement and deep belief in me, even during times when I
doubted myself. I also thank my two cousins Manuela and Nicole and my aunt Helga who I
havealwayshadaspecialbondwith-regardlessofdistanceandhoursapart, wehavealways
managed to remain like sisters. Thanks also to my German friends Anna, Karin, Sabrina,
and Patrick who have always made me feel welcome and made time to see me when I was
back home in Germany. I thank my cat Chipsy for providing emotional support without
even knowing it. Last but certainly not least, I want to thank myself. This may not be
common or customary, but I want to officially acknowledge all the hard work, persistence,
patience, and passion I have put into this PhD and dissertation project.
iv
Table of Contents
Dedication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Chapter 1: The Rise of Public-Private Cooperation in Refugee Protection & Empow-
erment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 A Drop of Corporate Water on Hot Stone: Why Care? . . . . . . . . . . . . 4
1.2 The Firm in World Politics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2.1 Corporate Authority in Global Governance . . . . . . . . . . . . . . . 6
1.2.2 Public-Private Partnerships (PPPs) in Global Governance . . . . . . 7
1.2.3 The For-Profit Private Sector in the Refugee Regime . . . . . . . . . 9
1.3 Definition of Refugee Protection and Empowerment PPPs . . . . . . . . . . 10
1.4 Dissertation Overview and Structure . . . . . . . . . . . . . . . . . . . . . . 11
Chapter 2: Who Cares? Measuring US Public Attitudes Towards Corporate Social
Responsibility for Protecting and Empowering Refugees . . . . . . . . . . . . . . . 14
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2 Contribution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.3 Literature Review on CSR in World Politics . . . . . . . . . . . . . . . . . . 17
2.3.1 The For-Profit Sector in Global Governance . . . . . . . . . . . . . . 17
2.3.2 The Refugee Regime and the For-Profit Private Sector . . . . . . . . 18
2.3.3 A Brief History on CSR and Corporate Activism . . . . . . . . . . . 20
2.4 Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.4.1 Refugee Protection and Empowerment CSR Compared to Other CSR
Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.4.2 Support for Corporate Commitment Towards Refugees . . . . . . . . 23
2.4.2.1 DemographicFactors: Age,Gender,andPoliticalIdentification 23
2.4.2.2 Personal Social Responsibility & Skepticism About Motives
as Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.4.3 Support For Different Types of Commitments . . . . . . . . . . . . . 26
2.5 Research Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.5.1 Survey Design and Participant Recruitment . . . . . . . . . . . . . . 29
2.5.2 Measurement of Concepts and Analyses. . . . . . . . . . . . . . . . . 32
v
2.5.2.1 Variables and Measurements . . . . . . . . . . . . . . . . . . 32
2.5.2.2 Data Cleaning and Analysis . . . . . . . . . . . . . . . . . . 33
2.6 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.6.1 Comparison Across Different CSR Activities . . . . . . . . . . . . . . 35
2.6.2 Factors That Influence Support for Refugee CSR activities . . . . . . 37
2.6.3 Regression Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.6.4 Preferences on Type of Support for Refugee CSR . . . . . . . . . . . 45
2.7 Empirical Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Chapter 3: Refugee Protection Inc.: Emergence of a New Public-Private Protection
& Empowerment Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.2 Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.2.1 New Forms of Cooperation in Global Governance . . . . . . . . . . . 52
3.2.2 The Refugee Regime and the For-Profit Private Sector . . . . . . . . 53
3.3 Contribution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.4 A Theory of Network Emergence and Tie Formation. . . . . . . . . . . . . . 56
3.4.1 Actor Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
3.4.2 Legitimacy and Reputation: Connection to Central Actor . . . . . . . 57
3.4.3 Similarity: Geographic Region and Actor Type . . . . . . . . . . . . 58
3.5 Research Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.5.1 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.5.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.5.3 Independent and Dependent Variables . . . . . . . . . . . . . . . . . 63
3.6 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
3.7 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Chapter 4: Partnering for Good: How and Why Actors Form Public-Private Partner-
ships as a Response to the Global Refugee Crisis. . . . . . . . . . . . . . . . . . . 72
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
4.2 Contribution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
4.3 Theory of Partnership Creation & Evaluation Process in Refugee Empower-
ment and Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
4.3.1 Who Initiates and Decides? . . . . . . . . . . . . . . . . . . . . . . . 75
4.3.2 Restricted Access: Reputation and Type of Industry . . . . . . . . . 76
4.3.3 Partnership Negotiation: Values, Goal, Risks, and Rewards . . . . . . 78
4.3.4 Partnership and Program Success: Evaluation of Performance . . . . 79
4.4 Research Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
4.4.1 Data Sources and Methods . . . . . . . . . . . . . . . . . . . . . . . . 82
4.4.2 Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.4.3 Ethical Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
4.5 Empirical Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
4.5.1 Why Partner? Motivations of Public and Private Actors . . . . . . . 85
vi
4.5.2 Partnership Entrepreneurs: Organizations and States as
Gatekeepers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
4.5.3 Partnership Dealbreakers: Non-Partnerships . . . . . . . . . . . . . . 90
4.5.4 Partnership Negotiation Processes . . . . . . . . . . . . . . . . . . . . 96
4.5.5 Program Monitoring and Evaluation . . . . . . . . . . . . . . . . . . 101
4.6 Empirical Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
4.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
Chapter 5: The Way Forward: The Future of Public-Private Cooperation in Refugee
Protection and Empowerment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
5.1 Chapter Synopsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
5.2 The Corporate Elephant in the Room: Is Business Involvement Good? . . . 120
5.3 The Future of Public-Private Cooperation in Refugee Issues . . . . . . . . . 123
5.4 A Future Research Agenda . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
A Appendix to Chapter 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
A.1 Treatment Text Survey . . . . . . . . . . . . . . . . . . . . . . . . . . 144
A.2 Summary Statistics for Treatments . . . . . . . . . . . . . . . . . . . 147
A.3 Difference in Means Test: Overall and Different Groups . . . . . . . . 148
A.4 Multicollinearity Checks for Regression Models. . . . . . . . . . . . . 151
A.5 Additional Robustness Check: Logistic Regression . . . . . . . . . . . 153
A.6 Additional Figures to Show Answer Distribution for Happiness About
CSR activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
B Appendix to Chapter 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
B.1 Glimpse into Datasets . . . . . . . . . . . . . . . . . . . . . . . . . . 156
B.2 2-Mode Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
B.3 Correlation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
B.4 Goodness of Fit Model 4 and MCMC diagnostics . . . . . . . . . . . 160
B.5 Odd Ratios for Model 2 and 3 . . . . . . . . . . . . . . . . . . . . . . 164
C Appendix to Chapter 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
C.1 Sample Interview Questions . . . . . . . . . . . . . . . . . . . . . . . 166
C.2 Example Recruitment Email for Interviews . . . . . . . . . . . . . . . 167
vii
List of Tables
2.1 Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.2 Variables and Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.3 Odds Ratios for Ordered Logit Regresssion Results . . . . . . . . . . . . . . 44
2.4 Summary of Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.1 Results for ERGMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.2 Odds Ratios Model 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.1 Short Summary of Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . 80
4.2 Overview of Interview Participants . . . . . . . . . . . . . . . . . . . . . . . 83
4.3 Summary of Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
5.1 Summary Statistics for Refugee CSR Treatment . . . . . . . . . . . . . . . . 147
5.2 Summary Statistics for CSR for the BLM movement . . . . . . . . . . . . . 147
5.3 Summary Statistics for CSR for the fight against climate change . . . . . . . 147
5.4 T-test: Diff-in-Means for Refugee and Climate Group . . . . . . . . . . . . . 148
5.5 T-test: Diff-in-Means for Refugee and BLM Group . . . . . . . . . . . . . . 148
5.6 T-test: Diff-in-Means for Climate and BLM Group . . . . . . . . . . . . . . 148
5.7 T-test: Diff-in-Means for Women vs. Men . . . . . . . . . . . . . . . . . . . 149
5.8 T-test: Diff-in-Means for Independents vs. Democrats . . . . . . . . . . . . . 149
5.9 T-test: Diff-in-Means for Democrats vs. Republicans (Level of Support) . . . 149
5.10 T-test: Diff-in-Means for Independents vs. Republicans . . . . . . . . . . . . 149
5.11 T-test: Diff-in-Means for High vs. Low Attention to Social Issues . . . . . . 150
5.12 T-test: Diff-in-MeansforHighvs. LowAttentiontoBrands’SocialEngagement151
5.13 T-test: Diff-in-Means for High vs. Low Skepticism about Companies’ Motives 151
5.14 Snapshot Meta Data of PPP Programs . . . . . . . . . . . . . . . . . . . . . 156
5.15 Snapshot 2-Mode Edgelist . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
5.16 Contingency Geographic Region - Actor Type . . . . . . . . . . . . . . . . . 158
5.17 Contingency UNHCR Connection - Actor Type . . . . . . . . . . . . . . . . 159
5.18 Contingency UNHCR Connection - Geographic Region . . . . . . . . . . . . 159
5.19 Main Effects (Model 2) Odd Ratios . . . . . . . . . . . . . . . . . . . . . . . 165
5.20 Similarity (Model 3) Odd Ratios . . . . . . . . . . . . . . . . . . . . . . . . . 165
viii
List of Figures
1.1 Dissertation Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.1 Theory of Refugee Empowerment & Protection CSR . . . . . . . . . . . . . 28
2.2 Issue Awareness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.3 Difference in Means for CSR Treatment Groups . . . . . . . . . . . . . . . . 37
2.4 Mean Level of Support By Gender. . . . . . . . . . . . . . . . . . . . . . . . 38
2.5 Mean Level of Support By Political Party Identification . . . . . . . . . . . . 39
2.6 Mean Level of Support By Age Groups . . . . . . . . . . . . . . . . . . . . . 40
2.7 Mean Level of Support Based on Personal Social Responsibility and Awareness 41
2.8 Mean Level of Support Based on Skepticism about Companies’ Motives . . . 42
2.9 Mean Level of Support Based on Skepticism about Companies’ Motives . . . 43
2.10 Type of Support People Prefer Companies to Provide to Refugees . . . . . . 46
3.1 Refugee Protection Network . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.1 Theory of Partnership Creation & Evaluation Process in Refugee Empower-
ment and Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.1 Refugee Treatment Text in Survey . . . . . . . . . . . . . . . . . . . . . . . 144
5.2 BLM Treatment Text in Survey . . . . . . . . . . . . . . . . . . . . . . . . . 145
5.3 Climate Change Treatment Text in Survey . . . . . . . . . . . . . . . . . . . 146
5.4 Multiple Comparison with Bonferroni Correction For Age . . . . . . . . . . . 150
5.5 Correlation Plot IVs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
5.6 Logistic Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
5.7 Level of Happiness by Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
5.8 Level of Happiness by Education . . . . . . . . . . . . . . . . . . . . . . . . 155
5.9 Level of Happiness by Political Party . . . . . . . . . . . . . . . . . . . . . . 155
5.10 2-Mode Refugee Protection Network . . . . . . . . . . . . . . . . . . . . . . 158
5.11 Goodness of Fit Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
5.12 Goodness of Fit Degree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
5.13 Goodness of Fit ESP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
ix
Abbreviations
BLM Black Lives Matter
CIs Confidence Intervals
CSA Corporate Social Activism
CSR Corporate Social Responsibility
ERGMs Exponential Random Graph Models
GG Global Governance
GLMs Generalized Linear Regression Models
IFC International Finance Cooperation
IGOs Inter-governmental Organizations
IR International Relations
ISO The International Organization for Standardization
LGBTQI+ Lesbian, Gay, Bisexual, Transgender, Queer and Intersex
MNCs Multinational Corporations
NGOs Non-governmental Organizations
NYU New York University
PPP Public-Private Partnership
R&D Research and Development
SNA Social Network Analysis
Tent Tent Partnership for Refugees
UN United Nations
UNDP United Nations Development Programme
UNHCR The United Nations High Commissioner for Refugees
x
Abstract
Theinternationalcommunityfacesadilemma: whilenumbersofrefugeesarerapidlyincreas-
ing, statesandorganizationshaveyettofindadequatesolutionstoaddressthiscrisis. Anew
trend is the active participation of the for-profit private sector consisting of businesses and
multi-national corporations. While the business community has long been engaged in inter-
national investment and philanthropic activities, their active corporation with public actors
for the purpose of creating and executing refugee protection and empowerment programs,
is rather novel. This dissertation project sets out to explore the emergence of cooperation
between businesses, governments, and organizations: a public-private network that devel-
ops collaborative solutions to protect, support, and empower asylum seekers and refugees.
It investigates why and how these public and private actors cooperate to tackle the global
refugee crisis. Through a multi-method approach that combines social network analysis,
content analysis of stakeholder interviews and public reports, and a survey experiment, it
exposes motivations, factors, and overall dynamics that are at play when various actors find
and form partnerships and evaluate joint projects and programs. It deconstructs the com-
plex partnership formation and evaluation process and finds that factors such as geography,
type of industry, goals, visibility, and reputation influence actors’ ability to join this network
and cooperate. While the partnership creation and program evaluation processes are ardu-
ous, the resulting public-private partnerships often create innovative solutions for refugee
situations and have the potential to revolutionize humanitarian program/project evaluation
metrics. This dissertation concludes by offering a discussion on the future of cooperation
in the refugee protection and empowerment space and laying out a research agenda beyond
this project.
xi
Chapter 1
The Rise of Public-Private Cooperation in Refugee
Protection & Empowerment
”
I have worked alongside my refugee brothers and sisters for many years,
and they are a testament to the power of the human spirit. I am grateful
that so many people are now paying attention to the refugee crisis, and we
cannot let this moment pass without change. It is critical that we make
sureourwordstranslateintoaction, andthatweworktogetheronconcrete
solutions.
— Hamdi Ulukaya
(Chobani Founder and CEO)
I had just started my masters degree at the University of Oklahoma when Europe ex-
perienced the arrival of millions of people fleeing their homes in 2015-16. I devoted my
graduate studies to understanding the economic, security, and legal aspects of forcibly dis-
placed people arriving in Europe. For my masters thesis, I spent the summer of 2016 in
Germany interning in the German Bundestag and working in refugee camps to trace the
rapidly changing immigration and asylum policy making process.
While millions of asylum seekers were arriving, European politicians stood in disagree-
ment over an appropriate response. The German government, which had initially advocated
and employed a “Willkommenskultur” (welcoming culture), was slowly returning to a more
conservativepolicyapproachdueto growinganti-refugeesentimentofthepublic, limitations
1
of infrastructure and resources, and an absent unified European approach. Public and pri-
vate non-profit entities including local/regional governments, organizations, and civil society
complained about lack of funding, insufficient resources, and unpreparedness. A solution, or
at a minimum a more promising approach, to support and protect asylum seekers seemed
distant.
In June 2016, President Obama issued a call of action inviting and encouraging the
corporate sector to actively participate in creating solutions for asylum seekers and refugees
because “more than government action” was necessary to tackle the escalating refugee crisis
(White House, 2016b). Only a few months after the call already 51 companies had issued
high-scale and noteworthy commitments to educate and employ asylum seekers and refugees
as well as support governments and organizations (White House, 2016a). In addition, the
Tent Partnership for Refugees, a global business coalition bringing together all kinds of
corporate actors interested and committed to creating sustainable solutions for refugees,
launched.
Attheheightoftherefugeecrisis,thebusinesssectoremergedasanewandrelevantactor
in the refugee support and protection system. As the refugee regime has often been seen
as a state-centric and, to a certain extent, organization-centric enterprise, with governments
as the law-making and asylum-granting entities and organizations delivering services and
protection and providing platforms for mediators and meetings, the active engagement and
the role of the for-profit sectors has only received limited attention.
This is where this dissertation steps in: it attempts to fill a critical void in the existing
theoretical and empirical scholarship on business engagement in the global refugee crisis.
As this is a huge endeavor and involves a research agenda beyond this dissertation, I focus
on the emergence of cooperation between businesses, governments, and organi-
zations: a public-private network that develops collaborative solutions to protect, support,
and empower asylum seekers and refugees.
2
Given that businesses and corporations have limited experience and expertise working
with forcibly displaced people, have trouble navigating the extensive legal landscape sur-
rounding immigration and asylum, and face difficulties understanding and adapting to com-
plex cultural challenges when engaging with asylum seekers and refugees, there is almost
always an element of cooperation between businesses and public or non-profit private actors.
Thus, this dissertation sets out to investigate the following questions:
1. Whatarethemotivationsforthecorporatesectortogetinvolvedinrefugeeandasylum
support and protection in the first place? How does the public react to business
engagement in refugee protection and empowerment?
2. What factors make forming partnerships more likely; and who partners up with whom
in this newly forming public-private network?
3. How and why do public (governments, international organizations), non-profit private
(non-profits), and for-profit (business, corporations) actors form partnerships? How
do they evaluate the performance (or success) of the resulting collaborative programs
or projects?
To answer these questions, I employ a cross-disciplinary approach (IR, migration &
refugee studies, American politics, business, and marketing) and rely on a multi-method
research design. This is essential as I found that no single theory, school of thought or
method could capture and account for the complexities of various actors from vastly differ-
ent backgrounds joining together to tackle one of the worst humanitarian crises. In other
words,drawingfromdifferenttheoriesandtheirassumptionsenabledmetoconstructholistic
theoreticalexpectationsforeachchapterandinvestigatemyresearchquestionsfrommultiple
angles.
Whatthisdissertationfindscanbebrieflysummarizedinthefollowingway: thepublic(in
the US) welcomes business engagement in the refugee protection and empowerment space,
and the for-profit sector is motivated to actively support, empower, and protect forcibly
3
displaced people. However, gaining access to this network (and to refugees and asylum
seekers) and being able to form partnerships is difficult, especially for corporate actors.
There are static factors (i.e. geographic location or type of industry/sector, etc.) as well as
more fluid factors (i.e. reputation, shared goals/values, etc.) that influence actors’ ability to
join and cooperate. Overall, the partnership creation and program evaluation processes are
arduous but rewarding for actors; there seems to be great potential for creating innovative
solutions and novel program evaluation metrics.
The next sections will first outline why we should care about corporate engagement in
refugeeempowermentandprotectiontobeginwith, beforesituatingthisdissertationproject
in the broader literature. The chapter concludes by outlining the remaining chapters and
addressing how they fit together to answer the above-stated research questions.
1.1 A Drop of Corporate Water on Hot Stone: Why
Care?
Why should we care about businesses becoming active in refugee protection and empower-
ment? Scholarsandpractitionersbelievethattheinvolvementofbusinessesandcorporations
could produce more sustainable and promising solutions to the crisis (Betts et al., 2017).
However, could this be true? Asked differently, are the contributions of the for-profit sector
sufficient and significant enough to make a difference; do they outweigh, or at a minimum,
measure up to those of governments or organizations? Why should we take the business
sector seriously, if their contributions are merely a drop of water on a hot stone?
Not only are the contributions of the for-profit sector diverse and noteworthy, businesses
are often also able to mobilize money and donations more quickly than traditional actors.
Forexample, whentheUnitedNationsHighCommissionerforRefugees(UNHCR)calledfor
emergency funding for the escalating humanitarian crisis in the Ukraine, the private sector
donated a “record-breaking” amount of US $200 million in a little over two weeks (UNHCR,
4
2022). With this, the for-profit private sector mobilized and delivered almost half of the US
$510 million initially requested emergency funding for the Ukrainian region.
However, corporate commitment extends beyond financial donations and includes inno-
vative, multi-million dollar projects and programs to help and empower forcibly displaced
people. The following list only aims to exemplify how the business sector has been and
continues to work towards improving refugees’ educational, employment, and/or living situ-
ations:
• Better Shelter: The IKEA foundation has developed a sustainable, safe, and inno-
vative temporary shelter solution.
1
The solar-powered shelter easily outlasts a typical
emergency tent (6 times longer). The project has won the Beazley design of the year
2016 by London’s Design Museum (Wainwright, 2017).
• Amazon: The company has announced its intention to hire 5,000 refugees over the
next three years. Employees gain access to Citizenship Assistance Portal and will get
reimbursed for certain government filing fees ( Welcome Door program). Amazon sup-
ports refugees with pre-paid college tuition and ESL support (Career Choice program)
(Tent, 2022a).
• Cisco: Together with NetHope, Cisco improved connectivity for over 600,000 asylum
seekers and refugees in Greece by installing over 80 Meraki-based Wifi hotspots (Tent,
2022b).
• Coursera: The company is expanding its existing program Coursera for Refugees,
which has already served more than 150,000 learners and included more than 5,000
courses. Thenewcommitmentincludesprovidingtrainingresourcesfor15,000refugees
worldwide in the next three years (Tent, 2022c).
• LinkedIn: With its Welcome Talent program, LinkedIn has already supported thou-
sands refugees in countries like Sweden and Canada. The company is expanding its
1
See Better Shelter Homepage.
5
efforts and will provide training and employment resources to 18,000 refugees in the
coming three years. Additionally, LinkedIn will offer 2,000 LinkedIn Premium licenses
to refugees, one-on-one coaching to 1,000 refugees, and assist refugees to find a job
through its tailored refugee-specific jobs portal (Tent, 2022d).
Undeniably, the private for-profit sector’s involvement in refugee protection and empow-
erment is significant and diverse, touching almost every governing aspect of refugee policy
making, crisis management, and integration efforts. Not only is the private for-profit sec-
tor a considerable financial donor, often contributing more than other non-state actors or
even governments
2
, the sector also offers innovative approaches, products, and solutions
that governments and organizations direly seek and need. It is therefore sensible to consider
businesses as an increasingly important and influential actor in the humanitarian field.
Scholars have addressed shifting legitimacy and authority patterns in the international
system and the rise of non-state actors, with particular attention to corporate authority.
Similarly, research has investigated the phenomenon of public-private partnerships - the
tendency of public, not-for profit private, and corporate actors to join forces and develop
collaborative projects and programs. The next section will review the scholarship landscape
and situate this dissertation project in the larger literature.
1.2 The Firm in World Politics
1.2.1 Corporate Authority in Global Governance
Thephenomenonoftransnationalactorssettingand/orchangingnormsandrules,addressing
issues, and providing goods and services on the world stage has been a central theme in
International Relations (IR) literature. Scholars often refer to this as global governance
2
See for example budget contributions to the UNHCR. Considering the list of all major donors on pages
13-15, we see that private donations figure centrally in the UNHCR budget. In 2020 for example, GAP Inc.
donated US $22,680,054, which is more than what USA for UNHCR donated (pg.17).
6
(GG),whichis“thecollectiveeffortbysovereignstates,internationalorganizations,andnon-
state actors to address common challenges and seize opportunities that transcend national
frontiers” (Patrick, 2014, pg.59).
Especially private entities have gained key governance roles, prompting scholars to an-
alyze how multinational corporations (MNCs) and businesses/firms shape standards/rules
and practices (Abbott and Snidal, 2010; B¨ uthe and Mattli, 2011; Hall and Biersteker, 2002;
Roger and Dauvergne, 2016; Ruggie, 2014). The business sector is involved in global rule-
making(B¨ utheandMattli,2011)aswellasdomesticlobbyingandpolicymaking(Newhouse,
2009; Hafner-Burton and McNamara, 2019; Facchini et al., 2011; Freeman, 1995; Menz,
2011;Tichenor,2002). Theprivatefor-profitsectorhasalsobecomeimportantfordelivering
government-type services and security functions (Haufler, 2010; Jones Luong, 2018). How-
ever, the private for-profit sector often does not govern alone: businesses tend to collaborate
with governments and organizations to produce, offer, and update existing/new policies,
products, projects, and/or services.
1.2.2 Public-Private Partnerships (PPPs) in Global Governance
With the emergence of the private for-profit sectors as a central actor in international rela-
tions, we see more diverse forms of cooperation. State and non-state actors do not operate
in isolation; they work together to find and develop solutions to global issues and to provide
goods/services. Scholars and practitioners refer to this type of hybird governance as public-
privatepartnerships(PPPs). Thesenovelpartnershipshavebeendiscussedasaninstrument
to enhance governance effectiveness and good governance (Andonova, 2017; Chan, 2012).
PPPs usually consist of at least one private for-profit actor and at least one not-for-profit
or public entity. However, what is included in public and private? Public entities include
governments and international organizations such as the UNHCR, World Bank, etc. The
private sector category captures for-profit businesses and corporations. But what about
NGOs? Scholars that study PPPs recognize that NGOs are “private in the sense that
7
they do not belong to a governmental structure, yet they seek to promote public interests”
(Reich, 2002, pg.4-5). As a result, to more accurately capture public vs. private, scholars
categorize such organizations as “belonging to the public side of the equation of public-
private partnerships, while recognizing that NGOs are often considered as a third sector”
(Reich, 2002, pg.5). In my dissertation, I follow this distinction between public and private
actors.
The rise of global public-private partnerships (PPPs) is commonly attributed to struc-
tural factors such as increased globalization and intensified crises. With the world becoming
moreinterconnected,privateactorshavesuccessfullyextendedtheirreachbeyondtraditional
businessoperationsintovariousglobalgovernanceissuesconcerningtheenvironment,health,
and human rights. In addition, the escalation of crises has activated the private sector to
remedy “market and intergovernmental failures” (Reinicke, 1998). Consequently, scholars
suggest that PPPs are formed consciously and purposefully by rational actors who aim to
remedy global issues. Due to the complexity of global problems, neither public nor private
actors alone are able to effectively address problems, thus making PPPs especially appealing
(Reinicke et al., 2000; Reinicke and Witte, 2000; Nelson, 2002).
However, it appears unlikely that PPPs mainly form on a normative basis of wanting to
provideglobalsolutions. Assomestudieshaveshown, PPPsoftenemergeinareaswherethe
interests of powerful actors such as northern states and international organizations converge
(Andonova and Levy, 2003; Hoering, 2003; Sch¨ aferhoff et al., 2009). Other studies have
investigated the motivations and interests of the respective actors involved.
InternationalorganizationsareincentivizedtocreatePPPsinordertoremedythepublic
fundingandlegitimacycrisis. Therisingdemandintechnicalexpertiseandspecificsolutions
makespartneringupwithbusinessesandcorporations,thathavethenecessaryresourcesand
know-how, attractive for IGOs (Bull et al., 2004; Tesner and Kell, 2000). NGOs may also
want to diversify their funding resources and advance their status, however, they are also
8
interested in PPPs because of their desire to more actively shape and monitor state and
business policies and behavior (Andonova and Levy, 2003).
Corporate motivations for entering PPPs may revolve around accessing new markets and
businesspartnersaswellassignalingsocialresponsibilityandenhancingreputationandpub-
lic image (Andonova and Levy, 2003; United Nations Industrial Development Organization,
2002). Finally, state actors welcome the formation of PPPs as they allow governments to
outsource certain responsibilities and tasks and to tap into information and knowledge of
non-state actors. Thus, PPPs may not purely be a a result of closing governance gaps, but
instead a product of fostering actors’ interests and capacities (Andonova, 2006; Reinicke and
Witte, 2000; Sch¨ aferhoff et al., 2009).
While the structural/normative account may offer some insights into the potential mo-
tives of actors to join and form PPPs, it tells us little about the actual factors, processes,
and dynamics of partnership creation. Put differently, this account may shed light onto why
they form, but not how they form and who partners up with whom and why.
1.2.3 The For-Profit Private Sector in the Refugee Regime
As pointed out earlier in this introduction, the private for-profit sector has become increas-
ingly relevant and active in refugee empowerment and protection. Nevertheless in scholar-
ship, there is still a tendency to focus on the role and contributions of the not-for profit
sector, while sidelining the for-profit sector (Cammett et al., 2014; Campbell et al., 2019;
Weiss, 2013; Zyck and Armstrong, 2014)
While scholarly interest in the business sector has burgeoned in the past few years, re-
searchisoftenlimitedtocasestudiesthatilluminatethepotentialofthefor-privatesectorin
refugee protection and empowerment (Drummond and Crawford, 2014; Omata and Kaplan,
2013; Zyck and Armstrong, 2014). This research also tends to undermine the complexity
and diversity of private actor engagement and portrays private actors “uniquely as donors,
suppliers, or technical advisers to humanitarian agencies” (Betts et al., 2017, pg.188). Other
9
scholarship has often focused on the private sector’s engagement in migration control re-
sponsibilities including offshoring, outsourcing, and regionalizing, which are practices geared
towards limiting the arrival of irregular migrants and asylum seekers (Gammeltoft-Hansen
and Sørensen, 2013; Hern´ andez-Le´ on, 2008).
This survey of the literature exposes two main shortcomings: 1) limited understand-
ing of PPP emergence (why and how do they emerge, and who partners with whom and
why?) and 2) lack of theoretical and empirical consideration of the for-profit sector’s role in
refugee protection and empowerment. This is where this dissertation steps in to bring it all
together: it investigates the rise of public-private cooperation in the refugee empowerment
and protection space.
This dissertation centers its attention on the newly emerging for-profit sector and the
sector’s effort to become involved in refugee protection and empowerment and form partner-
ships with states and organizations. The next section will offer a brief definition of refugee
protection and empowerment PPPs, before the final section provides an overview of the
dissertation and summarizes the focus of the dissertation chapters.
1.3 DefinitionofRefugeeProtectionandEmpowerment
PPPs
For the purpose of this dissertation, refugee protection and empowerment is defined as any
activity that aims to enhance the rights, livelihood, and integration of forced migrants.
This may range from providing financial or humanitarian support for humanitarian crises
and/orrefugeesituationstocreatingjobortrainingprogramsforforcedmigrantstoenhance
social and/or economic integration in host countries.
Refugee protection and empowerment PPPs are then collaborative projects/initiatives
between public actors and private actors that specifically focus on improving the rights and
lives of forced migrants. It is important to note that these PPPs do not exclusively focus on
10
refugees as the term refugee refers to a specific legal status in International Relations and
Law.
3
Instead, these PPPs are often designed to help, support, and empower asylum seekers,
refugees, internally displaced people, and other types of forced migrants. This is sensible
as global challenges and humanitarian crises have been consistently and rapidly changing
and have given rise to more diverse hardships, dangers, and human insecurity that are not
necessarily limited to and/or a result of war or conflict.
Thus, this dissertation uses the broader understanding of “refugee” and investigates
public-private partnership programs and initiatives that serve various types of forcibly dis-
placed people.
1.4 Dissertation Overview and Structure
Figure 1.1 provides a visual representation of how the dissertation unfolds over the course of
the following three chapters. The three chapters address different research questions that all
centeraroundthetopicof emergingpublic-privatecooperationinrefugeeprotection
andempowerment. Asmentionedearlier,theydrawupondifferenttheoretical,evencross-
disciplinary, perspectives including international relations, migration studies, business, and
marketingaswellasdiffersignificantlyintheirempiricalapproachdependingontheresearch
question at hand.
Chapter 2 sets the stage by exploring one of the most pressing questions: why would
businessesevengetinvolvedinrefugeeprotectionandempowermentinthefirstplace? While
there are different answers to this question, one of the most common ones is: pressure from
the public (or in this case: from customers and consumers). We know from marketing and
3
The 1951 Refugee Convention is a key legal document and that defines a refugee as “someone who is
unable or unwilling to return to their country of origin owing to a well-founded fear of being persecuted for
reasons of race, religion, nationality, membership of a particular social group, or political opinion.” Refugees
therefore have often fled war, violence, conflict or persecution and have crossed international borders to seek
asylum in a safe country. To receive refugee status, people need to go through a legal and administrative
processwheregovernmentsortheUNHCRdetermineeligibility. Thus, refugeestatusisanarrowlegalstatus
that does not capture other types of forced migrants such as internally displaced people or asylum seekers.
11
Figure 1.1: Dissertation Overview
corporate social responsibility literature that there is a link between stakeholder pressure
and firms becoming more socially and environmentally active (Mohr et al., 2001; Sen and
Bhattacharya, 2001; Smith, 2016). However, how does the public feel about companies
gettinginvolvedinrefugeeempowermentandprotection? Throughasurveyexperiment, the
chapter investigates attitudes and opinions of the US public towards business engagement in
refugee protection and empowerment. It argues and finds that pressure from consumers and
customersisindeedpresent: manyAmericanssupportbusinessessupportingandempowering
refugees. Inaddition,thechapterrevealshowAmericanswouldliketoseecompaniessupport
and empower refugees (i.e. the type of support Americans would like to see).
Given that the public welcomes companies’ engagement and support for refugees, and
companies are motivated to contribute, how do companies best participate? As briefly
addressed earlier, gaining access to forcibly displaced people and navigating the legal and
12
cultural landscape is difficult for companies as they have often had limited experience and
exposure to working with such a vulnerable population. This is why most private for-
profit actors (need to) partner with organizations and/or governments. Consequently, we
are witnessing the emergence of a public-private partnership network in refugee protection,
which is the focus of chapter 3 and 4.
Chapter 3 analyzes the formation of this emerging public-private partnership network.
Particularly, it investigates what factors make partnerships in this network more likely, and
who partners up with whom? It claims that businesses have an overall harder time forming
partnerships than any other actors (organisations, governments, academic intstituions), but
that being affiliated with a big organization like the UNHCR increases partnership chances.
Moreover, it argues that actors with similar attributes more readily partner with one an-
other rather than with unlike actors. The chapter introduces my original refugee protection
public-private partnership consisting of over 200 actors and more than 100 verified collabo-
rative programs/projects and employs social network analysis to analyze network formation
dynamics.
Chapter4picksupwherechapter3leftoff,investigatingtheactualpartnershipformation
process, motivations of various actors, and challenges of forming partnerships and programs.
This chapter underlines some of the findings of chapter 3 while revealing additional (often
quantitatively difficult-to-measure)factors includingreputation, visibility, shared valuesand
goals. It confirms how difficult it is for businesses to break into the refugee protection and
empowerment space. It further illuminates the challenging process of finding partners and
forming partnerships as well as creating and evaluating collaborative projects and programs.
The chapter relies on semi-structured interviews with various public and private actors who
aredirectlyresponsibleforpartnershipandprogramsinrefugeeprotectionandempowerment
complemented by evidence from conferences/meetings and reports.
13
Chapter 2
Who Cares? Measuring US Public Attitudes Towards
Corporate Social Responsibility for Protecting and
Empowering Refugees
2.1 Introduction
Over the past few years a new trend has emerged: corporate actors have increasingly com-
mitted to protecting and empowering refugees and asylum seekers. Companies engage in
versatile activities to support forcibly displaced people. For example, Amazon, Hilton, and
Marriott have employed and pledged to hire thousands of refugees; Airbnb has provided free
short-term housing for displaced people with its Open Homes project; HP, Microsoft, and
Google continue to provide training and skill courses to refugees to enhance their digital, IT,
and business skills.
1
However, why would companies and brands decide to get involved in refugee protection
and empowerment? Engaging in refugee protection and empowerment is rather costly: it
takes time and money to support, educate, train, and employ forced migrants compared
to other societal groups. Out of all potential social and environmental causes, why would
companies choose refugee protection and empowerment?
1
See the Tent Partnership website for more detailed information and a list of companies and their com-
mitments.
14
One of the predominant reasons discussed in literature is that consumers and customers
demand and expect companies to act socially and environmentally responsible (Mohr et al.,
2001; Sen and Bhattacharya, 2001). However thus far, there is no research on whether and
to what extent American consumers and customers support corporate social responsibility
(CSR) initiatives for refugee protection and empowerment.
This project fills this gap and explores attitudes and perceptions of existing and poten-
tial US customers towards companies supporting and empowering refugees. It investigates
whether stakeholder pressure is actually present and could be a valid reason/explanation
for companies/brands getting involved in refugee protection and empowerment. Through
a survey experiment, I answer the following two questions: 1) how and to what extent
do consumers approve of companies supporting and empowering refugees and 2) whether
consumers tend to support corporate activism for refugees more/less compared to other
social/environmental causes (such as the BLM movement and climate change).
I argue that there are differences in support between the treatment groups. While I
expect to find that CSR activities around climate change and the BLM movement receive
the highest support, I actually find that people support CSR for climate change and refugee
protection the most. Americans support companies stepping up and becoming engaged in
refugee protection and empowerment. The high support levels for CSR activities in refugee
protection and empowerment help explain why companies may have become increasingly
involved and continue to expand their engagement in the refugee cause.
IextendthisanalysistofurtherexplorewhichfactorsinfluenceAmericanlevelsofsupport
for refugee protection and empowerment CSR activities, and what type of support people
would like to see from companies. I contend that the most important drivers for support
are 1) gender, age, and political ideology, 2) peoples’ personal social responsibility and
awareness, and 3) skepticism about companies’ motives to get engaged in refugee protection.
Myanalysisshowsthatgender, politicalpartyidentification, andpersonalresponsibilityand
awareness are indeed most influential for higher support levels. I also find that the more
15
skeptical people are about companies’ motives/intentions as to why they get involved in
refugee protection the less they support CSR for refugee protection. In terms of type of
support, I find that Americans prefer companies to actively train and hire refugees as well
as donate food and clothes.
The paper proceeds with the contribution/literature review and theoretical expectations
before introducing the research design. This is followed by the empirical analysis and dis-
cussion. The study concludes by suggesting future avenues for research.
2.2 Contribution
Thisprojectoffersempirical, theoretical, andpolicycontributions. Empirically, itrepresents
one of the only studies that investigates how Americans feel about companies supporting
and empowering refugees. It offers a novel body of data that reveals levels of support for
different CSR activities, and in particular for CSR activities for refugee protection. Further,
it provides an in-depth analysis of factors that may influence peoples’ support for CSR
activities for refugee protection.
Theoretically, this study bridges scholarship in international relations, refugee and mi-
gration studies, and business. It draws upon knowledge across these different disciplines to
develop a comprehensive theoretical framework. It goes beyond existing literature on cor-
porate activism and CSR by comparing different CSR activities to one another, therewith
offering theoretical justifications on how and why companies choose certain CSR activities.
It lends additional support to the foundational assumption that companies engage in CSR
activities because of consumer pressure.
When it comes to policy implications of the findings, this study suggests that corporate
involvement in the humanitarian space will likely continue to grow. As many participants
supportthatcompaniesstepupandhelp,thefor-profitsectorwillplayanincreasinglyactive
and important role in addressing social and environmental issues. This affects country and
16
regional laws and regulations, national and regional integration and immigration strategies,
andthedistributionofresourcesandaid. Astherestofthisdissertationilluminates, govern-
ments and organizations have been actively partnering up with corporate actors to develop
new solutions to the refugee crisis.
2.3 Literature Review on CSR in World Politics
2.3.1 The For-Profit Sector in Global Governance
Scholars have engaged in fruitful theoretical and empirical endeavors that explore global
governance (GG) as well as the involvement of transnational actors to set/change norms
and rules, address global problems, and/or to provide common goods. Broadly, GG refers
to “the collective effort by sovereign states, international organizations, and nonstate actors
to address common challenges and seize opportunities that transcend national frontiers”
(Patrick, 2014, pg.59).
More recently, literature has engaged with the changing architecture of GG given the in-
creased diversification and emergence of new central actors as well as institutional fragmen-
tation. Especially private entities have gained key governance roles, prompting scholars to
analyzehowmultinationalcorporations(MNCs)andbusinesses/firmsshapestandards/rules
and practices (Abbott and Snidal, 2010; Haufler, 2010; B¨ uthe and Mattli, 2011; Graz and
N¨ olke, 2008; Hall and Biersteker, 2002; Roger and Dauvergne, 2016; Ruggie, 2014). B¨ uthe
and Mattli (2011) engage in a detailed analysis of the development of private regulation
and the role of private actors in global rule making. In The Emergence of Private Author-
ity in Global Governance, the authors of the various chapters document the move towards
privatization and investigate practices of diverse actors (financial institutions, corporations,
terrorists etc.) and consequences of their involvement in world politics.
Theimpactofthefor-profitprivatesectorisalsofeltthroughlobbyingandinterestgroup
activitiesonthedomesticlevel. Newhouse(2009)claimsthatthemostpowerfullobbyistson
17
CapitolHillconsistoffactionsrepresentingoverseasbusinessinterests. Theauthordescribes
howtheselobbyistsactivelyinfluenceUSforeignpolicymaking. Thisalignswiththeworkof
Hafner-Burton and McNamara who investigate the relationship between corporate lobbying
and US human rights policy. The authors find that lobbying campaigns may lead to policy
changes (Hafner-Burton and McNamara, 2019, pg.116). Other literature finds support for
the influence of the business sector on immigration policy and employment rights (Facchini
et al., 2011; Freeman, 1995; Menz, 2011; Tichenor, 2002).
Besides more indirect influence, the private sector has increasingly moved towards direct
engagement: it delivers social services and carries out state security functions. Jones Luong
(2018) illustrates how oil companies may actually improve social and economic conditions
in states by providing public goods and social services. Post investigates the “contracting
out” of social welfare and health services to private firms and illustrates how this practice
changes responsibilities between pubic and private sector and is essential to understanding
the development of the welfare state (Post, 2018). Haufler analyzes how the private sector
gained influence in conflict, corruption, and criminality management. The author finds that
one reason for private actors to get involved in conflict resolution is the growing trend of
corporate social responsibility (Haufler, 2010, pg.113).
Undeniably, the private sector has become a central player in world affairs. The next
section will briefly address the for-profit sector in refugee protection and empowerment.
2.3.2 The Refugee Regime and the For-Profit Private Sector
In light of institutional proliferation, the diversification of actors in the refugee regime
2
has
also received significant attention in migration and refugee studies. While often seen as
a state-enteric enterprise, given that it traditionally has been public actors such as states
and IGOs that are supposed to provide protection and asylum, focus has shifted towards
2
The refugee regime refers to institutional arrangements that ensure migrant and refugee rights and
lay out state responsibilities. The main treaty is the Refugee Convention with the UNHCR as the main
organization.
18
the importance of non-state actors. For example, research has illustrated the importance of
NGOs stepping in to provide services and protection (Cammett et al., 2014; Campbell et al.,
2019).
However, the role of private for-profit actors in the refugee regime has received little
attention. This stands “in contrast to the vast literature that now exists relating to the
role of the private sector in other areas of global governance” (Betts et al., 2017). Most
existing literature explores private actor participation in international political economy
areas such as international trade, health, environment (Brown and Woods, 2007; Clapp,
2009; Cutler et al., 1999; Falkner, 2005; Fuchs, 2007; Hall and Biersteker, 2002; Levy and
Newell, 2005; May, 2006; Ruggie, 2007), while sidelining the private sector in the human
rights and humanitarian regime (Weiss, 2013; Zyck and Armstrong, 2014). Besides some
work on offshoring and outsourcing, practices that refer to transferring migration control
responsibilities to private actors and third states, there has been a lack of considering and
investigating the role of for-profit actors in the refugee regime. (Gammeltoft-Hansen and
Sørensen, 2013; Hern´ andez-Le´ on, 2008).
3
Whilesomecasestudiesonthepotentialoftheprivatesectorandrefugeeprotectionexist,
(DrummondandCrawford,2014;OmataandKaplan,2013;ZyckandArmstrong,2014),this
research undermines the complexity and diversity of private actor engagement and portrays
privateactors“uniquelyasdonors,suppliers,ortechnicaladviserstohumanitarianagencies”
(Betts et al., 2017, pg.188). However, private sector involvement is eclectic and goes far
beyond indirect contributions.
Considering the growing involvement and commitment of the for-profit private sector
in the refugee protection and empowerment space, this study aims to investigate how the
public perceives such CSR and corporate activism. To set the stage, the next section offers
a succinct overview of CSR and corporate activism.
3
This point is also stressed in (Betts et al., 2017).
19
2.3.3 A Brief History on CSR and Corporate Activism
As illustrated above, company and brand involvement in social and environmental issues is
not a new development. However, the concepts and practices of CSR and corporate activism
haveevolvedovertime(Carroll,2021;Latap´ ıAgudeloetal.,2019). Whilescholarsandprac-
titioners disagree on a universal definition of CSR (Dahlsrud, 2008), a commonly referenced
and accepted definition was set forth by The International Organization for Standardization
(ISO). The ISO (International Organization for Standardization, 2010, para 2.18) describes
CSR as:
Theresponsibilityofanorganizationfortheimpactsofitsdecisionsandactivities
on society and the environment, resulting in ethical behavior and transparency
which contributes to sustainable development, including the health and well-
being of society; takes into account the expectations of stakeholders; complies
with current laws and is consistent with international standards of behavior; and
is integrated throughout the organization and implemented in its relations.
The concept and practice of corporate activism is related but distinct to CSR. Corporate
activism refers to companies taking a visible and vocal stance on sensitive economic, social,
and/or environmental topics (Olkkonen and J¨ a¨ askel¨ ainen, 2019). Corporate social activism
(CSA)thuscomplimentsandexpandsCSRasitnotonlycapturesethicalandsocialbusiness
practicesbutrequiresfirmstoreinforceandpublicizetheirvalueson, often, contestedissues.
Traditional CSR and corporate activism started with the profit sector’s commitment and
engagement in more ethical and sustainable practices. Early CSR activities revolved around
environmental sustainability of production and products (Joireman et al., 2015; Yoon et al.,
2006), fair labour standards and diversity (Lichtenstein et al., 2004; Sen and Bhattacharya,
2001), and promoting health causes (Du et al., 2010; Robinson et al., 2012; Simmons and
Becker-Olsen, 2006). All of these traditional CSR campaigns were focused on activities that
were core to the business. Given that corporations pollute, produce waste, hire cheap labor,
and/or develop unhealthy/health-risky products, various stakeholders including consumers,
20
policymakers,andboardmembersdemandedtransparencyandresponsibilityfrombusinesses
in those areas.
However,overthepastdecade,firmshavegonebeyondsocialandenvironmentalcommit-
ments that are core to their business and have become actively engaged in issues unrelated
to every-day business operations. The social corporate agenda now includes issues such as
LGBTQI+, racism, and gender equality. Put differently, there has been a push towards
broadening the corporate sector’s responsibility and accountability not only towards its im-
mediate direct and indirect stakeholders but also towards overall society, future generations,
and even the global community (Bhattacharya and Korschun, 2008; Erdem et al., 2018;
Hoeffler et al., 2010; Laczniak and Murphy, 2012; Raghubir et al., 2010; Smith et al., 2010).
Givenconsiderablestakeholderpressure, CSRandcorporateactivismisessentialtocom-
panies’ performance and success. Corporate social and environmental initiatives are con-
nected to customer satisfaction and response, competitive positioning, and stock market
performance (Erdem et al., 2018). Scholars have found that CSR activities influence cus-
tomers’ purchase intentions (Sen and Bhattacharya, 2001; Robinson et al., 2012), donation
behavior(Lichtensteinetal.,2004),attitudetowardthefirm(Wagneretal.,2009),emotional
reaction (Joireman et al., 2015), and willingness to pay for the product (Koschate-Fischer
et al., 2012). CSR initiatives also influence brand positioning (Simmons and Becker-Olsen,
2006), brand equity (Torres et al., 2012), competitive position of the firm (Du et al., 2010),
and financial market performance (Du et al., 2017; Luo and Bhattacharya, 2006; Orlitzky
et al., 2003; Servaes and Tamayo, 2013).
WhileextantscholarshiphasexploredvariousfacetsofCSRandcorporateactivism,there
has not been an investigation into CSR and corporate activism in the refugee protection and
empowerment space. Additionally, we know little about whether certain CSR initiatives are
perceived better by consumers than others. This study aims to fill this gap by introducing
a theory of CSR and corporate activism in refugee protection and empowerment.
21
2.4 Theory
This section introduces the theoretical expectations for company and brand involvement in
CSRandcorporateactivismtosupportandempowerrefugees. Italsosetsforthexpectations
of varying consumer preferences for different CSR activities. As pointed out above, there is
no literature on corporate activism or CSR initiatives in refugee protection and support. I
therefore take an interdisciplinary approach and draw upon IR and business scholarship to
derive the following hypotheses.
Overall, Iexpecttheretobedifferencesbetweenindividuals’supportforthefightagainst
climate change, support for refugees, and support for the BLM movement. Comparing
support levels for different social and environmental causes illuminates how companies may
make decisions about where to get involved.
Focusingontherefugeecause,Icontendthatcertaindemographicfeaturesfigurecentrally
for consumer support for CSR and corporate activism. Precisely, the most important factors
should be gender, age, and political party identification. I also argue that personal social
responsibilityandawarenessaswellasskepticismaboutcompanies’motivesmatter. Finally,
Iexpectthatpeopleprefercertaintypesofsupportmorethanothers. Thefollowingsections
will discuss the relevant hypotheses in detail.
2.4.1 Refugee Protection and Empowerment CSR Compared to
Other CSR Activities
CSR and corporate activism has become an integral part of business activities. However,
there are a raft of social and environmental causes to support. Compared to other causes,
why would businesses choose refugees and asylum seekers? We know little about whether
and to what extent consumers support certain CSR activities more than others. I argue that
consumersactuallypreferandsupportcertainCSRactivitiesmorethanothers. Iexpectthat
22
US consumers prefer CSR activities to support the BLM movement and the fight against
climate change more than CSR initiatives for refugees.
Especially for US consumers, refugees remain a rather foreign and distant topic. Given
that most refugee situations are located in Africa and the Middle East, far removed from
North America, the American public has limited experience and exposure to this crisis.
WhiletherehasbeenmoreattentiondirectedtowardsCentralandSouthAmericanmigration
towardstheUS,peoplearrivingattheUSborderaremainlyreferredtoasillegalimmigrants
rather than asylum seekers. As a result of this limited exposure and experience with refugee
flows and refugees, I anticipate that:
Hypothesis 1: On average, CSR in climate change and BLM should receive more support
than CSR and corporate activism in refugee protection and empowerment.
However, this does not mean that CSR activities and corporate activism for the refugee
causeisnotbeneficialorworthwhile. EspeciallywiththeescalationofthewarintheUkraine,
which has triggered a new refugee situation in Europe, attention towards the global refugee
crisishasbeenrevived. IsuspectthatUSconsumersexpectcompaniesandbrandstostepup
andsupportandempowerrefugees. Thenextsectionswillsetupthetheoreticalexpectations
for factors that influence support for refugee protection and empowerment CSR.
2.4.2 Support for Corporate Commitment Towards Refugees
2.4.2.1 Demographic Factors: Age, Gender, and Political Identification
IarguethattheUSpublicsupportsandapprovesofcompaniesandbrandsbecomingengaged
in refugee protection and empowerment. However, certain demographic characteristics in-
crease support for such corporate commitment and activism. Consulting relevant literature,
the most important demographic variables for consumer attitudes and preferences are age
(Diamantopoulosetal.,2003;Dickson,2001;Elias,2004),educationallevel(Diamantopoulos
et al., 2003; Kelly et al., 1990; Roberts, 1996), economic status (Erdem et al., 2018; Roberts,
23
1996), and gender (Gonz´ alez-Rodr´ ıguez Ma et al., 2014; Meyers-Levy, 1989; Moosmayer and
Fuljahn,2010;Sharmaetal.,2012;Skoeetal.,2002;WangandJuslin,2012). Ianticipatethe
most important factors to be age and gender. From this I derive the following hypotheses:
Hypothesis 2a: On average, younger people support CSR and corporate activism in refugee
protection and empowerment more than older people (as age increases, support for CSR and
corporate activism in refugee protection and empowerment decreases).
Hypothesis 2b: On average, women should support CSR and corporate activism in refugee
protection and empowerment more than men.
What is less explored in the business, marketing, or CSR literature is the link between
political ideology and consumers’ attitudes and preferences towards CSR activities. How-
ever, the NYU study also showed that political ideology of consumers matters increasingly
given that many CSR activities involve highly politicised social and environmental issues
(Erdem et al., 2018). Drawing upon American politics scholarship, research has shown how
Americans’ political ideology is associated with supporting social or environmental causes
(Funk and Hefferon, 2019; Grose, 2005; Miller and Stokes, 1963). As immigration and asy-
lum matters are sensitive and highly politicized topics in the US, it is sensible to assume
that political ideology is a key factor for peoples’ attitudes and preferences towards CSR
activities. I thus expect:
Hypothesis 2c: On average, liberal participants (Democrats) should support CSR and cor-
porate activism in refugee protection and empowerment more than conservative participants
(Republicans).
2.4.2.2 Personal Social Responsibility & Skepticism About Motives as Factors
Besides demographic factors, literature has also hypothesized about another important fac-
tor: consumer social responsibility. To be precise, consumers who themselves are socially
more responsible and aware should be more inclined to support companies and brands’ CSR
24
activities (Caruana and Chatzidakis, 2014; Caruana, 2010; Devinney et al., 2006). People
who are interested and informed about social and societal issues as well as those who pay
attention to companies’ social and societal engagement should have a positive reaction to
companies supporting and empowering refugees. From this, I derive:
Hypothesis 3a: On average, people who are socially responsible and aware should support
CSR and corporate activism in refugee protection and empowerment more than people who
are not.
Finally, people and thus consumers have been rather torn about the motives and inten-
tions behind companies getting involved in social and environmental issues. However, how
does such consumer skepticism influence their attitudes towards CSR initiatives? Business
literaturehasexploredthelinkbetweenmotivesforcompanies’CSRinitiativesandconsumer
reactions (Chernev and Blair, 2015; Habel et al., 2016). For example, scholars found that
when consumers feel companies’ engagement is not altruistic and/or genuine, it negatively
impactsattitudesandbuyingintentions(Becker-Olsenetal.,2006;BrownandWoods,2007;
Ellen et al., 2017).
However, other scholars have challenged these findings and argued that consumer skepti-
cism about companies intentions to get involved is not negatively associated with consumer
attitudes and buying preferences. For example, Forehand and Grier (2003) show that con-
sumers do not even consider companies’ and brands’ motives. Similarly, Zasuwa (2018)
confirm that consumer responses are often independent of motives for CSR activities. I join
the latter camp and argue that while consumers’ may be skeptical of companies’ intentions
and motives, it does not decrease their support for CSR activities. Thus, I hypothesize:
Hypothesis 3b: On average, consumer skepticism about companies intentions to support
and empower refugees does not have a negative impact on consumer support for CSR and
corporate activism in refugee protection and empowerment.
25
2.4.3 Support For Different Types of Commitments
I also expect that consumers prefer certain refugee protection and empowerment CSR ac-
tivities over others. For example, it should matter whether a company merely advocates for
refugees or whether a company actively trains and hires refugees. There is no peer-reviewed
research on which CSR activities for refugees should receive more support. However, the
NYU study explored something similar and found that consumers tend to support charity
and service delivery as well as hiring refugees (Erdem et al., 2018).
Drawing upon these results, I anticipate that consumers find monetary donations and
hiring refugees as especially appealing for two reasons. First, consumers may expect compa-
nies to give back and devote some of their large profits to refugee causes. Consumers may
see these monetary donations also as an indirect way of giving back themselves. In other
words, their purchase indirectly supports the refugee cause. Second, consumers may also
desire a more direct way of companies supporting refugees. Given that refugees will need
to be integrated in host societies, consumers may find companies and brands responsible for
training and hiring refugees and thus speed up the integration process. From this, I derive:
Hypothesis 4: On average, hiring refugees and donating money is likely to receive more
support than other CSR activities.
Table 2.1 briefly summarizes the discussed hypotheses:
26
Hypotheses
H1: On average, CSR in climate change and BLM should receive more support than CSR
and corporate activism in refugee protection and empowerment.
H2a: On average, younger people support CSR and corporate activism in refugee
protection and empowerment more than older people (as age increases, support
for CSR and corporate activism in refugee protection and empowerment decreases).
H2b: On average, women should support CSR and corporate activism in refugee
protection and empowerment more than men.
H2c: On average, liberal participants (Democrats) should support CSR and
corporate activism in refugee protection and empowerment more than conservative
participants (Republicans).
H3a: On average, people who are socially responsible and aware should support CSR
and corporate activism in refugee protection and empowerment more than people who are not.
H3b: On average, consumer skepticism about companies intentions to support and empower
refugees does not have a negative impact on consumer support for CSR and corporate
activism in refugee protection and empowerment.
H4: On average, hiring refugees and donating money is likely to receive more support
than other CSR activities.
Table 2.1: Hypotheses
To visually capture the hypotheses, I present my theory of refugee empowerment and
protection CSR in figure 2.1. The upper panel reflects H1, where I theorize about which
type of CSR activities should receive more customer support on average. The ‘greater than’
sign reflects my expectation that, on average climate change CSR and BLM CSR, receive
more support than refugee empowerment and protection CSR.
The lower panel then illuminates the proposed analysis of individual level factors that
influencepeoples’supportforrefugeeempowermentandprotectionCSRandactivism. They
areallindividuallevelfactors,however,theycanbefurtherdividedintodemographicfactors
and consume-related factors. The outcome variable is support for refugee CSR & activism
(H1-3) as well as type of support (H4).
27
Age
Consumer personal
social responsibility &
awareness
Consumer skepticism
Political identification
Gender
Support for refugee CSR &
activism
Demographic Factors
Consume-related Factors
Individual-Level Factors
Type of support
Support for climate
change CSR
Support for BLM CSR
Support refugee
empowerment &
protection CSR
On average, support for BLM CSR and climate change CSR should be greater than support for refugee CSR
Analysis of factors
that influence refugee
CSR support
Figure 2.1: Theory of Refugee Empowerment & Protection CSR
Having introduced the theoretical expectations, I now turn to the research design. The
next section discusses the survey design and analysis.
28
2.5 Research Design
I conducted a survey experiment to test the above presented hypotheses. A survey exper-
iment is well-suited as it allows me to capture peoples’ attitudes and preferences towards
refugee protection and empowerment CSR activities while comparing them to other social
and environmental CSR initiatives such as the BLM movement and the fight against climate
change. A plethora of scholars have utilized survey and experimental data to explore similar
questions (Latif and Sajjad, 2018; Tian et al., 2020; Yamane and Kaneko, 2021). The fol-
lowingsectionswilldetailparticipantrecruitment, sampleselection, andsurveyconstruction
and analysis.
2.5.1 Survey Design and Participant Recruitment
I constructed an approximately 7-minute survey experiment in Qualtrics. The survey was
administeredthroughProlificanddeployedtoapproximately1,500Americanadults;roughly
500 participants per treatment group.
4
The study population is a representative sample of
the US population (see Appendix tables for summary statistics 5.1, 5.2, 5.3). The sample
size was mainly determined by budget, however a power analysis was conducted to ensure
enough participants per treatment group.
5
ThefocusonAmericanconsumersandcustomershastwomainreasons. First,theUnited
States represents one of the largest, and thus most important, consumer markets worldwide.
ItisthereforesensiblethatcompaniescareaboutUSconsumeropinionsandattitudes,which
in turn satisfies the core assumption that US consumers indeed exert stakeholder pressure
over corporations and could influence CSR activities.
Second, similar to other countries, attitudes towards refugees remain highly polarized in
the US. A recent poll found that while Americans overall believe that refugees are able to
4
A total of 1,574 people completed the survey. This was to remedy the impact of missing data and to
ensure a minimum of 500 people per treatment group
5
Power (1- β ) = 0.80, and Type I error rate (α ) = 5 %.
29
integrateandcontributetotheUS(around60percent),thereremainsalotofskepticism. For
example,50percentofUSpollparticipantsagreedthatrefugeescometotheUSforeconomic
reasons and to exploit welfare services; 53 percent think that allowing refugees to work while
they await their asylum decision could encourage people without a genuine asylum claim
to come to the US (Boyon, 2022). This polarization makes the US an interesting case for
exploring attitudes towards CSR given that companies often offer training, skill, and hiring
opportunities for forcibly displaced people.
Before launching the official survey mid April 2022, I deployed a pilot survey with 90
respondentstotroubleshootthesurvey. Thisaidssurveyvaliditybyexposingandmitigating
technicalproblemsandupdatingandrephrasingpotentiallyconfusingquestions(participants
were able to give feedback on the survey). The hypotheses and analysis for the study were
pre-registered using the Open Science Framework in March 2022.
The survey consisted of multiple choice, multi-answer, and ranking questions. After be-
ing asked a set of demographic questions and general questions on product and purchase
preferences and behavior, participants were randomly assigned to one of the three treat-
ments: 1) companies supporting refugees, 2) companies supporting the BLM movement, or
3) companies supporting the fight against the climate crisis.
Each treatment group was first presented with a brief statement on either the global
refugee crisis, the BLM movement, or climate change (issue statement, definition, numbers,
and facts). Participants were then asked if they were aware/had heard of the refugee crisis,
theBLMmovement,orclimatechangerespectively. Thiswasfollowedwithastatementindi-
catingthatcompaniesandbrandssupportandempowereitherrefugees,theBLMmovement,
or the fight against climate change. Participants were then asked whether they knew that
companies and brands were supporting refugees, the BLM movement, or the fight against
climatechange. ThetreatmenttextpertreatmentgroupisavailableintheAppendix(figures
5.1, 5.2, 5.3). The rest of the survey proceeded to query participants about attitudes and
preferences on CSR activities and corporate activism specific to their assigned treatment
30
group. These questions were held constant across the different treatments with the only
difference being the respective social or environmental cause.
Treatment selection was based on providing 1) realistic and real-world CSR activities
and 2) topical events relevant to the US public. First, research has explored the trade-off
between abstract or detailed/real life experimental vignettes (Brutger et al., 2021; Gaines
et al., 2007). I chose vignettes that offer enough detail to introduce the respective issue yet
aregeneralenoughtoavoidsacrificingexternalvalidity. Inaddition,researchhasshownthat
realisticscenariosproducemorereliableresults(AguinisandBradley,2014;Atzm¨ ulleretal.,
2017). Consequently,participantsreceivedinformationonrealCSRinitiativespromotedand
carried out by real companies (which most, if not all, participants should be familiar with).
Second and related, the scenarios reflected social and environmental issues that are rel-
evant and relatable to the US public. Refugee protection and empowerment is the topic of
interest in this study and represents a socio-political issue. For the comparison treatments,
I introduced the BLM movement as another socio-political issue as well as climate change as
an environmental issue. Climate change has been a long-standing environmental issue that
has figured centrally in the US policy arena. Similarly, the BLM movement is an important
and highly politicized social issue. Presenting participants with both social and environmen-
tal CSR activities allowed me to compare peoples’ attitudes and preferences across some of
the most prevalent issue areas. With this, I am able to compare and contrast reactions to
refugee protection CSR with another prominent socio-political issue (BLM) as well as with
an issue area that is environmental rather than social.
31
2.5.2 Measurement of Concepts and Analyses
2.5.2.1 Variables and Measurements
To ensure internal validity and content validity of my survey, I consulted relevant literature
on survey instruments and measurements.
6
I hypothesize that certain demographic variables (gender, age, and political ideology)
influence attitudes towards refugee protection and empowerment CSR. Survey participants
were asked a number of questions pertaining to their demographic characteristics (age, gen-
der, income, region, politicalideology), fromwhichIconstructedtherespectivedemographic
variables of interest.
I also put forward that peoples’ social responsibility and awareness matters and increases
their support for refugee CSR activities. I capture this concept in two ways. First, I ask
participants to list their priorities/interests when they read/watch the news. This reveals
whether and to what extent participants are informed and invested in social and societal
issues. Second, Iaskparticipantstoranktheimportanceofvariousfactorswhentheydecide
to make a purchase (price, sustainability of product, brand name, brand social engagement,
etc.). This allows me to measure how important people find brand social engagement as
well as whether and to what extent they pay attention to companies’ social and societal
engagements.
Moreover, I assume there to be a relationship between skepticism about companies’ mo-
tives/intentions and support for refugee protection and empowerment CSR activities. I ask
surveyparticipantsseveralquestionsaboutmotivesofcompaniesforengaginginrefugeesup-
port as well as whether and to what extent they are skeptical about companies’ intentions.
This serves as the measurement for skepticism about the companies motives/intentions.
Finally, I expect that people prefer certain types of activities more than others. In par-
ticular, hiring refugees and monetary donations should receive more support than other
6
Latif KF, Sajjad A. Measuring corporate social responsibility: A critical review of survey instruments
offers a comprehensive overview and framework of empirical CSR studies and consumer preferences.
32
activities. I included a survey question that specifically asked participants to choose their
two preferred ways of how companies should support and empower refugees. I also ask a
follow-up question that measures whether a company committing to different CSR activi-
ties (hiring, donations, advocacy, etc.) makes participants more or less likely to purchase
products/services from that company.
The main dependent variable of interest is support for CSR and corporate activism in
refugeeprotectionandempowerment. Tocreateaholisticmeasureforthisconcept,Icapture
support for refugee CSR along three dimensions: 1) emotional reaction of participants (hap-
piness), 2) trust in companies, and 3) purchase intentions. First, I ask participants how they
feel about companies’ and brands’ supporting and empowering refugees through CSR. This
reveals an emotional reaction and thus represents an implicit facet of support for refugee
CSR activities. Second, I inquire to what extent knowing that a company supports and
empowers refugees increases the participants’ trust in the company. Finally, participants are
asked whether and to what extent knowing that a company supports and empowers refugees
makes them want to purchase from the respective company. Both trust and purchase in-
tentions reflect a direct expression of support for CSR activities: consumers are rewarding
companies for their social engagement by placing trust in them and buying from them. To-
gether, happiness, trust, and purchase intention form a compound measure for consumer
support for refugee CSR and corporate activism.
Table 2.2 provides an overview of the main variables of interests as well as their mea-
surements.
2.5.2.2 Data Cleaning and Analysis
The survey data was cleaned and analyzed in R. Before conducting any statistical tests, I
eliminated flatliners (participants who selected the same answer choice for every (or most)
questionsinthesurvey)andexcludedparticipantswhocompletedthesurveyinlessthan1.5
33
Variable Name Measurement Values
Independent Variables
Age
Gender
Political Party
Identification
Age Intervals
Categorical
Categorical
18-24, 25-34, (...), 75-84
Female, Male, Prefer not
to say/other,
Independent, Republican,
Democrat
Personal Social
Responsibility
& Awareness
Skepticism about
companies’ motives for
CSR engagement
Rank/interval
turned into binary
Binary and
interval
High or Low;
interval: companies’
social engagement as
not important at
all to very important
High or Low;
interval: not skeptical
at all to to very skeptical
Type of support for
refugees
Categorical
Hire or train refugees,
financial donations,
material donations,
advocate for refugees,
companies should not support
refugees at all
Dependent Variable
Support for CSR and
corporate activism in
refugee protection and
empowerment
Interval data turned
into additive
score
0-12 (no support at all
to highest support)
Table 2.2: Variables and Measurements
minutes.
7
Ialsoonlyconsideredsubmissionsthatweremostlycompleteandhadlittlemissing
data (completion rate of 89 percent or more). This avoids critical data gaps and ensures
consistent and sufficient data coverage for my main variables of interests. After cleaning
the data, the sample size decreased to 1,555 participants, with about 517-519 people per
treatment group.
A difference in means test reveals whether there is a statistically significant difference
between the three treatment groups. To explore correlations between the main independent
variables of interests (demographic factors, personal social awareness and responsibility, and
skepticism about motives) and the main dependent variable (support for CSR and corporate
activism in refugee protection and empowerment), I run multivariate regressions (ordered
logit and OLS). I complement this data through coefficient plots with CIs to illuminate
7
This time is based on average and median timing of completion from the pilot study.
34
differences across demographic groups and preferences towards different refugee protection
CSR activities.
2.6 Empirical Results
The following paragraphs present the results for the earlier introduced hypotheses. The first
section focuses on the comparison across different CSR activities and whether the US public
prefers certain CSR activities (support for refugees/the BLM movement/the fight against
climate change) more than others. The second section hones in on refugee CSR activities
and illuminates which factors increase support for refugee CSR activities. The last section
closes by presenting which types of support Americans prefer.
2.6.1 Comparison Across Different CSR Activities
Before presenting the results of the comparison, I show that most participants are actually
aware of the respective issue. This is important as it ensures that participants know about
the social or environmental issue at the heart of this study, and that they are able to form
(or have already formed) opinions, perceptions, and attitudes on companies engaging in the
respective cause. As figure 2.2 illuminates, awareness of the cause is high across all different
treatments. 86.6 percent of participants in the refugee treatment group indicate that they
know about the refugee crisis. Awareness is even higher for the climate change group (98.8
percent) and the BLM movement group (99.6 percent).
Having established that most participants are aware of the issues, we turn to the central
question: do Americans support certain company engagements in social or environmental
issues more than others? The difference in means tests reveals there is indeed a statistically
significant difference between support for CSR activities to support refugees, CSR activities
to support the BLM movement, and CSR activities to support the fight against climate
35
Figure2.2: Surveyparticipantswereaskedwhethertheyhaveheardoftherefugeecrisis(n =
517), the BLM movement (n = 519), or the climate change crisis (n = 519). Answer options
included Yes/No/Not sure. For the BLM question, none of the participants answered Not
Sure.
change. Turning to figure 2.3, we see the mean values of support with Confidence Intervals
(CIs) for each treatment group.
As hypothesized, climate change receives the most support with an average of 8.14.
However,unlikeexpected,refugeeprotectionreceivesthesecondmostsupport(7.89)followed
by the BLM movement (6.21). This may be a result of the BLM movement triggering very
strong emotions and opinions as this has been an ongoing, highly politicized issue.
The diff-in-means test confirms there to be statistically significant differences between
the BLM group and the refugee group as well as the BLM group and the climate group (see
Appendix tables 5.5 and 5.6). Given the similarity in support values for climate and refugee
36
Figure 2.3: Plot presents mean level of support with Confidence Intervals. CSR to support
refugees(n = 517), CSR to support the BLM movement (n = 519), CSR to support the fight
against climate change (n = 519).
groups, the diff-in-means test indicates that there is no statistical significant difference be-
tween these group means (see table 5.4 in the Appendix).
2.6.2 Factors That Influence Support for Refugee CSR activities
ThecomparisonbetweensupportforthedifferentCSRactivitieshasrevealedthatAmericans
indeed support certain corporate engagement more than others. In particular, the analysis
hasilluminatedthattheUSpublicstronglysupportscompanies’commitmenttoprotectand
empowerrefugees. First,CSRactivitiestosupportandempowerrefugeesreceivesignificantly
more support than the CSR activities for the BLM movement. Second, attitudes towards
37
CSR activities for refugees and CSR activities for the fight against the climate change are
so similar that there is no statistically significant difference between the levels of support.
This section turns to the factors that influence support for refugee CSR. The following
figuresillustratetheaveragesupportforrefugeeCSRbasedon gender, age, political ideology,
personal social awareness and responsibility and skepticism about companies’ motives.
Figure 2.4 presents the mean level of support based on gender and reveals that men have
a lower support level (7.18) compared to women (8.55). A diff-in-means test shows that the
differentsupportlevelsbetweenmenandwomenareindeedstatisticallysignificant(seetable
5.7 in Appendix). This confirms that women, on average, have higher levels of support for
companies CSR activities to protect and empower refugees than men.
Figure 2.4: Plot represents Mean Levels of Support with Confidence Intervals. Female: n =
264, Male: n = 242.
38
Figure 2.5 captures the average support levels based on participants political party iden-
tification. Republicansdisplaythelowestsupportwithanaverageof5.54points, followedby
Independents/others with 6.81, and Democrats with 8.93. This aligns with my theoretical
expectations: on average, liberal participants more readily support CSR activities to protect
and empower refugees than conservative participants. The differences in means between
Republicans, Democrats, and Independents/others are all statistically significant (refer to
tables 5.8, 5.10, 5.9).
Figure 2.5: Plot represents Mean Levels of Support with Confidence Intervals. Democrat: n
= 334, Republican: n = 119, Independent/other: n = 64
39
In figure 2.6, we observe the average levels of support across different age brackets. All
meansarefairlyclosetogetherwiththeCIsoverlapping. Amultiplecomparisontestconfirms
that differences across different age groups are not statistically significant (see figure 5.4 in
Appendix). Theplotalsorevealsthatnotonlyyoungpeopleexpresshighsupportforrefugee
CSR activities. Support for refugee CSR activities is particularly high in age groups 65-74
and 75-84, which indicates that the relationship between age and support for refugee CSR
activities may be non-linear.
Figure 2.6: Plot represents Mean Levels of Support with Confidence Intervals. 18-24: n =
63, 25-34: n = 118, 35-44: n = 92, 45-54: n= 74, 55-64: n = 102, 65-74: n= 56, 75-84: n=
12
40
When it comes to peoples’ personal social responsibility and awareness, figure 2.7 shows
that those who prioritize social/societal issues in their news consumption and those who pay
attention to companies’ social/societal engagement, on average, express more support for
companies’ refugee CSR activities. For example, people who are attentive to social issues
in the news have an average support level of 8.48 compared to 7.18 for those who only pay
little to no attention to social issues in the news. The differences in support are even more
striking between low and high attention groups for brands’ social engagement: Americans
who care about companies commitment to social causes have an average support level of
10.09 versus those who care little or not at all about companies’ social causes (7.33). The
differences are statistically significant (see tables 5.11 and 5.12).
Figure 2.7: Plot presents Mean Levels of support with Confidence Intervals. High attention
to social issues in news: (n = 281), Low attention to social issues in news: (n = 236),
High attention to brands’ social engagement (n = 105), Low attention to brands’ social
engagement: n = 412
41
I also hypothesized about the relationship between peoples’ skepticism about companies’
motives/intentionstogetengagedinrefugeeprotectionandtheirsupportforCSRactivities.
While I expected that increased skepticism does not negatively influence levels of support,
we observe the opposite impact. Precisely, people who are highly skeptical about companies’
motives have a much lower level of support (6.81) in relation to those with low skepticism
(9.02). The differences are statistically significant (see figure 2.8 and diff-in-means in Ap-
pendix table 5.13).
Figure2.8: PlotpresentsMeanLevelsofsupportwithConfidenceIntervals. Highskepticism
about companies’ motives: (n = 263), Low skepticism about companies’ motives: (n = 251)
42
2.6.3 Regression Results
Figure 2.9 represents a coefficient plot with results from multivariate regression models (Or-
dered Logit and OLS with cluster-robust standard errors (CRSE)). The statistical analyses
explore the relationship between the main independent variables of interest (gender, age,
political ideology, personal social responsibility and awareness, and skepticism about com-
panies’motives)andthemaindependentvariableofsupportforCSRandcorporateactivism
to support and empower refugees. To ease interpretation of the regression results of the or-
dered logit model, Table 2.3 presents the odd ratios.
Figure2.9: Estimateswhichdonotcrosstheverticallinearep<0.10(90%). Allindependent
variables were checked for their level of multicollinearity; see figure 5.5 in Appendix for a
correlation matrix.
As expected, the probability of men expressing high levels of support for CSR activities
for refugees is about 27 percent less than for women (significant at the 0.10 level). Even
43
more striking and in the expected direction are the results for political ideology: those who
identify as Independent/other or Republican are 68.6 percent and 79.6 percent less likely to
have high levels of support for companies’ refugee protection and empowerment activities.
The results for the age variable “Under 35” (coded as a binary variable here for simplicity)
are barely significant (0.10 level). Additionally, the mean barplot earlier that detailed the
various age brackets (Figure 2.6) suggests very similar levels of support across age groups
with older participants (ages 65-74 and 75-84) displaying high levels of support. Therefore,
I do not find sufficient support for the hypothesis that younger people on average support
CSR activities for refugees more than older people.
Turning to peoples’ personal social responsibility and awareness, the regression results
suggest the hypothesized, positive relationship. In particular, people who pay a lot of atten-
tion to brands’ social engagement are 2.3 times more likely to support CSR activities and
corporate activism for refugees (significant at the 0.000 level). Similar, people who prioritize
social issues in their news consumption are 1.5 times more likely to express high support for
refugee CSR than those who do not (significant at the 0.05 level).
Odds ratio 5 % 95 %
Male 0.729 0.547 0.973
Independent/other 0.314 0.202 0.487
Republican 0.204 0.139 0.300
Under 35 1.415 1.052 1.902
Bachelors or higher 1.115 0.832 1.495
White 1.474 1.085 2.003
Attention to brands’ social engagement 2.318 2.009 2.674
Attention to social issues in news 1.461 1.086 1.964
Skepticism about companies’ motives 0.535 0.474 0.604
Income Less than $20,000 0.929 0.559 1.544
Income $20,000-39,999 0.760 0.491 1.175
Income $40,000-59,000 0.879 0.582 1.327
Income $60,000-99,999 0.810 0.553 1.187
Know about refugee crisis 1.730 1.102 2.714
Table 2.3: Odds Ratios for Ordered Logit Regresssion Results
44
Lastly,whileIexpectedtoseethatincreasedskepticismaboutcompanies’motives/intentions
wouldnothaveanegativeimpactontheirsupportforrefugeeCSRactivities,Ifindtheexact
opposite. Peoples’ increased skepticism about corporate motives decreases the likelihood of
high support for CSR activities for refugee protection/empowerment by about 46 percent
(significant at the 0.000 level).
2.6.4 Preferences on Type of Support for Refugee CSR
Finally, I present the results for what type of corporate support Americans’ prefer. As
support for CSR activities and corporate activism is rather broad, I give people the choice
to choose across different support activities. Through this, I dehomogenize support and gain
insights into what type of support US participants would like to see from companies.
I hypothesized that people should prefer active engagement such as hiring and training
refugees as well as making financial donations to refugee causes over other activities. Figure
2.10 uncovers that, as suggested, Americans expect companies to step up and actively hire
and train refugees.
The second most supported CSR activity is material donations (food/clothes). The
least supported activities include financial support and activism. This overall confirms that
Americans prefer companies to become actively engaged in shaping refugee protection and
empowerment rather than remain passive money givers or symbolic supporters.
45
Figure 2.10: Multi-select question: Participants were asked to pick their top two choices
of how they believed companies should and could best support refugees. Overall n = 517;
they should not support/empower refugees in any way: n = 34 ; advocate for refugees and
refugee rights: n = 121; donate money to refugee causes: n = 230; donate goods such as
food/clothes to refugees: n = 277 ; hire or train refugees: n = 324. Percentage calculated:
n choosing the answer/overall n*100.
2.7 Empirical Discussion
ThisstudyhasanalyzedAmericanattitudesandpreferencestowardscompanies’engagement
invarioussocialandenvironmentalissueswithaparticularfocusontherefugeecrisis. Table
2.4 offers an overview of the findings.
This study is one of the first to compare the level of support for CSR activities and
corporate activism across different social and environmental causes. It is motivated by
the question of why companies would choose to support and empower refugees rather than
pick any other (potentially less sensitive/costly) cause. I show that not only is consumer
pressure present and could serve as motivating factor for companies to get engaged in the
46
Hypotheses Status Notes
H1: On average, CSR in climate change and BLM should
receive more support than CSR and corporate activism
in refugee protection and empowerment.
Partly confirmed
climate change receives
the most support,
however the refugee cause
receives second most support.
H2a: On average, younger people support CSR and
corporate activism in refugee protection and
empowerment more than older people (as age increases,
support for CSR and corporate activism in refugee
protection and empowerment decreases).
H2b: On average, women should support CSR and
corporate activism in refugee protection and
empowerment more than men.
H2c: On average, liberal participants (Democrats) should
support CSR and corporate activism in refugee
protection and empowerment more than conservative
participants (Republicans).
Not confirmed
Confirmed
Confirmed
H3a: On average, people who are socially responsible and
aware should support CSR and corporate activism in
refugee protection and empowerment more than people who
are not.
H3b: On average, consumer skepticism about companies
intentions to support and empower refugees does not
have a negative impact on consumer support for CSR and
corporate activism in refugee protection and empowerment.
Confirmed
Not Confirmed
H4: On average, hiring refugees and donating money is likely
to receive more support than other CSR activities.
Partly confirmed
Hiring refugees is the top
preferred type of support.
However, donating money only
comes in 3rd place.
Table 2.4: Summary of Findings
refugee cause, but also that there are significant differences in support for different CSR
activities. Individuals’ support for companies’ refugee protection and empowerment CSR
was unexpectedly high. Undoubtedly, the events in the Ukraine may have influenced the
levels of support. However, the NYU study conducted a few years ago, found comparable
attitudes and support for refugee CSR (Erdem et al., 2018). Thus, there seems to be a
consistent pattern and pressure for increased for-profit engagement in the refugee protection
space. Two main take-aways can be derived from this.
First, regardless of there being an ongoing crisis, investing in refugee protection and
empowerment CSR initiatives may be beneficial for companies because American consumers
have expressed consistent support for the cause and brand engagement. In short, helping
47
refugeeshelpscompanies. Highsupportthatisexpressedthroughincreasedhappiness,trust,
and purchase intentions creates pressure for companies to get involved in refugee causes.
Second, given the public’s support and the increasingly active role of the for-profit sector
in the refugee protection and empowerment space, it will be important for practitioners
and scholars to pay attention to these developments. In the light of this evolving reality of
the for-profit sector as an increasingly influential and powerful player in the humanitarian
space, governments, organizations, and scholars should continue to explore opportunities for
a multi-stakeholder approach.
This study is also the first comprehensive study that statistically tests which factors
influence levels of support for refugee protection and empowerment CSR. It is interesting,
yet maybe not as surprising, to see that women and politically liberal participants are most
supportive. However, knowing which demographic factors are most influential in shaping
support is essential for companies’ messaging and marketing strategies. As people are ex-
posed to companies’ advertising and marketing efforts on a daily basis, CSR campaigns can
mobilize and inform people about the cause (arguably even more than information efforts
by governments or organizations). This in turn can lead to more politically/socially aware
and engaged constituents, which is essential for policymakers and practitioners.
Besides demographic factors, the study verified that the more socially responsible and
aware people are, the more they support companies’ CSR efforts. This supports the afore-
mentioned point of the benefits arising from creating socially and politically more informed
consumers: companies, governments, and even refugees may profit. However, when people
are skeptical about companies’ intentions/motives to get involved in a social cause such as
refugee protection and empowerment, the level of support decreases. While contrary to my
hypothesis, this is good news for humanity (and maybe bad news for capitalism). The find-
ing suggests that although people support companies stepping up, they are cautious about
the why. This holds the for-profit sector accountable and creates demand and need for
transparency in CSR activities and processes. Put differently, companies cannot simply do
48
whatevertheywantbecauseconsumersdonotcare. Rather,companiesneedtoestablishand
provetopeoplethatcompanyengagementisauthenticandmeaningful, whichthenincreases
the chances of high-quality CSR initiatives.
Both the high support for refugee CSR and the desire for meaningful initiatives are
reflected in the type of support people would like to see from the for-profit sector. As
already mentioned above, the times of pure passive or symbolic support in form of financial
donations and public statements is over. People prefer companies to actively train and hire
refugees and to support them with material donations. This is interesting as it challenges
the notion of the global refugee crisis as a distant, geographically removed issue. Instead
people acknowledge and express that creating and finding a solution can and should happen
anywhere, withcompanieshiringandtrainingrefugeesrighthereintheUnitedStates. With
this,bothcompaniesandrefugeesbecomeactivecontributorsandparticipantsinthesolution
process.
2.8 Conclusion
Why would companies and brands decide to get involved in refugee protection and empow-
erment? As this study has shown, one potential answer is consumer pressure: Americans
supportandapprovewhencompaniesstepupandaresociallyandenvironmentallyconscious
and engaged. I put this consumer pressure to the test in a survey experiment to answer 1)
howandtowhatextentconsumerswouldliketoseecompaniessupportandempowerrefugees
and2)whetherconsumerstendtosupportCSRactivitiesmore/lesscomparedtoothersocial
and environmental causes.
This study barely scratches the surface of the potential research agenda that lies beneath
thephenomenonoffor-profitsectorengagementinthehumanitarianspace. Iclosethischap-
ter by pointing out three paths for the future. First, and already addressed in the empirical
discussion, is the acceptance and embracement of the for-profit sector as an influential and
49
active participant in the humanitarian field and, in particular, in refugee protection and em-
powerment. This goes for both scholars and practitioners. Thus far, literature on for-profit
involvement in refugee protection and empowerment is sparse. We know very little about
program creation, dynamics, and effectiveness.
Second, and related, there are huge opportunities for governments and organizations to
partner with the for-profit sector to create multi-stakeholder programs and learn from one
another. Not a single actor can remedy the effects of the global refugee crisis; various actors
with their differing expertise and experience need to work together to create sustainable
approaches. This not only benefits the respective actors but the overall cause.
Third, messaging and advertising CSR activities as an alternative mobilizing and infor-
mation tool for the public presents another research gap within public policy, politics, and
internationalrelations. Aspointedoutabove, peopleareconstantlyexposedtotheirfavorite
brands’ advertisement campaigns, which includes messaging about CSR activities. This is
an untapped opportunity for inquiry as there is no research on how and to what extent such
exposure and messaging (and what type of messaging) influences peoples’ awareness and
attitudes towards refugees.
To conclude, this chapter has illuminated how preferences and attitudes of the US pub-
lic connects to companies’ increasingly active role in the humanitarian space. With this,
I hope to have set the stage for a broader research agenda on for-profit engagement in
refugee protection and empowerment. In the following chapters, I attempt to fill some of
the aforementioned critical gaps on the creation of public-private partnerships in the refugee
protection space as well as program dynamics and performance. The next chapter focuses
on the factors that make partnership formation between governments, organizations, and
businesses in this burgeoning public-private network more likely, and who partners up with
whom.
50
Chapter 3
Refugee Protection Inc.: Emergence of a New
Public-Private Protection & Empowerment Network
3.1 Introduction
Thepreviouschaptershowedthatconsumersandcustomerssupportbusinessengagementin
refugeeprotectionandempowerment. However, whenbusinessesdecidetogetinvolved, they
often do not operate alone. Instead, they attempt to form partnerships with other, often
times more experienced, actors such as governments and organizations. Public actors such
asstatesandorganizationsmaywelcomepartnershipswiththeprivatefor-profitsectortodi-
versifyexpertiseandfundingopportunities. Asaresult,morediversepartnershipshavebeen
on the rise as they hold the potential to produce more sustainable and sophisticated solu-
tions. We are currently witnessing the emergence of a groundbreaking public-private refugee
protection & empowerment network to combat the global refugee crisis more effectively.
However, we know little about how this new public-private network comes into being:
what factors influence partnership formation; who partners up with whom? This chapter
sets out to answer these questions and argues that the most important factors for forming
partnerships are 1) actor type, 2) being connected to a central actor such as the UN Refugee
Agency (UNHCR), and 3) geographic location. It also expects actors with similar attributes
to be more likely to form connections with one another rather than with unlike actors.
51
Usingsocialnetworkanalysisandexponentialrandomgraphmodels(ERGMs),thischap-
ter tests a theory of partnership formation on an original dataset that records public-private
actor connections based on shared protection & empowerment programs and finds support
for my theoretical expectations. It claims and finds that for-profit actors such as businesses
have a more difficult time forming connections compared to traditional refugee protection
actors such as international organizations and states. The analysis also shows that being
connected to the UNHCR increases the chances of tie formation. There is also support for
geographic homophily: Northern actors more readily form ties with other Northern actors,
and Southern actors indeed link up more with Southern actors. When it comes to partner-
ships with similar actors, the results mostly conform with the hypothesis that actors are
more likely to connect with the same type of actor (for example organizations with other
organizations) rather than with different actor types.
This chapter first reviews the relevant literature before introducing a theory of network
emergence and tie formation. It then presents the research design and empirical results. Fi-
nally,thechapterengagesinadiscussionofthefindingsbeforeconcludingandforeshadowing
the next dissertation chapter.
3.2 Literature
The following section will survey the extant, relevant literature on more diverse forms of
cooperation in world politics, the rise of non-state actors as well as the role of the for-profit
sector in refugee protection.
3.2.1 New Forms of Cooperation in Global Governance
As already addressed in the previous chapter, Global Governance (GG) literature has paid
close attention to the changing patterns of actor participation, rule/norm setting, and policy
making in world politics (Acharya, 2016; Hasenclever et al., 1997; Patrick, 2014). To remind
52
the reader, GG is defined as “the collective effort by sovereign states, international orga-
nizations, and nonstate actors to address common challenges and seize opportunities that
transcend national frontiers” (Patrick, 2014, pg.59).
The increased participation of non-state actors has spurred more, diverse forms of coop-
eration in the international system. In particular, state and non-state actors are working
together to address global issues and to provide goods collectively. This trend reflects the
redistribution of authority in GG and the rising presence of for-profit entities such as cor-
porations in world politics (Cutler et al., 1999; Hall and Biersteker, 2002; Sch¨ aferhoff et al.,
2009). Non-stateactorsdonotonlyindirectlyinfluencedecision-makingthroughlobbyingor
naming/shaming, they are actively involved in the political process and co-govern alongside
state actors (Sch¨ aferhoff et al., 2009; Rosenau and Linder, 2000; Osborne, 2005).
Both scholars and practitioners have emphasized the potential of the for-profit sector in
global governance broadly, but also more specifically in the area of state and peace building.
A growing literature engages with the role of for-profit actors in conflict zones and post-
conflict state building (Backer, 2011; Barbara, 2006; Gruener et al., 2021). For example,
the private sector is essential for catalyzing private investment and thus fostering economic
growth in post-conflict states. However, the involvement of the for-profit private sector in
refugee protection and empowerment is a rather new development.
3.2.2 The Refugee Regime and the For-Profit Private Sector
Institutional proliferation on the global stage has also resulted in more diverse set of actors
participating in the global refugee regime
1
. For a long time, states and IGOs had been
considered the most important actors given these public entities are supposed to provide
protectionandasylum. However, scholarshiphasrefocuseditsattentiononnon-stateactors,
1
The refugee regime refers to institutional arrangements that ensure migrant and refugee rights and
lay out state responsibilities. The main treaty is the Refugee Convention with the UNHCR as the main
organization.
53
especially on the importance of NGOs for stepping up and providing services and protection
(often in areas where states and IGOs fail) (Cammett et al., 2014; Campbell et al., 2019).
Nevertheless, the role of the private for-profit sector in the refugee regime has received
limited attention. This is surprising given “the vast literature that now exists relating to the
roleoftheprivatesectorinotherareasofglobalgovernance”(Bettsetal.,2017). Comparing
the amount of scholarship conducted on business involvement in areas such as trade, health,
and environment (Hall and Biersteker, 2002; Brown and Woods, 2007; Cutler et al., 1999;
Fuchs, 2007; Ruggie, 2007; Clapp, 2009; Levy and Newell, 2005; May, 2006; Falkner, 2005),
researchsurroundingtheprivatesectorinthehumanrightsandhumanitarianregimeremains
sparse (Weiss, 2013; Zyck and Armstrong, 2014).
Asmentionedinthepreviouschapter,thereissomeworkontheprivatizationofmigration
control responsibilities that has allowed public actors to limit migration (Hern´ andez-Le´ on,
2008; Gammeltoft-Hansen and Sørensen, 2013) and a few case studies that explore oppor-
tunities of private actor engagement in the refugee cause (Zyck and Kent, 2014; Drummond
and Crawford, 2014; Omata and Kaplan, 2013). However, such research either details the
privatesectorasavitalactortocircumcisemigrationorfailstorecognizebusinessesas more
than donors.
Consideringthattheprivatefor-profitsectorcontributeseverystepalongthewaywhenit
comestoprovidingandupdatingrefugeeprotectionandempowerment,itisunsurprisingthat
public-privatecooperationhasbeenincreasing. Therehasbeenatrendtowardscollaborative
projectswherebusinessesandMNCsactivelyanddirectlyprovide,create,anddelivervarious
services (including innovative in-kind donations, skill, language, and job training, novel
technologies and apps that increase connectivity, access to loans, and mapping services).
In short, private actors are pivotal in the construction and implementation of innovative
solutions in these collaborative partnerships.
54
Private engagement can even aid state and peace-building efforts by addressing root
causes and fostering state and institutional capacity during and after conflict periods. Busi-
ness in-kind donations and financial support can enhance state and organizations’ ability
to better protect and empower refugees. With their education, employment, and training
programs, businessesareabletoassistwithrefugeeself-relianceandintegrationinhostcom-
munities. In other words, more resources and programs for refugees alleviate pressures on
publicactors,especiallyinconflict-proneandpost-conflictregions. Goingevenfurther,schol-
ars and practitioners have speculated that cooperation between public and private actors in
the refugee regime could potentially be “the” solution to the global refugee crisis (Betts
et al., 2017). The next section will briefly lay out the theoretical and empirical contributions
of this paper.
3.3 Contribution
Theoretically, thisprojectcontributestothedebateaboutchangingauthorityintheinterna-
tional system and the rise of more diverse forms of cooperation in the humanitarian regime.
Whileonesideclaimsthatinternationalorganizationsandstatesremainthemostimportant
actors, the other side argues that authority and power has been reconfigured with private
for-profitactorsrisingasthenewcentralactors(AbbottandSnidal,2010;Avantetal.,2010;
Mattli and Woods, 2011; Ruggie, 2014).
Theprojectalsofillsacriticalvoidinthemigrationandrefugeestudiesliteraturebecause,
thus far, scholars have failed to consider the full potential of the for-profit private sector and
the diversity and complexity of refugee protection PPPs (Betts et al., 2017). Particularly,
we know little about the factors that influence partnership formation; nor do we know what
this newly emerging network even looks like. This project therefore presents the first (and
to my knowledge the only) study that systematically investigates the factors that make
partnerships in the public-private refugee protection and empowerment network more likely.
55
Empirically, this study produces a new body of data on the refugee regime and presents
the first visualization of the refugee protection network. Employing social network analy-
sis and exponential random graph models (ERGMs) presents an innovative methodological
approach to studying this newly minted network as it allows to map and statistically test
factors that influence partnership formation.
Finally, the internal dynamics of the refugee protection network are connected to the
performance of these refugee protection programs. While beyond the scope of this paper,
future work could explore the link between the formation/ structure of the network and the
resulting protection programs and their effectiveness.
3.4 ATheoryofNetworkEmergenceandTieFormation
This section introduces the factors that influence tie/network formation. First, I theorize
aboutwhich factorsareassociatedwithbetterchancesofbeingconnected. Morespecifically,
if actors have certain attributes, they should have, on average, better chances to form ties
than actors who do not have these attributes.
Second, besides these main effects, I also hypothesize about who partners up with whom:
actors who share certain similar attributes should be on average more likely to form ties
than actors with dissimilar characteristics.
The most influential factors are (1) type of actor, (2) being connected to a central actor,
(3) being from the same geographic region and (4) being the same type of actor as the
partner.
3.4.1 Actor Type
Which actors have better chances of forming partnerships? Literature is divided on this
question,pointingtodifferentactors. Thefirstcamphighlightstheimportanceoftraditional
protection actors such as states and IGOs. Given that states and organizations have been
56
(considered) the central actors in the international system, especially when it comes to
safeguarding rights and addressing global issues such as health, environment, and human
rights, these scholars argue that these actors hold legitimacy and authority and thus have
better chances of forming partnerships (Andonova, 2010, 2017).
The second camp challenges this line of scholarship, pointing out that it underestimates
the impact and power of the for-private private sector. It is one-sided and gives little to
no agency to businesses. Private actors’ influence in global governance has been steadily
expanding: businesses provide expertise and resources essential to traditional protection
actors. Thus, scholars see the for-profit sector has authoritative and powerful, making them
more likely to form partnerships (Levy and Newell, 2005; Mert, 2015a).
Although private for-profit actors are indeed powerful and influential, I join the former
camp, which highlights the continued importance of traditional protection actors. Having
had a long history of working with refugees, states, IGOs, and NGOs have established
legitimacy and authority. They also often control access to refugees and refugee programs.
From this follows:
Hypothesis 1: Businesses are, on average, less likely to form ties/partnerships with other
actors in the network than traditional refugee protection actors (IGOs, NGOs, states).
3.4.2 Legitimacy and Reputation: Connection to Central Actor
In addition, legitimacy and reputation of an actor may play a role. If one has a good
reputation and is seen as legitimate, the actor may be more likely to connect to other actors.
Being connected to an influential, important actor in the network may signal legitimacy and
a good reputation to other actors, thus increasing the respective actor’s chances of forming
connections. In the refugee regime, the UNHCR is considered the most central actor given
its official standing as the UN Agency for Refugees and as the organization responsible
established to ensure rights and responsibilities in the UN Refugee Convention (Barnett and
Finnemore, 2004). Therefore:
57
Hypothesis 2: Actors connected to the UNHCR should be more likely to form ties than
actors not connected to the UNHCR.
3.4.3 Similarity: Geographic Region and Actor Type
Network literature often suggests that birds of a feather flock together, meaning that simi-
larity increases the probability of being connected (McPherson et al., 2001; Fu et al., 2012).
I anticipate this to hold true in terms of geographic region and actor type in the refugee
network.
Although it would be sensible to assume that actors want to form diverse geographic
connections, meaning that powerful Northern actors should value ties with local/Southern
actors,toprovideholisticprotection,Iclaimthatthenetworkwillactuallydepictgeographic
similarity. Instead of seeing North-South connections, I expect to observe more North-
North and South-South connections. Maintaining geographically distant relations may be
complicated and costly, thus prompting actors to connect with geographically proximate
actors (Murdie, 2014, pg.8). Thus:
Hypothesis 3a: Northern actors are more likely to form ties with other Northern Actors,
while Southern actors are more likely to form ties with other Southern actors.
Additionally, one could expect that the diversification and proliferation of actors in the
refugee regime may result in different actor types connecting with one another to diversify
expertise and funding. However, working with partners who are very different from one-
self is costly: it makes coordination, agreement, and potentially distribution of roles and
accountability challenging. In order to mitigate these costs, I expect that actors prefer to
join programs and connect with similar actors. To exemplify, I suspect that businesses will
partner up with other businesses, NGOs with other NGOs etc. I formulate the following
general hypothesis:
Hypothesis 3b: Actors are more likely to form ties with the same type of actors rather
than across actor types.
58
Having introduced the theoretical expectations with regards to tie formation in the
refugee network, I now introduce the research design.
3.5 Research Design
The recent proliferation of diverse actor collaboration involving public and private actors
allows for a unique opportunity to investigate the network emergence process and tie forma-
tion. I construct an original dataset to then explore the network visually and statistically
through social network analysis.
This section will first review the data and data gathering process, then turn to the
definition and operationalization of the explanatory and outcome variables, before finally
discussing the method.
3.5.1 Data
Toilluminatetheemergenceofthenetwork, Iconstructadatasetthatrecordspublic-private
partnerships based on collaborative refugee protection programs and initiatives. This en-
ables me to document multiple rather than simple dyadic-actor relationships and reveals the
diversity of actors coming together for a specific project. Further, collecting program-level
data allows for even more variation in my data: instead of just recording whether two ac-
tors connect overall, which would result in only one observation, partnership choice changes
depending on the program/initiative, resulting in several observations. The network is a 2-
modenetwork: actorsconnectedtoprograms(seefigure5.10inAppendix, foraglimpseinto
the datasets see tables 5.14 and 5.15), from which I derive a 1-mode network: actor-actor
connections based on shared programs (see figure below 3.1).
To construct the dataset, I rely on publicly available data on programs and initiatives
that can be found and accessed on websites of both public and private actors. The main
data source used to build the current network is the report by the International Finance
59
cooperation (IFC) - a sister organization of the World Bank and member of the World Bank
Group - which constitutes the largest global development institution focused on the private
sector in emerging markets.
2
The report identifies programs/initiatives of private sector
engagement and refugees.
The data collection and verification was conducted over a period of over a year (2019-20)
with three undergraduate research assistants double-checking the accuracy of the program
data as well as collecting additional attribute data of the respective public and private
actors.
3
In particular, as I am interested in the factors that make partnership formation
more likely, my research team and I collected meta data on actors’ 1) geographic location, 2)
actortype(business,IGO,etc),and3)connectiontotheUNHCRamongothers. Thisprocess
included researching the actors’ profile on LinkedIn and corroborating the information with
their official websites.
The resulting network dataset consists of 199 refugee protection actors, creating an
overall 199 x 199 adjacency matrix. A tie between two actors represents being part of at
least 1 shared protection program. Actors represented are public actors such as state entites
or IGOs as well as private actors including NGOs, foundations, and businesses. Figure 3.1
allows for an initial visual inspection of the refugee protection network.
Thevisualizationilluminatesvariationinpartnerships: whilesomeprotectionactorshave
many connections to other actors through shared protection programs, others only connect
to one or two other actors. At the core of the network, we observe a few actors with many
connections: thesearehighlycentralactorsastheyareformingtieswithvariousotheractors.
Some of these actors include the UNHCR, the International Rescue Committee, Save the
Children, IKEA, and Microsoft. The different color-coding also underlines the diversity in
the network. There are many different actors (IGOs, businesses, NGOs) participating and
cooperating in refugee protection.
2
Please see the IFC publication for more information.
3
This means that the dataset reflects publicly verified public-private protection and empowerment pro-
grams and projects as of June 2020.
60
Figure 3.1: Refugee Protection Network
Although a visual and descriptive exploration of the network is insightful, it mainly
allows us to observe variation and diversity. To move beyond merely observing variation
and diversity, a statistical modeling approach is necessary. The next section will discuss
exponential random graph models in the context of social network analysis as an appropri-
ate methodological tool to test actor and network properties statistically to understand tie
formation.
3.5.2 Methods
As illustrated in the previous section, I use tools from social network analysis (SNA) to
create the PPP refugee protection dataset and respective network. While the visualization
61
and certain descriptive measures give insights into the variation and clustering within the
network, a more systematic approach is necessary to explore tie formation and partnership
choice.
Luckily, SNA allows for statistical testing of actor characteristics on the propensity of
ties. The most common statistical modeling approach consists of exponential random graph
models (ERGMs). ERGMs are powerful as they are “true generative statistical model[s]
of network structure and characteristics” (Luke, 2015). This means that characteristics of
the individual elements (i.e. actors) and other structural properties can be used to predict
properties of the entire network. For example, individual actor level attributes such as actor
type, geographic region, and connections to important actors can be used to explore the
likelihood of a tie.
Going beyond the question of which factors influence tie formation, ERGMs also provide
a means for exploring who partners up with whom. This can be achieved through an inves-
tigation on the dyad level. The characteristics of two actors in a dyad may play a role in
observingaconnectionbetweenthem. Withthis,itispossibletoexploretieformationbased
on similarity or dissimilarity. This will allow to test for similarity in terms of geographic
region and actor type and will ultimately provide insights into partnership choice.
ERGMsareaclassofmodelssimiliartoregressionorgeneralizedlinearregressionmodels
(GLMs). ThegeneralformoftheERGMspecifiestheprobabilityofthenetworkasafunction
of the features we expect to occur more/less likely than they would by chance:
P(Y =y) =
exp(θ ′
g(y))
k(θ )
(3.1)
where Y is a random network on n nodes (in this case nodes are actors), y is the
observed network, g(y) is a vector of model statistics for network y,θ represents the vector
of coefficients, and k(θ ) is a normalizing constant to ensure that probabilities sum to 1.
The above introduced general ERGM expression for the likelihood of the whole network
can be re-written to reflect the dyad level:
62
logit(Y
ij
= 1|y
c
ij
) =θ ′
δ (y
ij
) (3.2)
in whichY
ij
is a random actor pair i,j with the realization ofy
ij
,y
c
ij
represents the rest
of the network (excluding actor pair i,j), and δ (y
ij
) is a vector of the change statistics for
each model term.
4
In the case of the refugee network, the dyad level ERGM model is specified as follows:
P(Y
ij
= 1|y
c
ij
) =logit(θ ′
edges
δ edges
+θ ′
actortype
δ actortype
+θ ′
connUNHCR
δ connUNHCR
+
θ ′
geohom
δ geohom
+θ ′
actorhom
δ actorhom
)
(3.3)
whereatiebetweenactori andj isconditionalontherestofthenetwork. Thecoefficients
(θ ) are multiplied by the change for the respective model terms (actor type, connUNHCR,
geohom, actorhom).
5
3.5.3 Independent and Dependent Variables
Havingdiscussedthemethodologicalapproach, Inowturntowardsdefiningandoperational-
izing the independent and dependent variables. I first review the IVs: actor type, UNHCR
connection, and geographic region, before turning to the DV: network ties.
4
δ (y
ij
) = g(y
+
ij
)− g(y
− ij
), where y
+
ij
is the same network as y except that y
ij
= 1 and y
− ij
is the same
network as y except that y
ij
= 0
5
Actor type: Type of actor such as IGO, business, foundation, state etc; connUNHCR: whether or not
an actor is connected to the UNHCR; geomhom: whether actors are from the same geographic region;
actorhom: whether two actors are the same type of actor. For variable definition and operationalization see
next section. Change statistics:
δ cat
=
2 if both i and j actor have the characteristic,
1 if either i or j actor have the characteristic,
0 neither i or j actor.
;
δ hom
=
(
1 if both i and j actor have same value for a certain category of a categorical covariate,
0 otherwise.
63
Independent Variable: Type of Actor
• Definition: Main entities involved in the governance of refugee protection.
• Operationalization: Falling into the mutually exclusive categories of being an IGO,
NGO, Business, Foundation, or State.
• Measurement: Categorical
Independent Variable: UNHCR connection
• Definition: Whether actors are part of the same protection program as the UNHCR.
• Operationalization: Actors forming/not forming a tie with the UNHCR.
• Measurement: Binary (0/1)
Independent Variable: Geographic region
• Definition: Actors belonging to either the Global North or Global South.
6
• Operationalization: Main headquarter location in the Global North or Global South.
• Measurement: Categorical/Binary (North/South)
Dependent Variable: Network Ties
• Definition: Cooperation between actors within the network based on being part of the
same refugee protection program.
• Operationalization: Presence/Absence of tie between pairs of actors.
• Measurement: Binary (0/1)
Having discussed the research design, the next section will present and review the em-
pirical results.
6
I use a binary classification for geographic region because adding a lot of model terms (different geo-
graphic regions) to an ERGM decreases the likelihood of model conversion. This is especially pronounced
when there are only a few actors in certain regions, which is the case in this network. Faced with a similar
problem, Murdie (2014) in her network study also uses a binary classification for geographic location.
64
3.6 Results
Table3.1displaystheresultsoftheERGMs. Column1presentsthesimplestmodelincluding
only the edges (=ties) term, which represents the density of the network. The negative
coefficient signals a rather sparse (not very dense) network. This is sensible and common in
larger networks: the more actors exist, the less likely it is to have connections to many/all
actors. The overall density of the refugee protection network is 0.0314.
7
Column 2 reports the estimates for the main effects of actor type and connection to
the UNHCR on the likelihood of tie formation. For actor type, business is used as the
baseline. It appears that being an educational institution, a foundation, an NGO, an IGO,
or a state increases the chances of forming ties compared to being a business. These results
are statistically significant. When it comes to connections to the UNHCR, the results are
statistically significant and align with my hypothesis. Actors connected to the UNHCR are
indeed more likely to form ties than actors who are not connected to the UNHCR.
Column 3-4 report the main effects and include the similarity terms. The results for
the main effects remain mostly robust: being any other actor type than a business overall
increases chances of forming ties in the network. However, only the results for educational
actors and IGOs are statistically significant in model 4. When it comes to being an NGO,
model 4 depicts a negative relationship, indicating that the likelihood of forming ties for
NGOs is lower than that of businesses (again not statistically significant). Connections
to the UNHCR and likelihood of tie formation continue to show a positive, statistically
significant, association.
With regards to geographic location, Northern actors are, on average, more likely to
partner up with other Northern actors. Similarly, Southern actors tend to form ties more
readily with other Southern actors. Column 3 reports the average effects for actor type
matching: there is an overall positive, statistically significant, correlation, indicating that
7
Density of a network is calculated as:
# of ties
# of all possible ties
. Existing ties 619, possible ties 19713. The
edges coefficient here is in log-odds. To get the respective density of 0.0314: exp(-3.428)/(1 + exp(-3.428))
65
the same type of actors tend to connect to other actors of the same type rather than across
actor types.
Column 4 disaggregates actor type and provides more detailed estimates for the each of
thesimilarityactorterms. WhatstandsouthereisthatespeciallyNGOsconnectmoreoften
to other NGOs, and foundations have better chances to collaborate with other foundations.
While not statistically significant, businesses and IGOs are not more likely to partner up
with other businesses or IGOs respectively.
Model 4 also reaffirms that businesses have a more difficult time forming partnerships in
the network. Both the coefficients for connection to the UNHCR and geographic matching
remain positive and statistically significant.
Finally,model2-4introduceGWDegree. Thistermrepresentsthedependence/geometric
termandreferstogeometricallyweighteddegreedistribution. Putdifferently, itaccountsfor
how actors with different centrality scores (higher centrality = more ties) connect to other
actors. This term accounts for dependencies in the network and captures the “popularity
effect.” It illuminates that connections are indeed not independent but dependent on actors’
centrality/popularity in the network. GWDegree estimates the change in tie likelihood
given the degree of the actors involved, but with marginally decreasing weighting as degree
increases. Thenegativeandstatisticallysignificantcoefficientreflectsanincreasedlikelihood
of actors to form ties to higher-degree actors.
Although the above discussed results show statistical significance and direction of the
estimates, it is difficult to interpret their magnitude. To ease interpretation, Table 3.2
provides odds ratios for the most specified model (4) (see tables 5.19 and 5.20 in Appendix
for odds ratios of model 2 and 3). Not being a business makes some actors more likely to
form ties in the network. For example, being an educational institution (1.697) makes it 1.7
times more likely to form ties; being an IGO (2.992) makes it 3 times more likely to form
ties compared to businesses. While not statistically significant, states and foundations also
have a better chance connecting with actors in the network compared to businesses. Only
66
Dependent variable:
Network Ties
(1) (2) (3) (4)
Edges − 3.428
∗∗∗ − 3.825
∗∗∗ − 4.494
∗∗∗ − 4.107
∗∗∗ (0.041) (0.116) (0.153) (0.190)
Education 0.536
∗∗∗ 0.701
∗∗∗ 0.529
∗∗∗ (0.157) (0.159) (0.178)
Foundation 0.243
∗∗ 0.265
∗∗∗ 0.053
(0.096) (0.101) (0.135)
IGO 1.097
∗∗∗ 1.203
∗∗∗ 1.096
∗∗∗ (0.111) (0.112) (0.138)
NGO 0.407
∗∗∗ 0.321
∗∗∗ − 0.091
(0.063) (0.062) (0.132)
State 0.176
∗ 0.305
∗∗∗ 0.150
(0.096) (0.102) (0.133)
Conn.UNHCR 0.121
∗∗ 0.130
∗∗ 0.132
∗∗ (0.051) (0.054) (0.054)
Match.Geo.North 0.652
∗∗∗ 0.656
∗∗∗ (0.095) (0.094)
Match.Geo.South 0.572
∗∗ 0.579
∗∗ (0.246) (0.245)
Match.ActorType 0.543
∗∗∗ (0.099)
Match.Business − 0.035
(0.215)
Match.Education 0.416
(1.089)
Match.Foundation 0.789
∗ (0.460)
Match.IGO − 0.234
(0.534)
Match.NGO 1.055
∗∗∗ (0.178)
Match.State 0.303
(0.569)
GWDegree − 2.514
∗∗∗ − 2.118
∗∗∗ − 2.130
∗∗∗ (0.230) (0.257) (0.253)
Akaike Inf. Crit. 5,504.220 5,264.502 5,180.624 5,176.071
Bayesian Inf. Crit. 5,512.108 5,327.609 5,267.397 5,302.286
Note:
∗ p<0.1;
∗∗ p<0.05;
∗∗∗ p<0.01
Table 3.1: Results for ERGMs
67
being an NGO appears to lower chances (8.7 percent) of forming partnerships compared to
businesses (again, not statistically significant). Having a tie to the UNHCR increases the
odds of forming partnerships by 14 percent.
Odds ratio 2.5 % 97.5 %
Edges 0.016 0.011 0.024
Education 1.697 1.196 2.407
Foundation 1.054 0.809 1.374
IGO 2.992 2.285 3.918
NGO 0.913 0.705 1.182
State 1.162 0.895 1.508
Conn.UNHCR 1.141 1.027 1.267
Match.Geo.North 1.927 1.603 2.315
Match.Geo.South 1.783 1.104 2.880
Match.Business 0.965 0.633 1.472
Match.Education 1.516 0.179 12.812
Match.Foundation 2.202 0.894 5.423
Match.IGO 0.791 0.278 2.255
Match.NGO 2.873 2.025 4.076
Match.state 1.353 0.444 4.128
GWDegree 0.119 0.072 0.195
Table 3.2: Odds Ratios Model 4
Turning to the odds ratios for the similarity terms, geographic proximity appears to
matter when forming partnerships. Northern actors are almost twice as likely (1.927) to
connect to other Northern actors. Similarly, Southern actors are 1.8 times (1.783) more
likely to form partnerships with other Southern actors.
Similarity based on actor type produces mixed results. In particular, it is present for
somebut notallactors. Forexample, foundationsareabout2.2timesmorelikelytoconnect
to other foundations, and NGOs are almost 3 times (2.873) more likely to collaborate with
other NGOs. On the contrary, businesses are about 3.5 percent less likely to partner up with
other businesses, and IGOs are almost 20 percent less likely to form ties with other IGOs
(however, these results are not statistically significant).
68
3.7 Discussion
The findings of the ERGMs align with my theoretical expectations. In Hypothesis 1 I
hypothesized that businesses are less likely to form ties in the refugee protection network
compared to other actors. I find preliminary support for this claim as being any actor
other than a business mostly increases the odds of tie formation. This lends support to the
literature that emphasizes the continued relevance and authority of IGOs and states. While
businesses are certainly powerful and influential, actors may be sceptical to partner up with
them given their lack of experience in human rights and humanitarian work as well as their
potentially profit-driven motives.
The results for UNHCR connections also lend support to Hypothesis 2 as I expected
there to be a positive association. All models show that actors affiliated with the UNHCR
have indeed better odds of forming ties in the refugee protection network than those uncon-
nected to the agency. This finding illuminates the importance of the UNHCR not only for
providing services and protection to refugees but also as for facilitating partnerships across
different actors. This is pivotal as a more inclusive network allows actors to pool resources,
diversify expertise, and thus create more sophisticated refugee protection programs.
When it comes to the homophily (similarity) effects, the results conform with Hypoth-
esis 3a. I anticipated within-geographic region connections to be more prominent than
across-geographic region connections. Northern actors are indeed more likely to tie with
other Northern actors, and Southern actors link up more with other Southern actors. While
supporting my hypothesis, this raises some concerns. Geographically diverse connections
may enhance communication and transparency in the overall refugee protection network.
Local actors in the Global South may need support from larger, powerful Northern actors.
Similarly, Northernactorscanbenefitfromthelocal, in-depthknowledgeofSouthernactors.
Cross-geographic connections could produce more fine-tuned programs that better cater to
refugee needs.
69
Finally, in terms of actor type homophily, I find mixed results for Hypothesis 3b. For
some actors, connections between the same type of actors are more common than across
actor types. This aligns with the literature that suggests that alike actors tend to form ties
more readily and cluster together (Fu et al., 2012; McPherson et al., 2001). It may be less
costly to partner up with similar actors who have a potentially similar bureaucratic culture,
motivations, and goals.
However, as seen in the analysis, actor homophily is not universal. Businesses and IGOs
do not more readily form ties with other businesses and IGOs respectively. When it comes
to businesses, this is sensible: given that businesses lack experience and expertise working
with refugees, they are more inclined to connect with actors who have been active in refugee
protection. Big, powerful IGOs may value diverse connections more than partnering up with
other powerful IGOs.
3.8 Conclusion
What influences the partnership dynamics within the refugee protection network, and who
partners up with whom? This chapter has illuminated that type of actor, geographic region,
and being connected to the UNHCR influences the likelihood of tie formation. As expected
businesses have a more difficult time forming ties compared to more traditional protection
actors. However, being connected to the UNHCR enhances one’s chances of forming ties in
the network. Similarity also appears to breed connections most of the time: actors within
the same region and actors of the same type flock together.
This project provides first insights into the changing nature of the refugee regime and
takes seriously the increased presence of for-profit actors, thus filling a critical void in the
refugee and migration literature. It also sheds light onto the main actors and authority in
the international system and contributes to the debate of who governs the globe.
70
The current data and results produces insights into the internal network dynamics. Once
actors have made it into the network, there are factors that play a role for the within-
network partnership variation. As we have seen, actor type, connection to important actors,
and homophily help explain the formation of partnerships.
While this chapter has shed light onto some of the factors that impact partnership for-
mation within the network, it falls short in some areas. For example, it cannot offer an
explanation on who makes it into the network in the first place. Thus, we do not know the
factors that influence actors gaining access to the network.
Further, this chapter also does not reveal the motivations and incentives for the different
actors to form partnerships. In chapter 2 we saw that potential pressure from consumers
and customers may incentivize businesses to get involved and start forming partnerships.
However, itislikelythattherearemultiplereasonsatplayastowhybusinessesgetinvolved.
Similarly, what are the reasons for states and organizations to partner with the for-profit
sector? Accepting the for-profit sector into this network may be risky as businesses and
corporations are inexperienced and profit-driven.
Finally, quantitativeresultspresentthebroadfactorsthatinfluencewithin-networkpart-
nership formation. However, the analysis does not provide in-depth insights into the actual
partnership formation processes and their complexities. Chapter 4, the final empirical chap-
ter of this dissertation, steps in to complement Chapter 3 by addressing these limitations
and exploring the motivations of different actors as well as the long partnership finding and
creation process and program evaluation practices.
71
Chapter 4
Partnering for Good: How and Why Actors Form
Public-Private Partnerships as a Response to the
Global Refugee Crisis
4.1 Introduction
How do public-private partnerships (PPPs) in the refugee protection and empowerment
space form? For the longest time, supporting and protecting refugees was mainly seen as
theresponsibilityofintergovernmentalorganizations(IGOs),nongovernmentalorganizations
(NGOs) and states. However more recently, the for-profit sector consisting of businesses and
multinational corporations has stepped in, creating projects and programs together with the
aforementioned traditional refugee protection actors.
Compared to other fields where PPPs have been prevalent (infrastructure, health care,
development), these multi-stakeholder partnerships are rather rare and underexplored in the
humanitarian space. Finding and forming partnerships is complex to begin with, however, it
isevenmorechallengingforactorsfromdifferentbackgroundsandsectorsattemptingtowork
togethertosupportthemostvulnerablepopulationsintheworld. Howdothesepartnerships
arise, and how do actors evaluate their success? This chapter complements chapter 3 as it
setsouttoexplorethedecision-makingprocesses,motivations,anddifficultiesofestablishing
these partnerships and evaluating collaborative programs.
72
Inthischapter,Iargueandfindthatformingpartnershipsremainsremarkablydifficultfor
businesses because organizations and states gatekeep access to asylum seekers and refugees.
The most important factors for having a chance of being considered as a partner are 1)
reputation, 2)typeofindustry/sector, and3)visibility. Whenactorsnegotiatepartnerships,
discovering overlapping values and goals as well as sharing risks and rewards are essential for
thesuccessfulfinalizationofpartnershipsandrespectiveprograms. Theprogrammonitoring
and evaluation process is highly contested, with actors disagreeing on how to measure and
report success. One surprising finding was the harsh criticism voiced by the business sector
about the vague performance frameworks of the humanitarian sector.
Through a qualitative analysis of interviews with various stakeholders complemented by
evidence from conferences/meetings and reports, this paper is able to go beyond publicly
availableinformationanduncoverthepartnershipformationandprogramevaluationprocess
as well as stakeholder sentiments and perceptions. The paper proceeds as follows: the first
sections review the contribution and theoretical expectations before turning to the research
design and empirical results; it concludes with a discussion of the empirical results.
4.2 Contribution
Extant literature has theorized and investigated why and under what conditions PPPs form,
and what makes them more ore less legitimate and effective (Beisheim and Liese, 2014;
Pattberg and Widerberg, 2016; Sch¨ aferhoff et al., 2009). For example, scholars have found
the rise of PPPs to be a result of increased globalization and the diversification of players
participating in the global arena. They are assumed to form in an effort to fill governance
gapsandtoaddressunresolved,pressingproblems(Andonova,2017;BuseandHarmer,2007;
Reinicke,1998). DifferentinstitutionalconstellationsarearguedtoaccountforPPPsvarying
legitimacy and effectivneess (Beisheim and Liese, 2014; Mert, 2015b; Pattberg et al., 2012;
Sch¨ aferhoff et al., 2009; Witte and Reinicke, 2005).
73
However, literature has yet to develop a more granular theory of the processes and dy-
namics that actually go into the successful formation and performance evaluation of such
PPPs. Weknowlittleabouthowandwhenpartnershipsformornotform, howactorsdecide
who to partner and not to partner with, and how these programs/projects are evaluated and
deemed successful across partners.
This paper attempts to fill this gap and develops a theory of PPP partnership forma-
tion and program evaluation. It takes into account the motivations, considerations, and
frustrations throughout the process. This theory transcends the refugee protection and em-
powerment space and can be applied to the entire humanitarian and development field and
potentially even to all PPPs.
In addition, the paper is also empirically valuable as it traces the development of this
newly minted refugee protection and empowerment PPP network as it unfolds. With this,
thedatarepresentsoneofthemostup-to-dateaccountsonthesepartnershipprocessesinthe
refugee protection and empowerment space. It offers rare insights into sentiments and per-
ceptionsofmultiplestakeholdersworkingtogethertoupdateanddevelopmoresophisticated
support and protection for some of the most vulnerable populations in the world.
With the relentless unfolding of the crises in Afghanistan and Ukraine that have resulted
in millions of forcibly displaced people, these multi-stakeholder partnerships have become
ever more relevant. Understanding the dynamics of these PPPs as well as reavling the
sentimentsandfrustrationsofpartners, allowstoaddressshortcomingsearlyonandincrease
chances of their success.
4.3 TheoryofPartnershipCreation&EvaluationProcess
in Refugee Empowerment and Protection
Thefollowingsectionwilladvanceatheoryonhowthisrefugeeprotectionandempowerment
PPP network emerges and how the resulting partnership programs are evaluated. This
74
includes theoretical expectations on who initiates and decides, what factors play a role for
forming and negotiating partnerships, and how actors evaluate and report program success.
To develop this theory, I take a multi-disciplinary approach, incorporating insights from the
international relations and business literature.
4.3.1 Who Initiates and Decides?
I argue that the emergence of this new public-private refugee protection and empowerment
network is mainly a result of organization and state initiation and entrepreneurship. This
aligns with the scholarship on PPP initiation that contends that public actors remain most
important, especially when it comes to safeguarding rights and addressing global issues such
as health, environment, and human rights (Andonova, 2010, 2017).
This does not mean that private actors such as businesses do not reach out, engage, and
attempt to form partnerships. However, I expect that organizations and state actors are
more likely to engage, and are the main decision-makers when it comes to partnering up
with private actors for several reasons.
First, these actors hold the monopoly of expertise and access. In terms of expertise, gov-
ernment agencies and humanitarian organizations have on-ground and often long-standing
experience working with forced migrants, thus being familiar with the various needs and cir-
cumstances of refugees as well as being able to manage and navigate both crisis/emergency
situations and more stable/protracted environments. As a result, they are more proactive
seeking respective partnerships that complement their endeavors. In regards to access, gov-
ernment agencies and organizations are the ones that control access to refugees and refugee
camps. Being in charge means that these organizations can somewhat limit and decide who
else may gain access, thus allowing them to choose more liberally whom to partner up with.
Second, states and organizations also initiate these partnerships to hold private actors
accountable and to ensure transparency. While states and organizations are often the gate-
keeperstorefugeesandrefugeecamps,itispossiblethatprivateactorsneverthelessengagein
75
directrefugeeprotection(withoutpartneringupwithanypublicactors). Inanefforttoavoid
this private takeover and to monitor and control private actions, states and organizations
initiate these partnerships.
Finally, states and organizations face resource shortages and funding limitations. To
diversify their funding resources and their service and product portfolio, they seek out part-
nerships with the private sector. The above discussed reasons lead me to hypothesize:
Hypothesis 1: Organizations and states are the main entrepreneurs of partnerships with
the private for-profit sector.
4.3.2 Restricted Access: Reputation and Type of Industry
In addition to who creates these partnership programs, it is essential to explain the varia-
tion in partnerships. This raises the question of partnership vs. no-partnerships. Current
literature mostly observes and analyzes partnerships, leaving out the counterfactual of non-
partnerships (and thus selecting on the dependent variable). However, understanding the
whole partnership formation process requires us to also explore the factors that influence
which actors even gain access to be part of the refugee protection and empowerment com-
munity, who is having difficulties joining and why, and who is excluded from ever joining.
Scholars have explored the importance of actor reputation for international cooperation
andcoordination(Axelrod,1984;GrayandHicks,2014;Keohane,1984;Rioux,2008;Wilson
and Sell, 1997). For example, Mercer (1996) has illustrated how reputation is constructed
through biases, thus making it difficult for actors to change their image once they have
acquired a certain reputation. Tomz (2007) also shows that actors are indeed treated differ-
entlydependingontheirreputation: theymayreceivebetter/worsedealsandtheymayloose
access to markets and have to go through an arduous signaling and reputation rebuilding
process.
Scholarshavealsoexploredtheimportanceofreputationspecificallyfornon-stateactors.
Reputation is especially vital in the non-profit space “in order to distinguish themselves
76
from the crowd by establishing an identity or brand” (Gent et al., 2015, pg.430). A good
reputation allows NGOs to secure funding and create trust with donors (Boulding, 2009;
Gugerty, 2009). In management and business research, corporate reputation has taken the
center stage since the 1990s. The reputation of a business touches all areas of operations: it
signalstrusttovariousstakeholders(vendors, partners, consumers)(Ebert,2009;Kimetal.,
2008), helps to set prices and increase profits (Rindova et al., 2005; Roberts and Dowling,
2002), provides a competitive advantage (Hall, 1992; Schwaiger and Raithel, 2014), builds
brand image/identity (Cretu and Brodie, 2007; Hur et al., 2014).
Itisthussensible toassumethatthedifferentactorspayspecialattentiontotheir poten-
tial partners’ reputation. I expect to see the use of reputation mostly refer to ethical/moral
practices and behavior. This could come in form of ethical business/organizational practices
and absence of scandals. This is especially important for entities desiring to work in the
humanitarian field and with vulnerable populations such as forcibly displaced people.
In short, actors will want to avoid partnering up with the wrong type of partner because
it may undermine and potentially harm their own reputation. Partner selection is a careful
and strategic process that may take a long time. Thus, reputation is one of the main partner
selection criteria:
Hypothesis 2a: Actors are more likely to partner up with partners that have a good repu-
tation.
An additional hurdle, which mainly the business sector faces when forming partnerships,
is the type of industry. States and organizations not only rule out certain private actors
based on their reputation, but also based on the industry sector. Any business actor that
operates in an industry that is seen as harmful and potentially counterproductive to refugee
protection and empowerment should have a difficult, if not impossible, time to find partners.
From this I hypothesize that the type of industry is a crucial factor in partner selection and
that:
77
Hypothesis 2b: State and organizations are unlikely to partner up with for-profit private
actors that are part of harmful industries.
4.3.3 Partnership Negotiation: Values, Goal, Risks, and Rewards
Once actors have carefully selected potential partners, the partnership negotiation process
begins. This arduous process has been scarcely theorized and explored in literature. I thus
draw from both IR and business scholarship to derive the following hypotheses.
First, and foremost, there needs to be an alignment of values and goals for a partnership
to be formed. Chapter 3 showed that alike actors more readily form partnerships. Litera-
turehasalsoexploredthatespeciallylike-mindedactorsaremorelikelytoformpartnerships;
this is also evident empirically if we consider various international coalitions of actors with
similar worldviews and missions (Murdie, 2014). This is even more relevant, yet ever more
challenging, in the case of public and private actors partnering with one another. States,
organizations, andbusinessescomefromandoperateinverydifferentworldsandhavediffer-
ent motives, values, and goals. Scholarship on PPPs has stressed the importance of finding
overlap in values and goals for partnership and project construction (Arya and Salk, 2006;
Nelson, 2002). I assume that identifying shared values and goals is an integral part of the
partnership formation process and hypothesize:
Hypothesis 3a: Private and public actors are more likely to partner with one another if
their values and goals align.
Second, the partnership needs to be mutually beneficial to all parties involved. IR liter-
ature has explored mutual gains and payoffs in cooperation (Axelrod, 1984; Keohane, 1984;
Keohane and Nye, 1977; Nelson, 2002). However, while shared rewards are sensible, risks
also need to be shared. Business scholars have stressed the concept of mutuality in terms
of both rewards and risks (Kernaghan, 1993; Koppenjan, 2005; Nelson, 2002). Sharing both
risksandrewardsissensiblegiventhatvariousactorswithdifferentspecialitiesandexpertise
78
are coming together. This allows actors to distribute and potentially balance out benefits
and risks. I therefore expect to see actors pointing towards shared rewards and risks in the
partnership process. From this follows:
Hypothesis 3b: Private and public actors are more likely to partner with one another if
there are mutual gains and shared risks.
4.3.4 PartnershipandProgramSuccess: EvaluationofPerformance
Oncepartnershipsareestablished,howdoactorsknowtheresulting,jointprograms/projects
are successful? To be clear, my theory does not test whether these programs or projects
are successful, but rather how actors evaluate, report, and classify success. It explores
the internal dynamics of these partnership and program evaluations rather than the actual
output. This is important because insights into the evaluation and reporting dynamics can
be indicative of how rigorous, valid, and potentially trustworthy the respective reported
“successes” of these projects are.
Especially because scholars have questioned the evaluation and performance strategies of
humanitarian organizations given that these actors have an incentive to over-report success
and under-report failures (Cavill and Sohail, 2007; Ebrahim, 2003; Reimann, 2005; Steffek
and Nanz, 2008). Looking “inside the machine” allows us to gain a better understanding of
what we should/can expect from performance reports.
GiventhatPPPsinvolveamultitudeofdifferentactors,theremaybemoreaccountability
and potentially higher standards when it comes to evaluating and reporting performance of
joint projects/programs (B¨ ackstrand, 2008). Nevertheless, we know very little about these
PPPevaluationandreportingdynamics: howdoactorsagreeonwhatisandisn’tsuccessful,
and how are findings reported?
Iexpecttheperformanceevaluationandreportingtobecomplicatedandhighlycontested
astheseactorshavevastlydifferentevaluationandreportingstructuresandstandards. Thus,
I expect to find disagreement amongst partners on how to best evaluate and report:
79
Hypothesis 4a: Actors often disagree on how to evaluate report, and classify success.
As a result of this contestation, I expect actors to produce their own evaluations and
reports rather than joint findings.
Hypothesis 4b: Actors tend to conduct internal evaluations and report their own findings
rather than the joint findings.
This concludes all theoretical expectations. To synthesize and recall the hypotheses and
the main variables of interest, table 4.1 provides a quick overview.
Hypothesis Stage of Process Expected Variables of Interest
H1: Entrepreneurs and Initiators Organizations and states
H2a,b: Partnership Initiation and Dealbreakers
Reputation;
Type of Sector
H3a,b: Partnership Negotiation
Shared values and goals;
Shared rewards and risks
H4a,b: Program Monitoring and Evaluation
Contested amongst partners;
Individual reports
Table 4.1: Short Summary of Hypotheses
The different stages of the partnership creation and program evaluation process can
also be visualized. Figure 4.1 presents the different steps along the partnership formation
and evaluation journey. The four quadrants show partnership entrepreneurship, partnership
dealbreakers, partnership negotiation, and finally program evaluation. To better disentangle
the different stages, the process is presented sequentially; however it is important to note
that these different steps/stages often overlap. As pointed out above, the process is also
iterative and can therefore be imagined as a cycle.
Theupperleftquadrantreflectstheinitiationandentrepreneurshipstagethatconsistsof
public and private actors. States and organizations are grouped together as I hypothesized
that they are the main entrepreneurs in the process.
The upper right quandrant moves on to the partnership initiation process by focusing
on partnership dealbreakers. The two main factors are reputation and type of industry that
complicates or even prevents actors from joining partnerships.
80
This then leads into the lower right quadrant, which presents shared values & goals as
well as mutual gains & risks as the most influential factors in the partnership negotiation
stage.
Finally,thelaststageistheprogramevaluationstage,whereIexpecttofindcontestation
amongst actors when it comes to evaluating and monitoring program performance as well as
predominantlyinternalprogramevaluationsandseparatereportingofresults. Afterorwhile
program monitoring and evaluation, there is an opportunity for actors to renew or construct
new partnerships, which then lead back into the first quadrant of entrepreneurship.
States & organizations
Businesses
Reputation
Type of Industry
Shared values & goals
Mutual gains & risks
Program performance
contestation
Internal evaluations &
separate reports
Partnership Entrepreneurship Partnership Dealbreakers
Program Evaluation Partnership Negotiation
Partnership Creation &
Program Evaluation Process
Figure 4.1: Theory of Partnership Creation & Evaluation Process in Refugee Empowerment
and Protection
81
4.4 Research Design
4.4.1 Data Sources and Methods
To understand the processes behind creating these diverse partnerships as well as the eval-
uation of program/project performance, I conducted a series of remote, semi-structured
interviews with representatives from NGOs, IGOs, governments, educational institutions,
foundations, and businesses. Semi-structured interviews are well-suited for this research
project as they allow a flexible and iterative exchange between interviewer and interviewee
(Roulston and Choi, 2018). The respective data goes beyond publicly available data and a
quantitativeinvestigationofthenetworkasituncovers1)partnershipsandnon-partnerships,
2) motivations for partnerships, 3) initiators of partnerships, 4) partnership processes and
hurdles, 5) program performance and perceived success.
A central issue with interview material concerns the recruitment of relevant participants
andthesamplingofsuchparticipants. Iidentifiedappropriatecontactsfrommyexisting,ex-
pansive public-private actor dataset. I then qualitatively identified partnership departments
and contact details found on public websites and reports. I used a combination of cold-
emailing respective participants as well as snowball-sampling, where interviewees pointed
me towards relevant future study participants (Goldstein, 2002; Tansey, 2007).
Overall, the interview data includes evidence from a heterogeneous set of actors across
different sectors, geographic location, and organizational size to increase representativeness
of the sample. A total of 15 participants volunteered to be interviewed for an average of 30
to 60 minutes. In order to maintain the participants’ anonymity and privacy, no identifying
information was collected or recorded. Table 4.2 summarizes a few characteristics of the
interviewees.
The primary interview data is complemented by insights gained from public recordings
of stakeholder meetings and reports on partnership creation and program performance. As
82
Actor Interviews Geographic Region
Size
(Employees)
IGO 3 Europe, US >10,000
NGO 4
US, Europe, Middle
East
5-5,000
State 2 US, Europe >10,000
Businesses 4
US, Europe, Middle
East
5-10,000
Foundations 1 Middle East 201-500
Academic 1 US 501-1,000
Table 4.2: Overview of Interview Participants
it is often difficult to reach and access interview participants from big companies or organi-
zations, these public meetings and interviews represent substitutes (to a certain extent) of
perspectives and viewpoints of these elite actors. Public documents often describe the part-
nership creation criteria and dynamics as well as program evaluation in more depth than the
interviews revealed. Additionally, these oral and written communications reveal how and
what actors choose to communicate to the public about these partnerships and programs.
4.4.2 Data Analysis
I followed a deductive approach for both constructing and carrying out the interviews as
well as for the subsequent data analysis. This approach is well-suited in my case as I have a
clear theoretical framework and expectations around which the interviews were constructed
(Mihas, 2019; Bingham and Witkowsky, 2022).
Asaresult,Ihadavarietyofthematiccategoriesinadvancetoguidetheinterviewprocess
and ensure relevance to my research topic. The broad categories included 1) characteristics
for partners, 2) non-partnerships (partnership dealbreakers), 3) partnership formation pro-
cess,4)structureofpartnerships,and5)evaluationofsuccessandperformanceofpartnership
programs (please see Appendix C for interview example questions). These broad categories
are directly related to my hypotheses, yet avoid priming or leading participants into specific
83
answer paths. Consequently, participants still had enough room to describe their own ex-
periences, motivations, and frustrations in the partnership creation and program evaluation
processes.
Following the completion of the interviews, I transcribed, stored, and analyzed the au-
dio recordings in NVivo. I used qualitative content analysis where I coded relevant data
excerpts that map onto the above mentioned categories.
1
This coding allows to identify pat-
terns as well as frequencies in the data. Through this conceptual analysis, I was able to con-
firm/disconfirmmyhypotheseswhilediscoveringuntheorizedand/oradditional/complementary
insights.
4.4.3 Ethical Concerns
Conducting research with people warrants a discussion of potential ethical concerns and
risks. My research as well as the specific interview tools were reviewed and approved by
the USC Institutional Review Board. In addition, people interviewed constitute elites and
representatives of organizations and businesses who are often used to being interviewed.
All participants were informed about the interview process, offered additional background
informationaboutthestudy(ifdesired),andgivencontactinformationofUSC’sInstitutional
Review Board. Study participants were also informed that the interviews are recorded and
given the option to opt-out of the recording or of the entire interview at any point of the
study.
While my topic revolves around asylum seekers and refugees, which constitute a vulner-
able population, no sensitive or deceptive questions were asked. Thus, the overall risk to
study participants is minimal. The data is kept private and anonymous with only myself
having access.
1
I relied on Galletta (2013) and Lester et al. (2020) to guide the qualitative interpretation and coding
process.
84
4.5 Empirical Findings
The following section presents the empirical results of this study grouping together the
themesbasedontheirrelevancetothetheoreticalexpectations. AsdiscussedintheResearch
Design section, I discuss the most commonly occurring themes that arose from interviews
and public reports and meetings/conferences. Before, diving into the partnership formation
process and the related findings, I start this section off with reviewing the motivations of
public and private actors for even forming partnerships in the first place.
4.5.1 Why Partner? Motivations of Public and Private Actors
Unsurprisingly, the different actors involved in refugee protection have different motivations
as to why they seek partnerships. Organizations, for example, unanimously identified access
to more funding and resources as one of the major reasons. One interviewee admitted “we
need the funding [...], and we are very keen that [private actors] provide[] that.” Another
interviewee confirmed this sentiment saying that partnerships with the private sector often
start “due to a lack of resources than anything else.”
These findings corroborate what non-profits have mentioned in reports and public meet-
ings, as well as what other scholars have found in their case studies (IFC and Bridgespan,
2019; Malik et al., 2018b). This is sensible given the severe funding gaps organizations face.
For example, the UNHCR reports a 53 percent funding gap ($4.803 billion) as of August
2021 (UNHCR, 2020, pg.7).
Whilefundingiscertainlyimportant,organizationsaswellasgovernmentrepresentatives
highlighted reasons that go beyond funding and donations. In particular, the narrative that
the for-profit private sector should merely be funding/donating partner has been slowly
changing in the humanitarian space.
One interviewee remembers the evolution of the private sector into more than a “cash
cow.” She stated that her team “used to joke and call it ‘grip and grin’ [...] somebody is
85
carrying a big oversized check and you grip and grin, shake hands and you give it. Thank
god, these days are way over.” Organizations and states recognize the potential of the pri-
vate sector in bringing expertise and developing innovative solutions. When asked about
why they partner with the for-profit sector, interviewees frequently mentioned expertise and
specializations of the private sector.
Whenitcomestoinnovativesolutions,oneparticipantsummarizedthismindsetwhenshe
said that partnerships are “not just combining each other’s expertise or resources. You are
tryingtocreate[asolution]thatdidn’texistbecauseyouaretogether. Becauseotherwisewhy
be together?” Government representatives pointed towards the essential role of the private
sector for “long-term, sustainable solutions” because of “jobs... they are more important
than grants” when it comes to successfully integrating asylum seekers and refugees.
Scholars and policy experts have identified a combination of factors as to why the pri-
vate for-profit sector has increasingly become active in the refugee protection space. The
most common ones can be summarized as 1) expanding market and business, 2) developing
research and development (R&D) insights, and 3) fostering corporate social responsibility
and/or social impact engagements (Bisong and Knoll, 2020). My findings support these
broad themes and reveal the melting of normative responsibility (“right thing to do”) rea-
sons with market-based/market-gap incentives.
These responsibility and market based motives became apparent throughout all inter-
views with the private sector. When asked why their companies decided to get engaged in
refugee protection, almost all stressed that the private sector had the responsibility to help
while seeing a clear market gap.
One interviewee said “we were the only solution that we knew, this was in 2015, it was
actually our responsibility to do something [...] We understood that we had a solution that
may work and the refugees are in a desperate situation and no-one else is doing anything in
this field.” In a similar vein, another participant stated that there was a “clear and urgent
need” and “a clear market gap and a clear opportunity gap.” Yet another one confirmed
86
that their decision to become engaged was a “humanitarian [one]” but also because “there
was a pretty good opportunity [...] there was a lot of talent sitting idle in refugee camps.”
While businesses may be motivated to join the refugee protection space and form hu-
manitarian partnerships, it is not given that they will be able to participate and/or even
find partners. As the next two empirical sections illustrate, organizations and states remain
largely in control of who joins. There are clear partnership dealbreakers that prevent certain
actors from joining in the first place.
4.5.2 Partnership Entrepreneurs: Organizations and States as
Gatekeepers
Interviewees were also asked about who initiates and who is the main entrepreneur of these
partnerships. Whilemanystudyparticipantssaiditisbothsides-publicandprivateactors-
who reach out and initiate, it became clear throughout conversations that organizations and
state actors are often the main entrepreneurs. Since organizations and states control access
toasylumseekersandrefugeesandarefamiliarwithlegalandpolicyframeworkssurrounding
refugeeprotectionandsupport, theytakeonanactiveroleintheinitialpartnershipcreation
stages. In particular, organizations and states often function as gatekeepers, making it
difficult for businesses to become part of the refugee protection space.
This was reflected in how study participants described the arduous vetting and procure-
mentprocess. Especiallyintervieweesfromthebusinesssectorexpressedextremefrustrations
with these processes. One participant described the partnership and procurement process as
follows:
You have to come with a proposal, and then you have to write the proposal, then
you have to plan the proposal so you think you know what you will be doing
and once you have developed that “Luftschloss” as the Germans say “castles of
thin air” , then you also have to create the methodology of how you will measure
that. [...] And that is the procurement process for all of the projects in this field
87
- so no wonder nothing comes out - because that whole process makes everything
impossible.
While there are good reasons for these detailed procurement processes, these processes
work especially well when it comes to delivering tangible goods and services to refugees and
asylum seekers like “tents, pencils, trucks because that is something that can be defined.”
However, whenever it comes to long-term, innovative solutions that involve training, skill
development, education, and employment, the “the classic procurement does not work.”
Two other interviewees corroborated these experiences and feelings towards the procure-
ment and partnership initiation process. One said, these processes are often so taxing and
long-winded that it is difficult for small and medium sized businesses to participate and go
through with the partnership process. It often involves “partnership teams and legal teams”
to write these proposals and follow all the rules. Another one said that he simply “can’t
go to our investors or my employees and say the money we have we’re spending on these
lawyers, they are working on these deals that will never materialize.”
When asked why their businesses not just simply avoid forming partnerships and instead
just support refugees and asylum seekers on their own, most interviewees replied that they
needed to go through organizations and governments to 1) gain access to asylum seekers and
refugees and 2) to navigate the cultural and policy landscape working with these vulnerable
populations. One said that they do not have direct access to asylum seekers and refugees
andneededtoworkwithpartners“whohavedirectaccesstobeneficiaries.”Anotheronesaid
that states and organizations are essential for navigating the “administrative procedures”
and local and national policy frameworks. He continued to highlight that his business had a
razorsharpfocus[...] wearedesigningaservice,wearedevelopingthetechnology
[...] We do nothing else. Everything else is not our business, it is not our turf
and someone else should take care. That means we need an ecosystem of players.
Another interviewee from a big company echoed this sentiment and said that “[interna-
tional organizations] are the ones on the ground, in the camps, taking care of refugees - that
88
is a completely different relationship and access” than a company can offer. He added that
it is not easy for businesses to know all regulations when it comes to working with asylum
seekers or refugees in different countries. Thus, companies need a “network of partners” who
they can rely on.
A study participant from a big organization confirmed this and said that it is difficult
for businesses to know the humanitarian policy world as well as legal frameworks, which
leads the private sector to be “put off by the fact they just don’t know, and it is going to be
complicated.” Another non-profit organization stated that it is indeed challenging for their
business partners to understand what “is and isn’t locally or culturally appropriate.”
Besides states and organizations taking on and reinforcing gate keeping functions, the
interview data pointed towards another important factor for initiating these joint partner-
ships: CEO and/or higher management level support and commitment to the issue. In
other words, many partnerships originated as a result of CEOs, managers, or founders from
businesses, organizations, and governmental agencies meeting and talking at conferences or
other social events.
One interviewee from an NGO recalled that many of their partnerships with the private
sector started when “the two CEOs were together at a meeting and said we want to do
something together. [...] it did not bubble up from the bottom, it is top down only.”
A business study participant affirmed this and stressed that the leadership level needs to
become active because “it does come from the top and their deep connection to the issue
[...].”
In a public conference, a representative of the Tent Partnership for Refugees further
emphasizedhowessentialitistohavetheCEOorleadershiptocometogetherandtocommit
to socially responsible partnerships (Urban Institute, 2018). This was further explored and
confirmed in research conducted by the Urban Institute. The report found that support
from the management level is necessary to ensure the success of these partnerships (Malik
et al., 2018a).
89
As this section has shown, the initiation of these partnerships cannot solely or mostly
be attributed to organizations or states. Various actors initiate and reach out. However,
traditionalrefugeeprotectionactorssuchasorganizationsandstatestakeonentrepreneurial
andgate-keepingrolesgiventheycontrolandcreateprocurementprocesses, accesstobenefi-
ciaries, andpolicyandlegalframeworks. Itisespeciallychallengingforsmallerandmedium-
sized companies to be active partnership initiaters and entrepreneurs. On a more granular
level,thesepartnershipsareofteninitiatedontheindividuallevel: whenthepeopleincharge
meet and discuss cooperation.
4.5.3 Partnership Dealbreakers: Non-Partnerships
Theprevioussectionhasalreadyilluminatedthatinitiatingthesepartnershipsiscomplicated
and often controlled by organizations and state actors. Going beyond the above discussed
hurdles, the interviews also exposed partnership dealbreakers as well as partnership failures.
The main themes emerging from the data are 1) mistrust of and bias against the business
sector, 2) reputation, 3) type of industry, and 3) visibility. While these three themes are
related and intertwined, they are not necessarily the same as a review of the evidence will
show.
Mistrust and bias
Mistrust of and bias against the business sector was one of the strongest themes re-
occurring throughout all interviews. Study participants from the business world blatantly
expressed their frustrations for being perceived as not trustworthy and having questionable
intentions. Inoneinterview, arepresentativesaidthatpeoplehavecalledthebusinesssector
“scorpios or rats,” referring to its untrustworthy and unreliable character. Another study
participant expanded and explained that
there is a lot of doubt... as we are a private company. We are working with
investor capital and revenues, so there is a lot of people who think that an actor
from the private sector is actually there trying to make a profit. And that [the
90
private sector] is trying to do something that is wicked. That is something that
is not acceptable, that is not humanitarian.
State and organizational study participants admitted to being deeply mistrusting of and
biased against the private for-profit sector. One interviewee from the humanitarian sector
stated that they wanted a partner that does not just want “an economic benefit,” and for
whom refugees “[are] not just a CSR piece.” He admitted that “to be frank, within the
nonprofitsector, thereisahugebiasagainsttheprivatesector. It’senormous, anditismore
in the humanitarian space than in the development space.”
Anotherparticipantfurtherunderlinedthatthereisstillalotof“pushbackaboutworking
with the private sector because people [don’t] trust it.” Another interviewee from a big IGO
added that there is always the risk and fear that businesses could exploit forced migrants
“for unqualified, low-paid jobs” and that states and organizations want to avoid that and
protect these vulnerable groups.
However, it also became clear throughout the interviews, that state and non-profit actors
do understand the value added from the for-profit sector. There seemed to be this dilemma
of state and organizational actors being skeptical and mistrusting yet acknowledging that
the private sector is part of the solution. For example, in the same vein one representative
from an IGO expressed mistrust about private sector partnerships, she also added that the
organization “should be partnering with the private sector because it is an invaluable sector
insocietyand[...] partofthesolution.”AnotherintervieweefromabigNGOsimilarlyvoiced
his doubts about the private sector, yet called out the overall non-profit sector for needing
to get “off its high horse [...] because we can massively benefit” from business engagement
in the humanitarian sector.
Nevertheless, the prevailing negative attitudes towards the participation of businesses in
refugeeprotectioncomplicates, ifnothalts, partnershipcreationprocesses. Thisisespecially
pronounced when it comes to partnerships where the business sector does not just take on
a passive donor/funder role but instead is an active implementation partner. This aligns
91
with the above discussed motivations of non-profit and state actors for choosing to partner
up with businesses: while times have been changing towards accepting the private for-profit
sector as an active, eye-level, participant in these partnerships and programs, the norm of
the business sector as a more than a funding source is only slowly adjusting.
Type of Industry/Sector
While mistrust and bias is problematic, an additional factor further complicates the
chances of forming partnerships in the first place: type of industry/sector. Being part of
the “wrong” type of industry not only diminishes chances but can be a dealbreaker for a
partnership. Bigorganizationshaveacriteriacatalogueaswellasalistofindustriesthatare
excluded from ever becoming a partner. For example, the various UN agencies have varying
exclusionary criteria. About 61 percent of UN agencies completely rule out businesses in
the tobacco industry (Martens and Seitz, 2019, pg.9). In their Corporate Code of Conduct,
the UNHCR explicitly states that their “minimum criteria for co-operation,” which excludes
companies that are involved in “weapons sales or manufacture [...], systemic and sustained
forced labour or child labour [...]” (UNHCR, 2016). Even more explicit (and public) is
the the UNDP when it comes to non-eligibility for certain sectors: the UN entity will not
partner with companies engaged in weapons, tobacco, pornography, gambling, or human
rights abuses/child labour (UNDP, 2013).
Interviewees also mentioned this no-go list during interviews. A representative of a big
IGO said that they have to be careful and consider the industry of the partner because if
they “go out and do this project together where there will be brand acknowledgment, you
don’t want it to look like endorsement.” Another representative from an IGO confirmed this
and said “partnering in [a] programmatic role [does] not happen if [the potential partner]
made weapons, tobacco, pornography [...].”
It is sensible that type of industry plays a central role and that public actors rule out
whole industries given the risk involved when working with certain sectors. Type of industry
is therefore an important factor for non-partnerships.
92
Reputation
Besides type of industry/sector, the reputation of a partner can also be a dealbreaker.
Reputation was mainly expressed in terms of moral/ethical reputation and respect. Put
differently, actors pay attention to their potential partner’s human rights record and ethi-
cal/moral practices.
Interviewees agreed that a main factor in partnership creation is minimizing potential
“reputational risks” stemming from partnerships. Reputational risks are assessed, and if it
is too “much too handle,” partnerships are avoided. Organizations as well as businesses said
that they pay “particular attention to sanctions.” Another business study participant said
that he “draw[s] lines at some places ...[and] d[oesn’t] take Saudi money because of their
human rights record.” He continued saying “we get requests from Saudi groups all the time,
and I basically say thanks but no thanks.” Just as businesses pay attention to sanctions, so
do organizations. The UNHCR and UNDP do not partner with actors in UN sanctioned
countries (UNDP, 2013; UNHCR, 2016).
All actors mentioned background checks and in-depth screenings to avoid reputational
damage stemming from partnerships. The business sector devotes time and effort to eval-
uating the risks associated with partnering up with different organizations or governmental
groups. Given that the businesses are fairly new to the humanitarian space, it is especially
important for them to not partner with a risky actor. Representatives from two big com-
panies confirmed this and stated that they have teams or outside firms do the background
checks for them.
While all actors indicated vetting their potential partners, these screening processes are
especially challenging for the business sector. Put differently, it is quite difficult to pass
the reputational risk threshold set by states and organizations. This is related to the above
discussed bias and negative attitudes towards the private sector. Often, businesses have
to go through a far-more detailed screening process. States and organizations do this to
93
ensure that their potential private for-profit partner does not engage in any questionable or
unethical business practices.
For example, an interviewee from a big IGO remembered that one time, a partnership
with a company did not happen “because of how they treated pregnant women in the work-
place. They were sued for it, so until [we] see the company makes a change about it, [we]
are not going to work with them [...].” Another organizational representative confirmed that
“organizations look at who the company is and their record [...] and there are firms that
look at the company for you; we call them screenings.”
This again, is reflected in organizations’ code of conduct or partnership guidelines. To
continuewiththeUNexample,theUNHCRstatesthatitmay“notengagewithanycompany
whose public image is severely compromised by past activity [...] which may be deemed, at
any given time, to reflect negatively on the agency” (UNHCR, 2016). Similarly, the UNDP
“should not partner with companies that are exposed to high levels of controversy, i.e.
companies that are systematically criticized (or face legal cases) for the way in which they
manage labor, community, environmental, governance, product related or ownership and
management issues” (UNDP, 2016).
Visibility
Related, yet slightly different to reputation, is the theme of visibility. While the reputa-
tion of a potential partner is actively observed and checked, visibility describes the absence
of being noticed or observed in the first place. Put differently, it matters whether and to
what extent actors are well-known or known at all. Visibility may be a function of the size
of the actor, what actor they are, and how long they have been active in the humanitarian
field. Unsurprisingly, the evidence suggests that smaller to medium-sized businesses that are
new or trying to break into the refugee protection space are often at a disadvantage.
An interviewee from a big IGO stressed that it may be easier for the “Google’s of this
world”toformpartnershipsgiventheirvisibility,howeverthatincludingsmallerandmedium
sized businesses is essential because “if you think about where [most people] work if they
94
are in the business sector, it is not a fortune 500 company [...] not everybody can work for
Starbucks.” Thus acknowledging and integrating smaller, not as well-known actors, more
actively in the refugee protection space is imperative.
A representative from a big company admitted that it is way easier for him to “establish
contact with [any actor] given our big company stamp. [...] as a global player, we have
a certain reputation that opens doors. This would be very different if we were a smaller
business that isn’t as well-known as we are.”
Other study participants concurred and added that it is helpful for less visible and well-
known actors to connect to bigger, more well-know actors, to become more visible and thus
gain access to partnerships. For example, a governmental interviewee admitted that it is
hard for them to know who is even out there. She added that there are so many actors
who do incredible work, yet who have a difficult time joining partnerships because they are
not well-known. The participant said that their office tries to do a lot of research into new
actors in the refugee protection space, and that the office tries to take on a bridge function
to “refer [interested actors] to other [actors] and tell them that they could partner up.”
Another interviewee from a smaller business recalled a particular project, where their
company was only able to form the partnership and participate in the project “through this
other [bigger] organization” which ensured that “[our business was] connected to the right
people.” She said that “for us as a smaller [actor] that doesn’t have much name recognition,
we would have not been able to get on that project.” She expected that “as we grow and
get more name recognition and as we have more visibility, it’ll be easier for us.”
Once actors have passed these initial tests and thresholds for being even considered as a
partnerinrefugeeprotection,thepartnershipnegotiationprocessbegins. Itisalong,taxing,
and costly process, which will be explored in the next section.
95
4.5.4 Partnership Negotiation Processes
Once actors have identified potential partners and have established contact, the real work
begins: negotiating the partnership and the respective partnership projects/programs. This
process is full of hurdles and set-backs; many partnerships fail during this process. This
section reviews the factors as well as hurdles potential partners face when finalizing partner-
ships. The data suggests that 1) partnership establishment is long-winded and costly, and
2) there need to be shared values and goals and well as shared risks and rewards.
Long and Costly Process
All interviewees described the partnership negotiation process as time-intensive and
costly. This is even more pronounced when different actors with different backgrounds from
different fields try to work together. Especially the clashing world views of more traditional
actors (IGOs, NGOs, states) and the for-profit business sector often present a problem.
These differing world views are reflected in varying levels of experience when it comes to
working in the humanitarian field as well as divergent bureaucratic and working cultures of
public and private actors.
For example, in a public discussion at the Urban Institute, where representatives from
the business sector and humanitarian organizations came together to discuss public-private
partnerships and the role of the business sector in refuge protection, one speaker stressed
that the partnership formation process is tricky because of “two worlds coming together”
(Urban Institute, 2018). He continued to say that not only do these sectors look at things
differently, buttimelinesaredifferent, andthatthereislittleexperienceinworkingtogether.
This point was confirmed in some of the interviews.
One study participant described the partnership process between public and private ac-
tors as a “big, complicated Venn diagram of [differing] ideologies, lack of understanding, and
lackofexperienceoroversimplificationoftheprivatesector.”Publicandprivatesectoroften
use very different language so agreeing and being “able to speak[ing] the language of each
96
other” is important because otherwise partners “can be both talking with the same words
and referring to different things.”
Another interviewee from a big company described a situation where his team and the
team of their humanitarian partner completely misunderstood one another. He laughed and
said “we had to completely switch our language to make sure that they understood us, and
they had to adjust their language so we would understand them. So we all had to first agree
on a common language and find a common denominator for the cooperation.”
There were also clear differences between humanitarian and state actor frustrations and
the frustrations expressed by the business sector when it comes to the negotiation process.
Humanitarian organizations and state actors highlighted the occasional inability of the pri-
vatesectortounderstandwhatisandisnotpossibleinthehumanitarianspace. Forexample,
one interviewee expressed that often times potential partners are “dreaming,” asking for and
expecting things that are simply “not applicable.” Examples included things that are not
doable or appropriate in host country culture or require high governmental involvement or
approval. This complicates the negotiation process because it is time-intensive for all part-
ners to see eye to eye on what is and is not possible and to then develop a respective project
or program.
On the other hand, grievances expressed by the business side mainly revolved around
the inefficient, long-winded partnership negotiation process. This speaks to the point of
different bureaucratic cultures and time horizons in the humanitarian, governmental, and
business sector. One representative from a medium-sized business said that the dragged
out partnership and program negotiation process was frustrating because his business is not
interestedinprogramsthat“wouldbematerializingin1.5years.”Hesaidhisbusinessfocuses
on “immediate access and impact” and “[does not] have time for blablaba [...], absolutely no
time for BS.” Another interviewee from a big private company echoed this sentiment saying
that the
97
downsides to [the humanitarian sector] is bureaucracy. [...] we think of ourselves
asadisrupter, andsometimeswhenweareworkingwithreallybigorganizations,
we get sick of the bureaucracy. We don’t want to spend 2 years negotiating a
partnership agreement, we want to [help] people tomorrow. And we are very
aggressive about that, and that doesn’t always work in the humanitarian sector,
and it pisses people off sometimes.
Similarly, another business representative said “if people keep on talking and talking and
talkinganddonothing. Yea,after3monthsIwillstopspendingmytimeandenergybecause
I believe in execution.” He added that working with “the non-profit [sector], you will find
people talk more than they work.” In a public conference, a representative from one of the
biggestNGOsadmittedthat“onthehumanitarianside, wearenotstructuredtoactuallybe
able to move forward with these partnerships [...] we’re are moving at a much slower pace”
(Urban Institute, 2018). The take-away point is that there needs to be a more meaningful
internal restructuring of humanitarian organizations to “absorb these partnerships to make
them meaningful. This takes a really clear understanding of the difference in approaches
[...] and what business process in the humanitarian space we can potentially circumvent or
accelerate to kind of move at that pace” (Urban Institute, 2018).
What stood out during the empirical investigation was the fact that although businesses
mayhavelessexperienceandknow-howinthehumanitarianspace,theyactuallycompensate
by having very efficient processes and well-trained teams devoted to humanitarian projects.
The empirical evidence suggested that the business sector takes the development and main-
tenanceofthesehumanitarianpartnershipsandinitiativesveryseriously,thusoftendevoting
a considerable amount of physical and financial resources. Put differently, it is often not just
a side-project or a passive engagement for these businesses.
Forexample,intheabovementionedconferencetherepresentativeoftheNGOaddressed
thattheirNGO,althoughbeingoneofthebiggestonesinthefield, wasnotaswell-staffedas
their private sector partner. He stated that the NGO “[wasn’t] in a position where they had
a full time McKinsey team of 5 staffers who [were] all spending 23.5 hours a day moving this
project forward.” A study participant from a big IGO confirmed this and added that these
98
partnership negotiations and program creation “shouldn’t be something that [employees] do
on the side of their desk. ‘Oh I’m the head of communications, and now I’m doing this
partnership.’ You need people who are experts in partnership design and execution.”
Similarly, the representative of a big company in an interview stressed how important
it was to work with an organization where there was at least one person, or better, a team
devoted to the partnership and partnership projects. The interviewee said that partnership
negotiation and maintenance is “fairly labor intensive on both sides. So on both sides
capacity ends up usually being the biggest issue” and “requires a big team to manage that.”
Another interviewee from the business world concurred and said that he noticed a severe
capacityprobleminhumanitarianorevengovernmentalgroups, whichdelaystimetablesand
the project being moved forward.
Going beyond the different time horizons, bureaucratic structures, and experiences, an-
otherpartofthepartnershipandprogramnegotiationprocessislearningaboutsharedvalues
and goals as well as shared risks and rewards. While having overlap in these areas is part
of identifying a partner in the first place, actors mainly discover their specific alignment in
valuesandgoalsmuchlateronwhentheyarealreadyinthepartnershipnegotiationprocess.
Shared Values & Goals, Shared Risks & Rewards
In interviews, almost all study participants named shared values and goals as a main
component of negotiating the partnership and the respective joint programs. Interviewees
said that it takes a while to really identify overlap and to see what is and is not possible in
the partnership. It becomes “a priority to understand their goals” as one interviewee put it.
Mostly these partnerships have multiple goals - some of them are mutual and some of
them are individual goals. An interviewee stated that “everyone involved in the partnership
needs to understand there is the goal of the partnership but then there are the goals for each
of the participants that they have to extract from the partnerships as well.” However it was
important that all of these goals were “shared openly and transparently with everybody”
because otherwise “you lose trust”.
99
In addition to shared values and goals, the negotiation process involves the weighing of
shared risks and shared rewards. The concept of mutual benefit/gains is sensible and was
confirmed throughout the interviews. One interviewee stated that in the negotiation process
the parties need to “make sure [all] have skin in the game. There has to be a benefit to each,
and it is ok for the private sector to get a benefit from it.” Another representative from the
business community agreed saying that the “core of all partnerships” is that it is “mutually
beneficial, [that] there is mutual value. [...] It has to make sense for [all] sides.”
However, beyond mutual gains, partners also need to identify, talk about, and agree on
shared risks of the partnership and the respective, joint programs. One interviewee empha-
sized that “shared risks, shared rewards is a core principle in partnerships.” While being
transparent and agreeing on shared rewards is already challenging, it is even more difficult
for shared risks. A study participant admitted that this can result in heated discussions
and disagreements amongst partners. She remembered a time when her team and a private
partner disagreed on sharing risks: “[the private partner] came up with a little device [...]
however, I had to say to the R&D people [of the private partner], when it comes to liability
aroundthedevice,youdevelopedit. Thatcan’tbeasharedrisk[...] andtheywereinsulted.”
Unsurprisingly, given the overall strenuous process of setting up and negotiating the
partnership and joint projects/programs, actors overall prefer a longer-term partnership.
One interviewee said that they prefer “stable and valuable partners over the long-term”
since it is a huge “sunk cost” if the partnership is only “for 6 months or 1 year.” Another
interviewee echoed this and said that because “the front end work is a lot” it is preferred to
have longer partnerships.
Asthissectionisshown,thepartnershipprocessdoesnotjustendatidentifyingpartners,
norisitastraightforward,quickprocesstonegotiateandfinalizethepartnership. Itinvolves
discussions and disagreements along the way. It can be long-winded and frustrating before
100
a partnership is finalized and projects/programs are decided upon. However, what happens
once partners have settled and carried out a project or program together? The next section
will review the evidence of how partners evaluate these joint initiatives.
4.5.5 Program Monitoring and Evaluation
The empirical data also gave insights into how actors approach monitoring and evaluating
joint projects and programs. Two main take-aways stand out. First, all actors claim that
theyhave a rigorousmonitoringand evaluation system. However, the conversations revealed
that there is often disagreements on what constitutes success, how to evaluate, and what
should be reported. Consequently, it is common to have several reports: one joint report as
well as individual actor reports.
Second and related to the first point, especially the business sector exposed the rather
inefficient, non-rigorous monitoring and evaluation processes of the humanitarian sector.
Given the rather advanced, data-driven methodology of the for-private sector, there is huge
potential for the humanitarian and state sector to learn and benefit from the industry’s
monitoring and evaluation approaches.
Whenaskedtodescribethemonitoringandevaluationprocessofthejointprojects/programs,
interviewees detailed the different steps. While monitoring and evaluation of projects and
programsvariessignificantlydependingonpartnershipandproject,allintervieweesindicated
that they followed a rigorous process. Study participants often mentioned a combination of
quantitativetoolsincludingstatisticsandregressionaswellasqualitativeapproachesinclud-
ing beneficiary interviews and focus groups. Often times progress is measured throughout
rather than just at the conclusion of projects and programs. This gives actors the flexibility
to adjust and update if necessary.
Most interviewees stated that these partnership programs often result in a number of
differentreports. This canbea function ofthe variousactors catering totheir specific target
audiences (stakeholders, boards, funders etc). It is therefore sensible that these different
101
public and private actors produce very different reports based on what they want to commu-
nicate to the public. When asked about the output reports, one interviewee said that “each
partner completely does their own thing, and then it is pulled together and reported out.”
However, another reason for the production of several reports goes beyond individual re-
portingpreferencesandneeds. Itisnotuncommonthatpartnersdisagreeonhowtomeasure
success, how it should be evaluated, and what is reported eventually. One interviewee from
a big IGO emphasizes that “everybody thinks evidence is clear, but it is not. It is open to
interpretation.” She remembered that in one project
[the partners] did the research together, and when it came time to write the
report, they couldn’t agree. They had collected all the evidence [..] and they had
to have a third party come in to write the report and get them to agree because
even though they both agree good stuff is happening, one wanted to say it this
way, the other wanted to say it that way. And they couldn’t agree, so it all got
stymied.
Some of the above described disagreements on what to report may also stem from the
frustrations expressed by the business sector. In particular, in interviews, representatives
from the business sector criticized the humanitarian sector especially for their non-rigorous
monitoring and evaluation approaches.
One interviewee said that success was difficult “to quantify [...] from the business stand-
point,yeathatisdrawbackbutthatishowlifeworksinthenon-profitspace.”Hecomplained
that his partners often “give some kind of statistics [...] but have nothing solid” or rigorous.
Another interviewee from a big company admitted that they were “a very demanding part-
ner” whenever it comes to the methodology and data they want to see. She continued to
say that for example, her team stays on top and checks progress every day using “internal
dashboards” and “advanced metrics,” however that she had learned not to expect the same
from their humanitarian partners as they may only report “on a quarterly basis.”
Some other interviewees were even more direct in their frustrations. One was shocked
to see how programs and projects were monitored, evaluated, and what was reported. He
102
said that some of these programs do not deliver what they promise but what matters is that
“they get a bunch of money, they actually deliver nothing, and there are no consequences
[...] So I’m not sure that there is much concern about the efficacy of how the money is being
spent, just that it is being spent.” He continued that there is often no follow-up on these
projects or programs - not even for those that promise sustainable impact. Instead
theydon’tmeasureROI,becausetheresultswouldbeabysmal. Buttheygoback
to these villages where they taught 100 women to sew and the sewing machine
broke down after the trainer left. And they go to these villages and teach 100
people to cut hair, and they go back and the hair clippers are dull and nobody
knows how to sharpen them. So you know they throw money at the problem,
and they are able to tick a few boxes and you know [the managers] are able to
tell their managers: yea, you know we touched 1000 people in this village and a
1000 people in that village. But did they do anything lasting or impressionable
and worth-wile? No.
Another business interviewee said he has had many conversations with his non-profit
partners about “impact measurement frameworks” but that most of these “organizations
[...] have no idea what they are talking about.” Even a representative of a big non-profit
agreed and admitted that the humanitarian sector had “a quite strong background in the
pseudo-scientific measure, of which we do a lot of. [...] we are not very good at sharing
failure.”
Interviewees from the business sector were not only concerned about potentially deriving
incorrect conclusions on whether and/or how successful a project or program is, some of
them also expressed that the potential incorrect results may be harmful to asylum seekers
andrefugees. Oneintervieweesaidthat“thereisnowaythat[humanitarianorganizationsor
governmentalactors]couldgettheKPIsthattheyareadvertising”and“inmanycasesthese
programs not only fail, they cause harm.” He mentioned refugee entrepreneurship programs
as an example:
They throw literally billions at youth to try to teach them how to do a business
plan, a marketing plan, and then tell them to go out and do their own startup.
103
Theaverageageofasuccessfulentrepreneuris45. Andtheyarethrowingbillions
of dollars at these kids that have no idea... they have 0 job experience, they
don’t know how things work. They don’t know all the parts... marketing, HR,
finance. And it’s like they are bringing a knife to a gun fight. And then they
realize they can’t succeed, and that they are worse off than before with a failed
business, no money, and huge debt.
Although some of these sentiments may sound like the business sector may have unreal-
istic expectations or may not be aware that performance evaluation is very different in the
humanitarian space, this was by no means the case. All business participants understood
that measuring success of humanitarian projects and programs is vastly different and cannot
be compared to their daily business evaluations. However, participants were shocked that to
this day, most of their humanitarian partners - whether organizations or governments - had
developed very few, rigours evaluation tools that are appropriate for these projects.
One representative of a big company said that he had to learn and now understands
that the traditional marketing tools and KPIs used in the business world “don’t fit or make
sense” for these projects. Nevertheless, it is possible to create a set of “social and/or soft
KPIs” that specifically capture “social skills and social factors” [KPIs]. He added that
becausetraditionalhumanitarianactorshaveyettodevelopthem,thebusinesssectorandbig
companies are now stepping in and coming up with more sound evaluation and performance
metrics that fit humanitarian projects and programs.
While the harsh criticism from the business sector is concerning, there is also a lot of
opportunity. Given the higher standards of the business sector when it comes to monitoring
and evaluation, these partnerships have the potential to remedy the shortcomings of the
humanitarian sector and produce overall more sophisticated methodologies to evaluate per-
formance. Most interviewees were hopeful. Humanitarian and state actors were very excited
about working with the R&D teams of the private sector. One interviewee was fascinated by
how sophisticated not only the products and programs developed together with the business
sector were, but also how advanced the measurements were. She indicated that working
104
together brings new avenues for methodologies that were unthinkable or unreachable before.
This also allows to update and upgrade current approaches of impact measurement.
Sherememberedarecentprojectwhereshedescribedthatwithoutthemeasurementtools
of their private sector, the NGO probably would have needed to resort to interviews and
focus groups to understand the impact. However, because of the involvement of the private
sector and their sophisticated R&D team, they were able to more directly and accurately
observe and measure the impact of the project. She said “the R&D people were stunning,
they came up with a little device [...] that would track all day how and when [the product]
wasused; forwhatpurposetheproductwasused.”Thedevicewasintegratedintheproduct
so it could track movements and motions of people who were using it to then record for
what people used it. This data was then combined with some interviews, overall resulting
in a holistic methodology of measuring impact. In addition, the teams were able to monitor
the progress of the project on a daily basis without necessarily disturbing the beneficiaries,
thus giving them the chance to adapt and change and detect errors early on (instead of only
having sporadic feedback).
Another interviewee from the business sector agreed and added that while upgrading the
currentstandardofperformancemeasurementistaxing, however, thatitis“importanttobe
patient,andhopefullywe’llbeabletoopennewdoorstogetherandpioneerthehumanitarian
impact measurement field.”
This section has uncovered and discussed the most common themes throughout the em-
pirical data. It detailed the motivations, decisions, considerations, and frustrations that
actors experience in the partnership selection and formation process as well as during the
monitoring and evaluation phase. The subsequent section will synthesize the empirical re-
sults and connect them to the theoretical expectations.
105
4.6 Empirical Discussion
While it may appear that the partnership process is straight forward with like-minded ac-
tors coming together and becoming partners, this chapter has exposed how convoluted this
process is in reality. This is especially relevant in the refugee protection space as we wit-
ness this new multi-stakeholder network emerge. The empirical data aligns with many of
the hypotheses introduced in this chapter. However, some of the findings go beyond these
hypotheses and add nuance and new revelations.
Motivations and Incentives: Good vs. Bad
While not hypothesized, the empirical data reveals the different motivations and incen-
tivesofstates,organizations,andthebusinesssectortopartnerintherefugeeprotectionand
empowerment space. There seems to be a clear pattern of what is deemed an “appropriate”
vs. “inappropriate” reason or motivation to get involved in refugee protection and empow-
erment or the desire to form partnerships in this space. This black-and-white thinking is
prohibitive and affects the process of partnership formation and certain actors’ ability to
join the network. For example, humanitarian organizations and state actors seemed to be
perceived and portrayed as the ones with the pure/good intentions for wanting to partner
with the profit sector due to resource constraints and desire to develop innovative solutions,
while the business sector, with its potential market/business motives, is located on the other
end of the spectrum (bad intentions).
However,thedatarevealedthatmotivationsofthedifferentactorsarenoteasilyclassified
inbinarytermsofgoodorbad,butbetterasameltingofvariousreasons. Especiallythemo-
tivations of the business sector are diverse and a combination of genuine social responsibility
aswellasbusinessopportunity. Whilehumanitarianandstateactorsoftenquestionthegen-
uineness of businesses for wanting to do good, the empirical data suggested that the payoff
and benefits for businesses are often ambiguous and uncertain (especially in the beginning).
Many businesses get engaged ad hoc during emergency situations. An example presents the
business engagement during the humanitarian crisis in 2015 when millions of asylum seekers
106
and refugees arrived in Europe, which was mainly a result of feeling a responsibility rather
than seeing an opportunity for immediate or even long-term benefit/profit.
Simultaneously, the intentions of state or organizational actors can also not be classified
as purely “humanitarian.” Non-profits have performance pressures, stakeholders, advisor
boards etc.; they are often structured in a similar manner as businesses; and they cannot
freely distribute resources or create programs/projects where they are most needed. The
empirical data illustrates high overhead and organizational costs very similar to business
operations. Partneringwiththefor-profitsectorbecomesthusnotjustabouthelpingrefugees
but also financing the functionality of these non-profits.
Similarly,becausestateactorshavetheresponsibilitytograntasylumandprotectrefugees,
it is often assumed that their motivations are good. However, states often fall short of this
responsibility and commonly outsource support and protection services. Their motivations
to partner with for-profit actors is thus not only due to lack of capacity but also due to a
lack of willingness.
Keepingthisinmind, thischapterencouragesreaderstotakeamorenuancedviewofthe
various incentives and motivations of organizations, states, and businesses to get engaged in
refugee support and protection. Stepping away from this above-described binary thinking,
enables us to better understand partnership formation processes.
Review of Hypotheses and Take-Aways
Turning to the specific hypotheses, the empirical data partly confirms that organizations
and states are often the main entrepreneurs of these partnerships. Various actors initiate
and reach out; the business sector is indeed very active and motivated to break into this
refugee protection and empowerment network. Thus the initiation process cannot solely or
mostly be attributed to public actors.
107
However, state actors and organizations hold the majority of control in the initiation
stages. As they have access to asylum seekers and refugees, they can restrict who gains
and does not gain access to the network. Thus, these traditional refugee protection ac-
tors function as gatekeepers and often as main entrepreneurs. Businesses also rely heavily
on humanitarian organizations and states to navigate the complicated humanitarian policy
world and cultural hurdles that arise when working with vulnerable populations. Given that
states and organizations have historically been the ones directly working and engaging with
refugees, businesses need the expertise and experience of these actors.
Althoughnothypothesized,theempiricaldatarevealedanotherimportantdriverofpart-
nershipinitiation: theimportanceofupperlevelCEOormanagementcommitmenttorefugee
protection and support. Often, the grounds for partnerships are laid out by CEOs or man-
agers meeting at conferences or talks and then spread from top down throughout the entire
company, organization, or governmental agency. This means that there needs to be sys-
temic, cultural commitment for the business sector to become engaged as well as to start
these partnership processes. Only once this is in place, teams are created or trained to take
on refugee protection partnerships and projects.
When it comes to gaining access to the network and partnership dealbreakers, the results
align with the hypothesis on the centrality of reputation. Actors devote a lot of time and
effort to screening their potential partners and regularly turn down those with a question-
able moral/ethical reputation. All actors are affected by this: humanitarian organizations,
governmental groups or governments overall, and businesses. Nevertheless, it is often par-
ticularly difficult for businesses to meet the standards set by the traditional humanitarian
actors. This is often rooted in the sustained deep mistrust in the business sector.
It is also reflected in the fact that humanitarian and state actors rule out whole business
sectors from the get-go. Consequently, certain sectors are precluded from ever joining the
refugee protection space. This again aligns with my theoretical expectation. It is sensible
that actors focus on reputation and types of industries to rule out partners. Any actor that
108
engagesinunethicalpracticesorthatispartofanindustrythatcouldcausedirectorindirect
harm to forced migrants should be at a minimum restricted from participating or accessing
asylum seekers or refugees.
Additionally,theresultspointtowardsthepowerfulinfluenceofbeingavisibleactor. This
is a new revelation that was not part of the theoretical expectations. Moreover, visibility
emerged as a distinct theme from reputation. While actors may have an intact reputation,
they may still be unable to form partnerships given they are not as visible or well-known
as other actors. This is a problem for smaller actors in particular. Especially smaller or
medium sized businesses who are new to (or potentially not yet part) of the network are
affected by visibility issues. A potential remedy for less visible actors is to connect to some
bigger actors or to actors who take on coordination and bridge functions. For example, the
TentPartnershipforRefugeesisabusinesscoalitionthatspecializesinconnectingbusinesses
with like-minded actors.
I also hypothesized about the partnership negotiation process. First, the empirical data
illuminated how long-winded and costly the partnership negotiation process is. Various ac-
tors experienced the partnership negotiation differently and had thus differing frustrations
with the process. On one hand, humanitarian organizations and state actors conveyed that
the inexperience and lack of local/cultural sensitivity of the private for-profit sector slows
down negotiation and program creation. On the other hand, businesses criticized the ineffi-
cient and unpragmatic procurement and partnership process.
As expected, finding and sharing similar values is integral in the negotiation process
as it sets the stage for subsequent goal setting and program/project creation. Though,
finding this overlap in values is not as easy as it appears. While actors may have superficial
overlap, identifying how these shared values can be solidified and serve as the foundation of
a partnership is challenging.
109
Second, partnership negotiation heavily involves the review of shared risks and shared
rewards as hypothesized. It is sensible that all actors should and need to gain from a
partnership given the setup process is costly and time-consuming. Similarly, actors also
discuss areas of shared risks of the partnership and respective programs/projects. All actors
aim to minimize individual and shared risks when negotiating roles and responsibilities.
To successfully negotiate shared risks and rewards, all actors need to clearly lay out and
communicate their expectations. This includes being fully transparent about what actors
aim to gain individually and together, and how risks are distributed across all parties. Only
if reward and risk expectations are lucid can actors trust one another and avoid or mitigate
tensions/problems arising at later stages of the projects/programs.
Finally, the empirical data also shed light onto project/program evaluation processes.
I expected to see disagreements amongst actors on how to evaluate, report, and classify
success. I indeed found the program/project performance process to be contested amongst
partners. This illuminates that evidence and data is subject to interpretation, and that
interpretation seems to be even more complicated when partners come from very different
sectors (humanitarian, business, state etc). As expected, actors produce multiple reports to
cater to their specific stakeholders and to highlight what they find most important and most
successful.
While it is sensible that evaluating and measuring the success of humanitarian pro-
grams/projectsisdifficultgivenitinvolvesmanyimmeasurableorhard-to-measurevariables,
this was not necessarily the core of the disagreements. Instead, the business sector was very
honestabouttheshortcomingsoftraditionalhumanitarianactors’evaluationpractices. This
finding goes beyond my theoretical expectations and problematizes overall impact measure-
ment frameworks in the humanitarian space.
Scholars have theorized and explored the effectiveness and accountability of NGOs and
IGOs and identified mixed results (Ebrahim, 2003; Reimann, 2005). More recently, human-
itarian organizations have been a focal point of criticism for being rather untransparent and
110
ambiguous about how their funding is used, who benefits, and how scalable their solutions
are. The performance of governmental groups or governments in refugee protection has not
been explored in much detail.
My findings provide rare insights into the perceptions of and sentiments towards the
performance of collaborative humanitarian projects/programs. Given that the business sec-
tor comes with a different methodology and a razor-sharp focus on data, it was fascinating
to observe their opinions and reactions towards the performance measurements of various
humanitarian actors.
While these revelations seem dooming, they must not be. With the private for-profit
sector entering and actively participating in the refugee protection and empowerment space,
thereisanopportunitytoreviseandupgradecurrentperformanceandsuccessmeasurements.
As seen in the empirical examples, the private sector is already offering sophisticated data
tracking tools that compliment more traditional evaluation methods such as interviews and
focus groups. Further, combining the knowledge and background of the for-profit sector and
its strength in data, marketing, and R&D with the experience of the humanitarian sector
could lead to the development of even more sophisticated, appropriate KPIs for refugee
protection and support projects/programs. Table 4.3 briefly summarizes all findings.
111
Hypotheses Status Notes/Additional Findings
H1: Organizations and states are the main entrepreneurs of
partnerships with the private for-profit sector.
Partly
Confirmed
Both sides initiate;
business sector initiates
quite a bit;
however, organizations and
states are the main
entrepreneurs and gatekeepers
CEO and management level
initiation of partnerships
H2a: Actors are more likely to partner up with partners that
have a good reputation.
H2b: States and organizations are unlikely to partner up with
for-profit private actors that are part of harmful industries.
Confirmed
All actors conduct screenings
Especially harmful industries
are tobacco and alcohol
H3a: Private and public actors are more likely to partner
with one another if their values and goals align.
H3b: Private and public actors are more likely to partner
with one another if there are mutual gains and shared risks.
Confirmed
Core to partnership
negotiation; long process
H4a: The program performance evaluation is highly contested.
Actors often disagree on how to evaluate, report, and classify
success.
H4b: Actors tend to conduct internal evaluations and report
their own findings rather than the joint findings.
Confirmed
Disagreements as a result of both
individual actors’ reporting needs
as well as diverging methods and
R&D standards
Traditional actors criticized for
ambiguous evaluation
frameworks
Table 4.3: Summary of Findings
4.7 Conclusion
How and why do multi-stakeholder partnerships in the refugee protection and empowerment
space emerge; and how do partners evaluate and report the performance of joint projects?
This chapter set out to provide answers to these questions. Engaging in qualitative analysis
of primary interview data and reports, I was able to uncover the dynamics and processes of
initiating, negotiating, and evaluating these partnerships and their projects.
While this chapter has illuminated the taxing process of securing partnerships and eval-
uating joint projects/programs, it also offers paths for opportunities. First, bigger, more
reputable and visible actors should consider and actively engage with smaller, less visible
actors. This enables smaller and medium-sized actors to access the network and become
more visible, thus making the network overall more inclusive and diverse. As supporting
112
refugees and asylum seekers requires a holistic approach, it remains imperative to involve
and invite all sizes and types of business actors into the space.
Second, aligning the partnership construction processes and managing capacity will be
centralgoingforward. Inotherwords, publicandprivatesectorwillhavetoworktogetherto
optimize the rather inefficient processes and find a middle ground that works for all parties.
While well-developed partnerships are certainly essential, partnership negotiations that take
years are time-consuming, labor-intensive, costly, and counterproductive for humanitarian
purposes. Reformingthesecurrentprocessesgoeshandinhandwithbuildingcapacity. Only
if enough resources/staff is devoted to these partnerships and programs, can they progress
and unfold in an efficient and productive manner. A way to increase capacity, especially
in underfunded humanitarian organizations, is to provide capacity grants. Several business
actors have already stepped up and provide capacity grants to their partners.
Third, and as pointed out in the empirical discussion, the various shortcomings of the
humanitariansectorwithregardstoperformanceframeworksmayberemediedbypartnering
with the for-profit private sector. This does not mean that these projects or programs
suddenly will be more successful or that actors will start reporting failures. However, it
opens up avenues into more rigours testing and measurement that could help actors to
adjust and update projects and programs before they fail.
More research is needed to understand the implications of these newly emerging PPPs in
therefugeeprotectionspace. Forexample,itstillmaybetooearlytosystematicallyevaluate
theperformanceofthesecollaborateprojects/programs. However, asthesepartnerships and
the respective projects mature, there will be more public information available on their
impact, allowing scholars to observe and trace their progress over time.
Moreover, this research has hinted at the benefits arising from these PPPs, yet we know
verylittleaboutpotentialandactualharmandrisksemanatingfromtheseprojects/programs.
Collaborative programs with the private sector offer huge opportunity for achieving long-
term refugee integration. However, questions such as how scalable and systematic are these
113
solutions (how many beneficiaries are really reached and impacted)? To what extent are
these services and projects non-discriminatory (or put differently, who are and aren’t the
beneficiaries of these products/projects; who receives products/services and who doesn’t?)?
Future research should actively target these questions.
Finally, andrelatedtothepointsonperformanceandrisksistheincreasinglycentralrole
of refugee voices in the active construction of solutions. Until very recently, refugees them-
selves as well as refugee-led organizations had been peripheral in the policy-making process
and/orprogram/projectcreationstages. Theyhadbeenmostlyseenasparticipants,receiver
of benefits, and passive feedback givers during or after the conclusion of projects/programs.
This has been rapidly changing: refugee voices and refugee-led organizations are receiving
a seat at discussion tables and are finally considered a vital part of finding solutions. This
also surfaced during my research when especially the business sector highlighted their desire
to have direct access and inclusion of beneficiaries during project/program creation. Fu-
ture research should investigate the role of refugees and refugee-led organizations and their
relationships with the humanitarian and/or business sector.
This chapter has set out to unpack the formation and evaluation dynamics of refugee
protection and empowerment PPPs. It has shown what motivates the various actors, who
makes it into the network, what factors matter during the initiation and negotiation stages,
and how contested the reporting and evaluation processes are. These partnerships offer
avenues for updating and upgrading current approaches and give a glimmer of hope that
refugee protection and support is becoming more sustainable.
With a new, relentless crisis unravelling in the Ukraine, these multi-stakeholder part-
nerships are more important than ever before. As discussed in the introduction, public-
private partnerships have already contributed to delivering emergency funds and goods to
the Ukrainian people. With millions of Ukrainian refugees arriving in European host coun-
tries, it is likely that we will see even more diverse public-partnerships that offer short and
long-term services and programs arise in the near future.
114
Chapter 5
The Way Forward: The Future of Public-Private
Cooperation in Refugee Protection and Empowerment
This dissertation set out to investigate why and how businesses, states, and organizations
cooperate to tackle the escalating humanitarian and environmental crises across the world
that lead to millions of people fleeing their homes. Although this project considered the
motivations,factors,andoveralldynamicsthatareatplaywhenvariousactorsfindandform
partnerships, it purposefully centered its attention around the perspective of the corporate
sector. The reason for this is, as mentioned in the introduction, that the active role of
businessesinrefugeeprotectionandempowermenthasreceivedlimitedattentioninliterature
(Betts et al., 2017; Weiss, 2013).
The overarching argument of this dissertation is that while businesses are interested, mo-
tivated,andreadytobecomeactiveprogramandprojectcontributors,itremainsremarkably
difficult for them to break into the refugee protection and empowerment space. Finding and
forming partnerships as well as creating programs/projects is a taxing and costly process.
Certain factors such as geographic location, reputation, shared values/goals influence the
chances of successful partnership creation. The resulting public-private partnerships often
produce innovative solutions for refugee situations and have the potential to revolutionize
humanitarian program/project evaluation metrics.
115
The following sections will first review the dissertation chapters, and how they overlap
andcomplementoneanother. Thischapterthenaddressestheelephantintheroom: towhat
extent is corporate engagement in refugee protection and empowerment even good? This
is followed by the two final sections, which discuss the future of cooperation in the refugee
protection and empowerment space and lay out a research agenda beyond this dissertation
project.
5.1 Chapter Synopsis
The increasingly present and influential role of the private for-profit sector in world politics
is not an entirely new phenomenon. Scholars have analyzed and discussed the various ways
of why and how the business sector shapes rules/norms, policies as well as produces services
and goods (Abbott and Snidal, 2010; Avant et al., 2010; B¨ uthe and Mattli, 2011; Graz
and N¨ olke, 2008; Hall and Biersteker, 2002; Roger and Dauvergne, 2016; Ruggie, 2014).
However, the involvement of the business sector as an active rather than passive participant
in creating and maintaining projects and programs for refugee protection and empowerment
isarathernewdevelopment. Itwasmostlyspurredthroughtheunfoldingofthedetrimental
humanitarian crisis in the Middle East (Syria), when millions of asylum seekers and refugees
arrived in Europe.
However, why would businesses choose to get actively involved in the first place? Given
businesses’ limited experience and exposure, becoming involved in refugee protection and
empowerment is complicated and costly. So why would they choose this cause compared to
any other environmental or social cause? Business and marketing literature has speculated
aboutthedriversforcorporatecommitmentinsocialissues. Motivationsforcorporatesocial
responsibility (CSR) and activism range from normative desire (businesses genuinely want
to do good), to accessing new markets, improving reputation and the public image.
116
While it is likely that multiple factors play a role, one of the most common ones is
stakeholder pressure (i.e. pressure from customers and consumers) on firms to act socially
and environmentally responsible. Given that this pressure is connected to brand loyalty,
public image, reputation, and buying intentions (thus profits), scholars find it likely that
it incentivizes businesses to become socially and environmentally engaged (Erdem et al.,
2018; Koschate-Fischer et al., 2012; Robinson et al., 2012; Sen and Bhattacharya, 2001;
Wagner et al., 2009). However, is this pressure also present for the refugee cause? In other
words, howdoesthepublicthinkaboutbusinessesgettinginvolvedinrefugeeprotectionand
empowerment, and how would they want corporations to support refugees?
Chapter2setsouttoinvestigatethesequestions. Ittestswhetherstakeholderpressurefor
business engagement in the refugee cause is actually present and could serve as a motivating
factor for increased business engagement. It argues that consumers and customers welcome
CSR activities to protect and empower refugees and that certain factors including gender,
political ideology as well as peoples’ awareness of social/environmental issues and skepticism
about corporate motives influence levels of support for refugee protection and empowerment
CSR.
Through a survey experiment, the chapter first compares levels of support from the US
public across three different issue areas: climate change, the BLM movement, and refugee
protection and empowerment. It finds that the level of public support for CSR activities in
refugee protection and empowerment is almost as high as that for CSR activities in climate
change (they are so similar that a difference in support between refugee CSR and climate
CSR are statistically not significant).
Thechapterthenzoomsintoparticipants’opinionsandattitudestowardsCSRforrefugee
protection and empowerment as well as the factors that influence levels of support for corpo-
rations supporting and empowering refugees. Maybe unsurprisingly, women, liberal partici-
pants, and participants with high personal social responsibility and awareness, on average,
117
have the highest levels of support for CSR in refugee protection and empowerment. Ad-
ditionally, the more skeptical people are about companies’ motives/intentions as to why
businesses get involved in refugee protection the less people support CSR for refugee pro-
tection. Finally, the chapter inquires about peoples’ preferences on the type of support
companies should provide to refugees and finds that Americans would like to see companies
actively train and hire refugees as well as donate food and clothes.
Chapter 2 serves as the foundation of corporate engagement in refugee protection and
empowerment. Given the high levels of support from the public, stakeholder pressure is
likely present and could incentivize businesses to get involved in the refugee cause. However,
when businesses decide to protect and empower refugees, they often do not develop these
programs and projects on their own. Instead, they partner with experienced actors such as
governments and organizations to plan, develop, and execute these projects and programs.
This is where chapter 3 and 4 come in: they focus on the emergence of a novel public-private
partnership network in refugee protection and empowerment (chapter 3) and the decision-
making processes behind finding and forming partnerships as well as challenges encountered
along the way (chapter 4).
Chapter 3 presents a purely quantitative analysis of the burgeoning public-private net-
work in refugee protection and empowerment. Utilizing social network analysis and expo-
nential random graph models (ERGMs), it explores what factors influence tie (partnership)
formation, and who partners up with whom. It contends that being a certain type of actor
and being connected to a central actor such as the UNHCR positively influences the like-
lihood of partnership formation. It also expects actors with similar attributes to be more
likely to form connections with one another rather than with unlike actors. The results
confirm that for-profit private actors are indeed less likely to form partnerships compared to
other types of actors. Further, being connected to the UNHCR is positively correlated with
forming ties. There is also evidence that alike actors are more likely to form partnerships
with one another (for example, actors in the same geographic region more readily form ties).
118
This chapter offers an unprecedented look into the growing public-private refugee pro-
tection and empowerment network as it maps how various actors are connected with one
another. It relies on an original dataset that records public-private partnerships based on
collaborative programs and projects and produces one of the first statistical analyses of this
network; it exposes observable and measurable factors that contribute to partnership for-
mation. Further, the chapter systematically explores who is more likely to partner up with
whom, uncovering geographic and actor type homophily (similarity). However, this chapter
is limited to an investigation of actors who have already made it into the network as well as
to factors that are measurable in a statistical way. The chapter cannot speak to why and
how businesses and other actors go about finding and forming partnerships, what challenges
they encounter, and how collaborative programs and projects are created and evaluated.
Chapter 4 therefore provides the perfect complement to chapter 3.
Chapter 4 continues with the exploration into public-private cooperation in refugee pro-
tection and empowerment, however with a focus on the actual decision-making processes
and dynamics of creating these partnerships. With its purely qualitative approach, this
chapter offers answers to the why and how questions. It contends and shows that finding
and forming partnerships is a taxing, long process. It speculates that there are additional,
difficult-to-measure/quantify factors that figure centrally in the partnership process. The
analysis reveals that actors, especially states and organizations, pay close attention to their
potential partners’ reputation, visibility, and type of sector/industry. In addition, actors
need to find and agree on shared values and goals as well as negotiate who carries which risk
in the partnership.
These factors and processes not only complicate gaining access to the network but also
prolong partnership and program creation. Businesses are especially affected by this, which
confirms the finding from chapter 3 that private for-profit actors have a more difficult time
forming partnerships. While not the main focus of this project, chapter 4 also provides some
insights into how actors attempt to evaluate the performance of the resulting collaborative
119
programs and projects. It exposes challenges and frustrations with developing and applying
appropriate evaluation methods and metrics. However, with the business sector as an active
program contributor, there lies great potential in developing novel solutions and updating
existing evaluation practices and metrics.
Bringingitalltogether, thisdissertationhasattemptedtoshedlightontherisingpublic-
private cooperation in the refugee empowerment and protection space. This project has fo-
cusedontheexperienceofthecorporatesector, therelativelynewactorinthisnewlyemerg-
ing network. It illuminated business motivations such as consumer pressure and showed how
the public feels about CSR activities in refugee protection and empowerment. The project
also deconstructed the partnership creation process from multiple angles. In chapter 3, it
took a holistic and systematic look at the overall structure of the network and tested factors
that influence the chances of forming partnerships and who partners up with whom. Chap-
ter 4 then built upon chapter 3 by engaging in an in-depth analysis of the decision making
processes, motivations, and challenges of finding potential partners, forming partnerships,
and evaluating the respective project/programs.
It is likely that public-private collaboration in refugee protection and empowerment will
continue to grow in the coming years. This raises another important questions: (to what
extent) is business involvement in the refugee cause actually good? The next section will
offer some thoughts on this.
5.2 The Corporate Elephant in the Room: Is Business
Involvement Good?
Whether and to what extent corporate involvement in refugee protection and empowerment
is good and/or ethical is a whole research agenda in and of itself. Nevertheless, this disser-
tation would feel incomplete without offering, at a minimum, a discussion of the potential
120
consequences of increased corporate influence in the refugee protection and empowerment
space.
For what or whom are there, potentially negative, consequences? Undoubtedly, multiple
actors could be affected including organizations, states, as well as vulnerable populations
themselves. As pointed out in the literature review, the diversification of actors and shifting
patterns of authority in the international system could change some of the traditional ac-
tors’ standing or status in the refugee protection and empowerment space. There could be
increased competition amongst organizations and businesses for landing a partnership or a
project, which most likely would affect smaller actors more than bigger ones. Buse and Walt
even go as far as proclaiming that PPPs could “further fragment international cooperation
inhealthandundermineUNaimsforcooperationandequityamongstates”(BuseandWalt,
2002, pg.177).
However, given the vastly different functions and purpose of businesses, competition may
belimited. Instead,competitionislikelymostintenseamongstsameactortypes(i.eamongst
organizations themselves and amongst businesses themselves) as these actors have similar
goals and functions. When it comes to the potential impact of PPPs on international co-
operation, it is true that it may spur institutional fragmentation. However, institutional
fragmentation is already under way and is not necessarily a bad thing. Roberts et al. (2002)
forexample expressthat becauseof thisdiversification, theinternationalcommunityis actu-
ally able to develop innovative solutions, consider critical issues from multiple perspectives,
and tackle global problems better than before.
Maybe even more importantly, what are the consequences of increased business involve-
ment for the actual target population: refugees and asylum seekers? Negative consequences
are often associated with the motivations of the business sector. It is a logical fallacy to
assume that business profit motivations necessarily produce harmful outcomes for forcibly
displaced people. As shown in chapter 4, having multiple motivations and goals is not
121
uncommon and does not preclude businesses contributing significantly to the well-being of
asylum seekers and refugees.
However, there may still be chance that the involvement of businesses may actually exac-
erbate a humanitarian crisis. If businesses provide services and products to or, alternatively,
purchase products or commodities from belligerent actors that are responsible and/or at the
root of a crisis, these businesses could prolong or worsen the humanitarian crisis (Collier
et al., 2004, 2008). An example of this is the case of the cement company Lafarge SA, which
was charged to be complicit in crimes against humanity by the Paris High Court in 2018
after it was discovered that the company paid the Islamic State (ISIS or ISIL) millions of
dollars for oil and raw materials to ensure the continuation of company operations during
the Syrian Civil War (Alderman, 2018).
PoliticalscienceandIRscholarshiphasbeencriticalofbusinessinvolvementandhasoften
held the view that corporations (and foreign direct investment) is linked to civil conflict and
human rights abuses (Collier and Hoeffler, 2004; Collier and Sambanis, 2005; Davenport,
2007; Hook and Ganguly, 2000). However, these studies produce mixed results and are
all conducted at the macro-level. They do not actually investigate corporate behavior and
policy-making that shows the negative impact of firms in crisis situations.
In contrast, businesses literature has devoted attention to the actual business practices
inconflictpreventionandcrisismanagementandhasidentifiedpositivewaysthatbusinesses
havecontributedandcancontributetostabilityandpreventionofhumanitariancrises(Ben-
nett,2002;Nelson,2000;Oetzeletal.,2009). Especiallyrelevantfortheinvolvementoffirms
in refugee protection and empowerment are the findings of Nelson (2000), who finds that
firms work to address root causes of conflict (inequality, ethnic tensions, etc.) through the
creation of programs and projects.
Firmscanalsoplayacrucialroleinthepost-conflictspaceastheyareabletohelprebuild
communities and societies (OCHA, 2017). The business sector is able to provide essential
goods and services as well as more advanced skill and employment programs/projects that
122
can give the affected population a “renewed sense of purpose, perspective, and social cohe-
sion” (Hotho and Girschik, 2019, pg. 207). In addition, especially when businesses work
with other entities including governments and organizations, there tend to be more checks
and balances amongst partners. This may remedy some potential negative consequences. In
interviews, I asked most of my study participants about the ethical concerns and potential
negative consequences about business engagement in refugee protection and empowerment.
Mostofthemcouldthinkofmoreadvantagesthandisadvantagesandconfirmedthatthrough
these PPPs there is overall more transparency and accountability amongst partners.
Now this is not to say that there are no ethical concerns or negative consequences when
businesses get involved in the humanitarian space. As pointed out above, there may be
negative consequences, and there are examples of unethical behavior of companies involved
in crisis situations. However, in light of the difficult access to the refugee protection and
empowerment space, the increased transparency and accountability through these PPPs,
as well as the type of businesses that self-select into this space, there is hope that ethical
concerns and negative consequences are somewhat manageable. Undoubtedly, more research
is necessary to evaluate the actual impact and consequences of the business sector in the
refugee protection and empowerment space.
Having provided some thoughts on the ethical concerns and potential consequences of
business involvement in the refugee protection and empowerment space, the next section
addresses what we could and should expect from public-private cooperation in the refugee
protection and empowerment space in the future.
5.3 TheFutureofPublic-PrivateCooperationinRefugee
Issues
Over the past few years, public-private cooperation in the refugee protection and empower-
mentspacehasconsistentlygrownandislikelytocontinueonthistrend. Withnewlyarising
123
crises and refugee situations around the world, there is increased need and opportunity for
public and private actors to work together and develop novel solutions.
As addressed in the introduction, the business sector has already played a critical role
in the emergency response to the escalating humanitarian crisis in the Ukraine. Businesses,
organizations, and governments are actively working together to deliver emergency services
and goods (shelter, food, clothing, etc.). Businesses havealso beeninstrumental inenforcing
sanctions and limiting or even halting business operations in Russia (Smith, 2022).
While especially financial contributions of the business sector to organizations and state
actors have been essential for financing, maintaining, and expanding projects and programs,
this dissertation has shown that collaboration extends far beyond passive involvement of
firms. Businesses have taken on active program management, oversight, and execution roles
and are often even the main program/project developer in these collaborative partnerships.
The results are often novel, innovative solutions and programs that may be better suited
to address the escalating global humanitarian crises that force people to flee their homes.
Especially in light of increasingly complex issues and diverse needs ranging from access to
basic goods and services (food, health care, shelter) to long-term integration measures, it
is sensible to assume that the private sector will play an even more active role in these
partnerships.
Although this dissertation has illuminated that often states and organizations initiate
these partnerships in the refugee protection and empowerment space, this may change in
the future. Big corporations often have enough resources and manpower to plan and initiate
projects and programs - especially those projects and programs that are closely related to
the firm’s expertise or core business operations (i.e. IKEA manufacturing shelter, Google
creating technology for mapping refugee movements, etc.). In addition, the more companies
are entering the refugee protection and empowerment space and the longer these firms oper-
ate in this space, the more practical knowledge they gain about creating, maintaining, and
executing programs and projects for forcibly displaced people. Consequently, firms may be
124
moreinclinedtospearheadevenmoreofthesecollaborativeprojectsandprogramsgiventheir
growing knowledge and their already existing partnerships and access to asylum seekers and
refugees. We are already observing this trend: many businesses have been renewing and/or
expanding their commitments to the refugee cause, which has strengthened existing public-
private cooperation and resulted in new partnerships.
There is also opportunity for smaller and medium-sized businesses to more easily par-
ticipate and form partnerships in the future. Organizations and state actors have expressed
how important it is and will be to include corporate actors of different sizes and levels.
They realize that especially for successful integration efforts, cooperation with local and re-
gional companies is vital. Consequently, we see steadily more cooperation amongst cities,
municipalities, NGOs, and private businesses (OECD, 2020).
What are the practical implications for burgeoning public-private cooperation in the
refugee protection and empowerment space? Including the private for-profit sector in the
creation, development, and execution of solutions for one of the worst humanitarian crises
in the world has ample benefits. Businesses have become increasingly authoritative actors
on the global stage; it is therefore sensible to include this important actor as part of the
solution development. The participation in PPPs holds all actors accountable and fosters
transparency. Giventhatnoactoralonecandevelopsufficientsolutionstothevariousrefugee
situations across the globe, increased public-private cooperation presents a more promising
approach.
These PPPs distribute responsibilities, benefits, and risks across partners, which eases
the overall burden on individual actors. They allow actors to pool expertise, experience, and
resources to establish more sophisticated, sustainable projects and programs. The results
may include faster mobilization of resources and funding in emergency situations (Ukraine
as an example), innovative technology that benefits asylum seekers, refugees and local com-
munities
125
(connectivity/internet projects, mapping of conflicts and refugee movements, updated pro-
grammeasurementmetrics,etc.),andbettereconomicinclusion(skill/language/employment
programs, loans and coaching for refugee entrepreneurs and businesses).
The implications of increased public-private cooperation in refugee protection and em-
powermentseemtobeencouraging. However,moreresearchisnecessarytoreallyunderstand
the actual implications and consequences of these PPPs. The next section will therefore lay
out avenues for future research.
5.4 A Future Research Agenda
This dissertation project has made the first steps towards investigating this burgeoning
public-private refugee protection and empowerment network. Simultaneously, this project
has discovered gaps and questions that warrant further exploration. I see great research
potential in the following three areas: 1) a focus on the sector-specific and region-focused
development of this PPP network, 2) an analysis of the performance and/or success of these
collaborative PPP projects and programs, and connected to this 3) the ethics and potential
and actual consequences of these PPPs.
First, as this public-private network of active participants in refugee protection and em-
powerment propagates, it will grow even more diverse with entities of different sizes, sectors,
andgeographicregions. Thisdissertationhastakenaverybroad,globalapproachbecauseits
goal was to analyze PPP emergence in the refugee protection and empowerment overall. It
therefore did not differentiate between sectors or geographic regions. Undoubtedly however,
there may be different dynamics and levels of difficulty within/across sectors and geographic
regionswhenitcomestocreatingthesePPPsandtherespectivecollaborativeprograms. For
example, establishing public-private relationships and collaborative programs around deliv-
ering material goods or food may be easier than creating PPPs that focus on teaching more
intangible skills such as job training or language. Consequently, certain industries may have
126
easier access to the network and find it easier to form partnerships. Future research should
therefore consider this sector-specific variation in access and establishing partnerships.
As this dissertation has also uncovered in chapter 3, there tends to be geographic ho-
mophily in the PPP network. In other words, actors located in the same geographic region
are more likely to form partnerships than actors across regions. This finding gives reason
to speculate that these inter-region partnerships are more difficult to build and maintain
compared to intra-region ones. Understanding these potential geographic barriers is vital
for two main reasons. First, refugee crises are international and involve at a minimum two
states (home state and host state), but often far more states due to asylum seekers crossing
multiple borders. This means that it would be important to have geographically diverse
partnerships and to discover what potentially inhibits their development.
Second, over 70 percent of refugees reside in developing countries - often in countries
that have limited resources and capacity and sometimes face political, social, and/or eco-
nomic instability. This makes it especially difficult for regional and local public and private
actors to operate and provide services and programs that forcibly displaced people need.
On the other hand, the wealthiest and most influential actors, which can provide necessary
resources, funding, and expertise, are headquartered in Europe and the US. Transnational
andtransregionalPPPsarethereforeessentialtoensurethattheselocalandregionalentities
are able to function, and that asylum seekers and refugees have access to appropriate goods
and services. It would be important for scholars to identify the main obstacles of these
geographically diverse partnerships to then offer recommendations on how to remedy and
overcome them.
The second area of future research revolves around the performance of these PPPs in
refugee protection and empowerment. This dissertation has touched upon the topic of how
actors evaluate the performance of these collaborative programs and projects, but it did not
offer an in-depth investigation. Given the focus was on the various facets of the emergence
of public-private collaboration in the refugee cause, a more sophisticated analysis of the
127
performance or success was outside the scope of this project. An evaluation of the perfor-
mance of these projects and programs is vital yet challenging. We know little about the
performance or success of these PPP programs and projects. This is due to 1) incomplete or
biased data as these actors are inclined to overreport success and underreport failures and 2)
the difficulty of measuring success. Even the brief look of this dissertation at the evaluation
practices of actors revealed these dilemmas: public and private actors often disagree on the
appropriate ways to measure performance and engage in varying evaluation metrics.
However, understanding whether and to what extent these PPPs in refugee protection
and empowerment aresuccessful has important implications. For scholarship, it would mean
developing and advancing our theoretical insights of what makes multi-stakeholder cooper-
ation work. Researchers as well as practitioners could greatly benefit from a sophisticated
empirical approach to measuring the PPP performance. Most importantly perhaps are the
policy implications as practitioners would be able to adjust programs and projects early on
and/or create more success-promising programs.
While this seems to be an impossible task, I am confident that it can be accomplished
through a multi-methods approach. For example, it could be especially beneficial to con-
duct field research/participant observation to gain direct insights of the workings of these
projects/programs. In addition, text analysis of reports and speeches could simplify deci-
phering public evaluation techniques and output. I also believe, besides having quantifiable
metrics like numbers of refugees enrolled/employed, an evaluation would necessarily need to
include feedback from the actual beneficiaries (asylum seekers and refugees).
Finally and connected to the performance of these programs, there is a need for a deeper
examination of the ethics and potential consequences of these PPPs in refugee protection
and empowerment. Though this conclusion has briefly touched upon this matter, it merely
scratched the surface. What makes involvement of the business sector more or less ethical?
What are the subsequent ethical concerns for increased public-private cooperation in this
space? Questionsthatarebeyondthescopeofthisdissertation, however, shouldandneedto
128
be addressed in the future. A research program of the ethics would consider these questions
and discuss various consequences for different actors (organizations, states, beneficiaries,
asylum seekers, refugees and even businesses themselves).
Public-private cooperation in the refugee protection and empowerment space presents a
vastandrichresearchagenda. Withnewcrisesemergingaroundtheworld,thestudyofthese
more diverse forms of collaboration will become increasingly important. This dissertation
project has taken the first step into the promising and existing research landscape in the
hope of more scholars to follow course, compliment, and expand its findings.
129
Bibliography
Abbott, K. W. and Snidal, D. (2010). International regulation without international gov-
ernment: Improving IO performance through orchestration. The Review of International
Organizations, 5(3):315–344.
Acharya, A. (2016). The Future of Global Governance: Fragmentation May Be Inevitable
and Creative. Global Governance: A Review of Multilateralism and International Organi-
zations, 22(4):453–460.
Aguinis,H.andBradley,K.J.(2014). BestPracticeRecommendationsforDesigningandIm-
plementing Experimental Vignette Methodology Studies. Organizational Research Meth-
ods, 17(4):351–371.
Alderman, L. (2018). French Cement Giant Lafarge Indicted on Terror Financing Charge in
Syria. New York Times.
Andonova, L. B. (2006). Globalization, Agency and Institutional Innovation: The Bise and
Fall of Public-Private Partnerships in Global Governance.
Andonova, L. B. (2010). Public-Private Partnerships for the Earth: Politics and Patterns of
Hybrid Authority in the Multilateral System. Global Environmental Politics, 10(2):25–53.
Andonova, L. B. (2017). Governance Entrepreneurs: International Organizations and the
Rise of Global Public-Private Partnerships. Business and Public Policy. Cambridge Uni-
versityPress, CambridgeNewYorkMelbourneNewDelhiSingapore. OCLC:1020082249.
Andonova, L. B. and Levy, M. A. (2003). Franchising Global Governance: Making Sense
of the Johannesburg Type II Partnerships. In Stokke, O. and Thommenssen, O., editors,
In Yearbook of International Co-Operation on Environment and Development. Earthscan
Publishing, London.
Arya, B. and Salk, J. E. (2006). Cross-Sector Alliance Learning and Effectiveness of Volun-
tary Codes of Corporate Social Responsibility. Business Ethics Quarterly, 16(2):211–234.
Publisher: Cambridge University Press.
Atzm¨ uller, C., Su, D., and Steiner, P. (2017). Designing Valid and Reliable Vignette Exper-
iments for Survey Research: A Case Study on the Fair Gender Income Gap. Journal of
Methods and Measurement in the Social Sciences, 7(2).
Avant, D. D., Finnemore, M., Sell, S. K., and Cambridge University Press (2010). Who
Governs the Globe? Cambridge University Press, Cambridge. OCLC: 857982622.
130
Axelrod,R.M.(1984). The evolution of cooperation. BasicBooks,NewYork,rev.ededition.
OCLC: ocm76963800.
Backer, L. C. (2011). Private Actors and Public Governance Beyond the State: The Multi-
national Corporation, the Financial Stability Board and the Global Governance Order.
Indiana Journal of Global Legal Studies, 18(751).
Barbara, J. (2006). Nation building and the role of the private sector as a political peace-
builder: Opinion. Conflict, Security & Development , 6(4):581–594.
Barnett, M. and Finnemore, M. (2004). Rules for the World: International Organizations
in Global Politics. Cornell University Press, Ithaca and London.
Becker-Olsen, K. L., Cudmore, B. A., and Hill, R. P. (2006). The impact of perceived corpo-
ratesocialresponsibilityonconsumerbehavior. JournalofBusinessResearch,59(1):46–53.
Beisheim, M. and Liese, A. (2014). Transnational partnerships effectively providing for sus-
tainable development? Palgrave Macmillan, Basingstoke. OCLC: 880416526.
Bennett, J. (2002). Multinational Corporations, Social Responsibility and Conflict. Journal
of International Affairs , 55(2):393–410. Publisher: Temporary Publisher.
Betts, A., Bloom, L., Kaplan, J. D., and Omata, N. (2017). Refugee Economies: Forced
Displacement and Development. Oxford University Press, Oxford, first edition edition.
OCLC: ocn955312871.
Bhattacharya, C. and Korschun, D. (2008). Stakeholder Marketing: Beyond the Four Ps
and the Customer. Journal of Public Policy & Marketing, 27(1):113–116.
Bingham, A. J. and Witkowsky, P. (2022). Deductive and inductive approaches to quali-
tative data analysis. In Vanover, C., Mihas, P., and Saldana, J., editors, Analyzing and
interpreting qualitative research: after the interview, pages 133–146. SAGE Publications,
Los Angeles, first edition edition.
Bisong, A. and Knoll, A. (2020). Mapping private sector engagement along the migration
cycle. Technicalreport,EuropeanCentreforDevelopmentPolicyManagement(ECDPM),
Maastricht, Netherlands.
Boulding, C. (2009). Accountability in Foreign Aid Delivery: Links between Donors and
NGOs. Toronto.
Boyon, N. (2022). World Refugee Day 2022: New Ipsos survey shows greater compassion for
forcibly displaced as war in Ukraine wears on| Ipsos.
Brown, D. L. and Woods, N., editors (2007). Making Global Self-Regulation Effective in
Developing Countries. Oxford University Press, Oxford [England] ; New York. OCLC:
ocn145396643.
Brutger, R., Kertzer, J. D., Renshon, J., Tingley, D., and Weiss, C. M. (2021). Abstraction
and Detail in Experimental Design. American Journal of Political Science.
131
Bull, B., Bø˚ as, M., and McNeill, D. (2004). Private Sector Influence in the Multilateral
System: A Changing Structure of World Governance? Global Governance, 10(4):481–498.
Buse, K. and Harmer, A. M. (2007). Seven habits of highly effective global public–private
health partnerships: Practice and potential. Social Science & Medicine, 64(2):259–271.
Buse, K. and Walt, G. (2002). The World Health Organization and Global Public-Private
Health Partnerships: In Search of ‘Good’ Global Health Governance. In Reich, M., editor,
Public-private partnerships for public health, Harvard series on population and interna-
tional health, pages 169–197. Harvard Center for Population and Development Studies,
Cambridge, Mass.
B¨ ackstrand, K. (2008). Accountability of Networked Climate Governance: The Rise of
Transnational Climate Partnerships. Global Environmental Politics, 8(3):74–102.
B¨ uthe, T. and Mattli, W. (2011). The new global rulers: the privatization of regulation
in the world economy. Princeton University Press, Oxford ; Princeton, N.J. OCLC:
ocn679940574.
Cammett, M. C., MacLean, L. M., and Gough, I., editors (2014). The Politics of Non-State
Social Welfare. Cornell University Press, Ithaca.
Campbell,S.,DiGiuseppe,M.,andMurdie,A.(2019). InternationalDevelopmentNGOsand
BureaucraticCapacity: FacilitatororDestroyer? Political Research Quarterly,72(1):3–18.
Carroll, A. B. (2021). Corporate Social Responsibility: Perspectives on the CSR Construct’s
Development and Future. Business & Society, 60(6):1258–1278.
Caruana, R. (2010). Morality and consumption: towards a multidisciplinary perspective.
Journal of Marketing Management. Publisher: Taylor & Francis Group.
Caruana, R. and Chatzidakis, A. (2014). Consumer Social Responsibility (CnSR): Toward
a Multi-Level, Multi-Agent Conceptualization of the “Other CSR”. Journal of Business
Ethics, 121(4):577–592. Company: Springer Distributor: Springer Institution: Springer
Label: Springer Number: 4 Publisher: Springer Netherlands.
Cavill, S. and Sohail, M. (2007). Increasing strategic accountability: a framework for inter-
national NGOs. Development in Practice, 17(2):231–248. Publisher: Routledge eprint:
https://doi.org/10.1080/09614520701196004.
Chan, S. (2012). Partnership for Sustainable Development Beyond the OECD World: China
and India. In Pattberg, P. H., Biermann, F., Sander, C., and Mert, A., editors, Public-
private partnerships for sustainable development: emergence, influence and legitimacy ,
pages 115–137. Edward Elgar, Cheltenham ; Northampton, Ma. OCLC: ocn769547078.
Chernev, A. and Blair, S. (2015). Doing Well by Doing Good: The Benevolent Halo of
Corporate Social Responsibility. Journal of Consumer Research, 41(6):1412–1425.
132
Clapp, J. (2009). The Global Food Crisis and International Agricultural Policy: Which Way
Forward? Global Governance: A Review of Multilateralism and International Organiza-
tions, 15(2):299–312.
Collier, P. and Hoeffler, A. (2004). Greed and Grievance in Civil War. Oxford Economic
Papers, 56(4):563–595. Publisher: Oxford University Press.
Collier, P., Hoeffler, A., and Rohner, D. (2008). Beyond greed and grievance: feasibility and
civil war. Oxford Economic Papers, 61(1):1–27.
Collier, P., Hoeffler, A., and S¨ oderbom, M. (2004). On the Duration of Civil War. Journal
of Peace Research, 41(3):253–273. Publisher: Sage Publications, Ltd.
Collier, P. and Sambanis, N. (2005). Understanding Civil War : Evidence and Analysis.
Technical Report 1, World Bank, Washington, DC. Accepted: 2012-06-07T19:08:39Z
ISBN: 9780821360477.
Cretu,A.E.andBrodie,R.J.(2007). Theinfluenceofbrandimageandcompanyreputation
where manufacturers market to small firms: A customer value perspective. Industrial
Marketing Management, 36(2):230–240.
Cutler,A.C.,Haufler,V.,andPorter,T.,editors(1999). PrivateAuthorityandInternational
Affairs . SUNY Series in Global Politics. State University of New York Press, Albany.
Dahlsrud, A. (2008). How corporate social responsibility is defined: an analysis of 37 defini-
tions. Corporate Social Responsibility and Environmental Management, 15(1):1–13.
Davenport, C. (2007). State Repression and Political Order. Annual Review of Political
Science, 10(1):1–23. eprint: https://doi.org/10.1146/annurev.polisci.10.101405.143216.
Devinney, T. M., Auger, P., Eckhardt, G., and Birtchnell, T. (2006). The Other CSR:
Consumer Social Responsibility. Stanford Social Innovation Review, 4:30–37.
Diamantopoulos, A., Schlegelmilch, B. B., Sinkovics, R. R., and Bohlen, G. M. (2003). Can
socio-demographics still play a role in profiling green consumers? A review of the evidence
and an empirical investigation. Journal of Business Research, 56(6):465–480.
Dickson, M. A. (2001). Utility of No Sweat Labels for Apparel Consumers: Profiling Label
Users and Predicting Their Purchases. Journal of Consumer Affairs , 35(1):96–119.
Drummond, J. and Crawford, N. (2014). Humanitarian crises, emergency preparedness and
response: The role of business and the private sector. Kenya case study.
Du, S., Bhattacharya, C., and Sen, S. (2010). Maximizing Business Returns to Corporate
Social Responsibility (CSR): The Role of CSR Communication. International Journal of
Management Reviews, 12(1):8–19.
Du, S., Yu, K., Bhattacharya, C., and Sen, S. (2017). The Business Case for Sustainability
Reporting: EvidencefromStockMarketReactions. Journal of Public Policy & Marketing,
36(2):313–330.
133
Ebert, T. (2009). Trust as the key to loyalty in business-to-consumer exchanges: trust
building measures in the banking industry. Gabler Verlag, Wiesbaden, 1 edition. ISBN:
9783834983077 Series: Gabler Edition Wissenschaft.
Ebrahim,A.(2003). AccountabilityInPractice: MechanismsforNGOs. WorldDevelopment,
31(5):813–829.
Elias, R. Z. (2004). An Examination of Business Students’ Perception of Corporate Social
ResponsibilitiesBeforeandAfterBankruptcies. JournalofBusinessEthics,52(3):267–281.
Ellen, P. S., Webb, D. J., and Mohr, L. A. (2017). Building Corporate Associations: Con-
sumerAttributionsforCorporateSociallyResponsiblePrograms:. Journal of the Academy
of Marketing Science. Publisher: Sage PublicationsSage CA: Thousand Oaks, CA.
Erdem, T., S ¸irin, C., Singh, V., and Zhang, Q. P. (2018). How Helping Refugees Helps
Brands. Technical report, NYU, New York.
Facchini,G.,Mayda,A.M.,andMishra,P.(2011). DointerestgroupsaffectUSimmigration
policy? Journal of International Economics, 85(1):114–128.
Falkner, R. (2005). The business of ozone layer protection: Corporate power in regime
evolution. In Newell, P. and Levy, D., editors, The Business of Global Environmental
Governance. Global Environmental Accord: Strategies for Sustainability and Institutional
Innovation., pages 105–34. MIT Press, Cambridge, MA.
Foreh, M. R. and Grier, S. (2003). When Is Honesty the Best Policy? The Effect of Stated
Company Intent on Consumer Skepticism. Journal of Consumer Psychology, 13(3):349–
356.
Freeman, G. P. (1995). Modes of Immigration Politics in Liberal Democratic States. Inter-
national Migration Review, 29(4):881.
Fu, F., Nowak, M. A., Christakis, N. A., and Fowler, J. H. (2012). The Evolution of
Homophily. Scientific Reports , 2(1).
Fuchs, D. A. (2007). Business power in global governance. Lynne Rienner Publishers, Inc,
Boulder. OCLC: ocm82472240.
Funk, C. and Hefferon, M. (2019). U.S. Public Views on Climate and Energy.
Gaines, B. J., Kuklinski, J. H., and Quirk, P. J. (2007). The Logic of the Survey Experiment
Reexamined. Political Analysis, 15(1):1–20.
Galletta, A. (2013). Mastering the semi-structured interview and beyond: from research
design to analysis and publication. Qualitative studies in psychology. New York Univ.
Press, New York.
Gammeltoft-Hansen, T. and Sørensen, N. N., editors (2013). The Migration Industry and
the Commercialization of International Migration. Routledge Global Institutions Series.
Routledge, London ; New York.
134
Gent, S. E., Crescenzi, M. J. C., Menninga, E. J., and Reid, L. (2015). The reputation trap
of NGO accountability. International Theory, 7(3):426–463.
Goldstein, K. (2002). Getting in the Door: Sampling and Completing Elite Interviews.
PS: Political Science and Politics, 35(4):669–672. Publisher: [American Political Science
Association, Cambridge University Press].
Gonz´ alez-Rodr´ ıguez Ma, R., D´ ıaz-Fern´ andez Ma, C., and Simonetti, B. (2014). Values and
corporate social initiative: An approach through Schwartz theory. International Journal
of Business and Society, 15(1):19.
Gray, J. and Hicks, R. P. (2014). Reputations, Perceptions, and International Economic
Agreements. International Interactions, 40(3):325–349.
Graz, J.-C. and N¨ olke, A., editors (2008). Transnational private governance and its limits.
Number51inRoutledge/ECPRstudiesinEuropeanpoliticalscience.Routledge, London.
OCLC: 255747629.
Grose, C. R. (2005). Disentangling Constituency and Legislator Effects in Legislative Repre-
sentation: BlackLegislatorsorBlackDistricts?*. Social Science Quarterly, 86(2):427–443.
Gruener, S., Berggrund, E., and Kantowitz, R. (2021). Financing Peacebuilding: The Role
of Private-Sector Actors. Development Dialogue Paper No. 29, Dag Hammarskj¨ old Foun-
dation, Uppsala, Sweden.
Gugerty,M.K.(2009). Signalingvirtue: voluntaryaccountabilityprogramsamongnonprofit
organizations. Policy Sciences, 42(3):243–273.
Habel, J., Schons, L. M., Alavi, S., and Wieseke, J. (2016). Warm Glow or Extra Charge?
The Ambivalent Effect of Corporate Social Responsibility Activities on Customers’ Per-
ceived Price Fairness. Journal of Marketing, 80(1):84–105.
Hafner-Burton, E. M. and McNamara, H. M. (2019). United States Human Rights Policy:
The Corporate Lobby. Human Rights Quarterly, 41(1):115–142.
Hall,R.(1992).Thestrategicanalysisofintangibleresources. StrategicManagementJournal,
13(2):135–144. eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/smj.4250130205.
Hall, R. B. and Biersteker, T. J. (2002). The Emergence of Private Authority in Global
Governance. Cambridge University Press, Cambridge. OCLC: 776977467.
Hasenclever, A., Mayer, P., and Rittberger, V. (1997). Theories of International Regimes.
Number 55 in Cambridge Studies in International Relations. Cambridge University Press,
Cambridge ; New York.
Haufler, V. (2010). Corporations in zones of conflict: issues, actors, and institutions. In
Avant, D. D., Finnemore, M., and Sell, S. K., editors, Who Governs the Globe?, pages
102–130. Cambridge University Press, Cambridge.
135
Hern´ andez-Le´ on, R. (2008). Metropolitan Migrants: The Migration of Urban Mexicans to
the United States. University of California Press, Berkeley.
Hoeffler,S.,Bloom,P.N.,andKeller,K.L.(2010). UnderstandingStakeholderResponsesto
CorporateCitizenshipInitiatives: ManagerialGuidelinesandResearchDirections. Journal
of Public Policy & Marketing, 29(1):78–88.
Hoering, U. (2003). Panacea PPP. Public Private Partnerships in German Development
Cooperation. Exec- utive Summary.
Hook, J. and Ganguly, R. (2000). Multinational corporations and ethnic conflict: Theory
and experience. Nationalism and Ethnic Politics, 6(1):48–71.
Hotho, J. and Girschik, V. (2019). Corporate engagement in humanitarian action: Con-
cepts, challenges, and areas for international business research. critical perspectives on
international business, 15(2/3):201–218. Publisher: Emerald Publishing Limited.
Hur, W.-M., Kim, H., andWoo, J.(2014). HowCSRLeadstoCorporateBrandEquity: Me-
diating Mechanisms of Corporate Brand Credibility and Reputation. Journal of Business
Ethics, 125(1):75–86.
IFCandBridgespan(2019). PrivateSector&Refugees.PathwaystoScale. Technicalreport,
IFC and the Bridgespan Group, Washington, D.C.
International Organization for Standardization, . (2010). ISO 26000:2010(en) Guidance on
social responsibility.
Joireman, J., Smith, D., Liu, R. L., and Arthurs, J. (2015). It’s All Good: Corporate Social
Responsibility Reduces Negative and Promotes Positive Responses to Service Failures
among Value-Aligned Customers. Journal of Public Policy & Marketing, 34(1):32–49.
Jones Luong, P. (2018). Empowering Local Communities and Enervating the State? Foreign
Oil Companies as Public Goods Providers in Azerbaijan and Kazakhstan. In Cammett,
M. and MacLean, L. M., editors, The Politics of Non-state Social Welfare, pages 57–76.
Cornell University Press.
Kelly, S. W., James H. Donnelly, J., and Skinner, S. J. (1990). Customer participation in
service production and delivery. Journal of Retailing, 66(3):315–336. Publisher: Elsevier
Advanced Technology Publications.
Keohane, R. O. (1984). After hegemony: cooperation and discord in the world political
economy. A Princeton classic edition. Princeton University Press, Princeton, N.J, 1st
princeton classic ed edition. OCLC: ocm57750488.
Keohane, R. O. and Nye, J. S. (1977). Power and interdependence: world politics in transi-
tion. Little, Brown, Boston.
Kernaghan, K. (1993). Partnership and public administration: conceptual and practi-
cal considerations. Canadian Public Administration/Administration publique du Canada,
36(1):57–76.
136
Kim, D. J., Ferrin, D. L., and Rao, H. R. (2008). A trust-based consumer decision-making
model in electronic commerce: The role of trust, perceived risk, and their antecedents.
Decision Support Systems, 44(2):544–564.
Koppenjan,J.J.F.(2005).TheFormationofPublic-PrivatePartnerships: LessonsfromNine
Transport Infrastructure Projects in The Netherlands. Public Administration, 83(1):135–
157.
Koschate-Fischer, N., Diamantopoulos, A., and Oldenkotte, K. (2012). Are Consumers
Really Willing to Pay More for a Favorable Country Image? A Study of Country-of-
Origin Effects on Willingness to Pay. Journal of International Marketing, 20(1):19–41.
Laczniak, G. R. and Murphy, P. E. (2012). Stakeholder Theory and Marketing: Moving
from a Firm-Centric to a Societal Perspective. Journal of Public Policy & Marketing,
31(2):284–292.
Latap´ ı Agudelo, M. A., J´ ohannsd´ ottir, L., and Dav´ ıdsd´ ottir, B. (2019). A literature review
of the history and evolution of corporate social responsibility. International Journal of
Corporate Social Responsibility, 4(1):1.
Latif, K. F. and Sajjad, A. (2018). Measuring corporate social responsibility: A critical
review of survey instruments. Corporate Social Responsibility and Environmental Man-
agement, 25(6):1174–1197.
Lester, J. N., Cho, Y., and Lochmiller, C. R. (2020). Learning to Do Qualitative Data
Analysis: A Starting Point. Human Resource Development Review, 19(1):94–106.
Levy, D. L. and Newell, P., editors (2005). The Business of Global Environmental Gover-
nance. Global Environmental Accord. MIT Press, Cambridge, Mass.
Lichtenstein, D. R., Drumwright, M. E., and Braig, B. M. (2004). The Effect of Corporate
SocialResponsibilityonCustomerDonationstoCorporate-SupportedNonprofits. Journal
of Marketing, 68(4):16–32.
Luke, D. (2015). A User’s Guide to Network Analysis in R. Use R! Springer International
Publishing, Cham.
Luo, X. and Bhattacharya, C. (2006). Corporate Social Responsibility, Customer Satisfac-
tion, and Market Value. Journal of Marketing, 70(4):1–18.
Malik, A. A., Bernstein, H., Mohr, E., and Irvin-Erickson, Y. (2018a). For-Profit Humani-
tarians: IKEA’s Partnership with the Jordan River Foundation. Technical report, Urban
Institute, Washington, D.C.
Malik, A. A., Mohr, E., and Irvin-Erickson, Y. (2018b). Private-Sector Humanitarians?
New Approaches in the Global Refugee Response. Technical report, Urban Institute,
Washington, D.C.
137
Martens, J. and Seitz, K. (2019). Rules of engagement between the UN and private actors
towardsaregulatoryandinstitutionalframework. Technicalreport, MISEREOR,Aachen,
Germany. OCLC: 1285504109.
Mattli, W. and Woods, N. (2011). The Politics of Global Regulation. Princeton University
Press, Princeton. OCLC: 741492617.
May, C. (2006). Global corporate power and the UN global compact. In May, C., editor,
Global Corporate Power, pages 273–82. Lynne Reinner, London.
McPherson, M., Smith-Lovin, L., and Cook, J. M. (2001). Birds of a Feather: Homophily in
Social Networks. Annual Review of Sociology, 27(1):415–444.
Menz,G.(2011). Thepoliticaleconomyofmanagedmigration: nonstateactors, Europeaniza-
tion, and the politics of designing migration policies. Oxford University Press, Oxford.
Mercer, J. (1996). Reputation and international politics. Cornell studies in security affairs.
Cornell University Press, Ithaca, N.Y.
Mert, A. (2015a). Environmental Governance through Partnerships. Edward Elgar Publish-
ing.
Mert, A. (2015b). Sustainable Development Partnerships in the UN System. In Net-
works for Prosperity: Advancing Sustainability through Partnerships, pages 55–67.
UNIDO/University of Leuven, Leuven, Belgium.
Meyers-Levy, J. (1989). Gender differences in information processing: a selectivity interpre-
tation in cognitive and affective responses to advertising. In Cafferata, P. and Tybout,
A., editors, Cognitive and Affective Responses to Advertising , pages 219–260. Lexington
Books, Lexington, MA.
Mihas, P. (2019). Qualitative Data Analysis. In Oxford Research Encyclopedia of Education.
Oxford University Press.
Miller, W. E. and Stokes, D. E. (1963). Constituency Influence in Congress. American
Political Science Review, 57(1):45–56.
Mohr, L. A., Webb, D. J., and Harris, K. E. (2001). Do Consumers Expect Companies to be
SociallyResponsible? TheImpactofCorporateSocialResponsibilityonBuyingBehavior.
Journal of Consumer Affairs , 35(1):45–72.
Moosmayer, D. C. and Fuljahn, A. (2010). Consumer perceptions of cause related marketing
campaigns. Journal of Consumer Marketing, 27(6):543–549.
Murdie, A. (2014). The Ties that Bind: A Network Analysis of Human Rights International
Nongovernmental Organizations. British Journal of Political Science, 44(1):1–27.
Nelson,J.(2000). The business of peace: the private sector as a partner in conflict prevention
and resolution. Prince of Wales Business Leaders Forum, London.
138
Nelson, J. (2002). Building partnerships: cooperation between the United Nations system
and the private sector. United Nations, Dept. of Public Information, New York. OCLC:
ocm50470260.
Newhouse, J. (2009). Diplomacy, Inc.: The Influence of Lobbies on U.S. Foreign Policy.
Foreign Affairs , 88(3):73–92. Publisher: Council on Foreign Relations.
OCHA (2017). The Business Case: A study of private sector engagement in humanitarian
action. Technical report.
OECD (2020). Local Inclusion of Migrants and Refugees. Technical report, Organisation for
Economic Co-operation and Development, Paris.
Oetzel, J., Westermann-Behaylo, M., Koerber, C., Fort, T. L., and Rivera, J. (2009). Busi-
ness and Peace: Sketching the Terrain. Journal of Business Ethics, 89(4):351–373.
Olkkonen, L. and J¨ a¨ askel¨ ainen, J. (2019). Corporate Activism: Exploring Corporate
Social Responsibility (CSR) Communication. Academy of Management Proceedings,
2019(1):17350. Publisher: Academy of Management.
Omata, N. and Kaplan, J. D. (2013). Refugee livelihoods in Kampala, Nakivale and Kyang-
wali refugee settlements: Patterns of engagement with the private sector.
Orlitzky, M., Schmidt, F. L., and Rynes, S. L. (2003). Corporate Social and Financial
Performance: A Meta-Analysis. Organization Studies, 24(3):403–441.
Osborne, S. P. (2005). Public-Private Partnerships Theory and Practice in International
Perspective. Routledge, London. OCLC: 1124365203.
Patrick, S. (2014). The Unruled World: The Case for Good Enough Global Governance.
Foreign Affairs , 93(1):58–73.
Pattberg, P. and Widerberg, O. (2016). Transnational multistakeholder partnerships for
sustainable development: Conditions for success. Ambio, 45(1):42–51.
Pattberg, P. H., Biermann, F., Sander, C., and Mert, A., editors (2012). Public-private
partnerships for sustainable development: emergence, influence and legitimacy . Edward
Elgar, Cheltenham ; Northampton, Ma. OCLC: ocn769547078.
Post, A. (2018). The Politics of “Contracting Out” to the Private Sector: Water and Sanita-
tioninArgentina. InCammett,M.andMacLean,L.M.,editors, The Politics of Non-state
Social Welfare, pages 77–98. Cornell University Press.
Raghubir, P., Roberts, J., Lemon, K. N., and Winer, R. S. (2010). Why, When, and How
should the Effect of Marketing be Measured? A Stakeholder Perspective for Corporate
Social Responsibility Metrics. Journal of Public Policy & Marketing, 29(1):66–77.
Reich, M., editor (2002). Public-private partnerships for public health. Harvard series on
population and international health. Harvard Center for Population and Development
Studies, Cambridge, Mass.
139
Reimann, K. D. (2005). Up to No Good? Recent Critics and Critiques of NGOs. In
Subcontracting Peace. Routledge. Num Pages: 17.
Reinicke, W. (1998). Global public policy: governing without government? Brookings Insti-
tution Press, Washington, D.C.
Reinicke,W.andWitte,J.M.(2000). Interdependence,Globalization,andSovereignty: The
Role of Non-Binding International Legal Accords. In Shelton, D. L., editor, In Commit-
ment and Compliance. The Role of Non-Binding Norms in the International System.
Reinicke, W. H., Deng, F. M., and Witte, J. M. (2000). Critical choices: the United Nations,
networks, andthefutureofglobalgovernance. InternationalDevelopmentResearchCentre,
Ottawa. OCLC: ocm43930327.
Rindova, V. P., Williamson, I. O., Petkova, A. P., and Sever, J. M. (2005). Being Good or
Being Known: An Empirical Examination of the Dimensions, Antecedents, and Conse-
quencesofOrganizationalReputation. AcademyofManagementJournal,48(6):1033–1049.
Publisher: Academy of Management.
Rioux, J. S. (2008). US Crises and Domestic Politics: Crisis Outcomes, Reputation, and
Domestic Consequences. Southeastern Political Review, 25(2):199–239.
Roberts, J. A. (1996). Will the real socially responsible consumer please step forward?
Business Horizons, 39(1):79–83.
Roberts, M. J., Breitenstein, A., and Roberts, C. S. (2002). The ethics of public-private
partnerships. In Reich, M., editor, Public-private partnerships for public health, Harvard
series on population and international health, pages 67–87. Harvard Center for Population
and Development Studies, Cambridge, Mass.
Roberts, P. W. and Dowling, G. R. (2002). Corporate reputation and sustained supe-
rior financial performance. Strategic Management Journal, 23(12):1077–1093. eprint:
https://onlinelibrary.wiley.com/doi/pdf/10.1002/smj.274.
Robinson, S. R., Irmak, C., and Jayachandran, S. (2012). Choice of Cause in Cause-Related
Marketing. Journal of Marketing, 76(4):126–139.
Roger, C. and Dauvergne, P. (2016). The Rise of Transnational Governance as a Field of
Study. International Studies Review, 18(3):415–437.
Rosenau, P. V. and Linder, S., editors (2000). Public-Private Policy Partnerships. MIT
Press, Cambridge, Mass.
Roulston, K. and Choi, M. (2018). Qualitative interviews. In Flick, U., editor, The Sage
handbook of qualitative data collection, pages 233–249. SAGE Publications Ltd, Los An-
geles. OCLC: on1010566433.
Ruggie, J. G. (2007). Global markets and global governance: The prospects for convergence.
In Bernstein, S. and Pauly, L., editors, Global Liberalism and Political Order: Toward a
New Grand Compromise?, pages 23–50. SUNY Press, Albany, NY.
140
Ruggie, J. G. (2014). Global Governance and “New Governance Theory”: Lessons from
Business and Human Rights. Global Governance: A Review of Multilateralism and Inter-
national Organizations, 20(1):5–17.
Schwaiger, M. and Raithel, S. (2014). Reputation und Unternehmenserfolg. Management
Review Quarterly, 64(4):225–259.
Sch¨ aferhoff, M., Campe, S., andKaan, C.(2009). TransnationalPublic-PrivatePartnerships
inInternationalRelations: MakingSenseofConcepts, ResearchFrameworks, andResults.
International Studies Review, 11(3):451–474.
Sen, S. and Bhattacharya, C. (2001). Does Doing Good Always Lead to Doing Better?
Consumer Reactions to Corporate Social Responsibility. Journal of Marketing Research,
38(2):225–243.
Servaes, H. and Tamayo, A. (2013). The Impact of Corporate Social Responsibility on Firm
Value: The Role of Customer Awareness. Management Science, 59(5):1045–1061.
Sharma, P., Chen, I. S., and Luk, S. T. (2012). Gender and age as moderators in the service
evaluation process. Journal of Services Marketing, 26(2):102–114.
Simmons, C. J. and Becker-Olsen, K. L. (2006). Achieving Marketing Objectives through
Social Sponsorships. Journal of Marketing, 70(4):154–169.
Skoe, E. E. A., Cumberland, A., Eisenberg, N., Hansen, K., and Perry, J. (2002). [No title
found]. Sex Roles, 46(9/10):295–309.
Smith (2022). How Companies Are Responding to the War in Ukraine: A Roundup.
Smith, N. C. (2016). Morality and the market: consumer pressure for corporate accountabil-
ity. Routledge, London. OCLC: 944381454.
Smith, N. C., Drumwright, M. E., and Gentile, M. C. (2010). The New Marketing Myopia.
Journal of Public Policy & Marketing, 29(1):4–11.
Steffek, J. and Nanz, P. (2008). Emergent Patterns of Civil Society Participation in Global
and European Governance. In Steffek, J., Kissling, C., and Nanz, P., editors, Civil Society
Participation in European and Global Governance: A Cure for the Democratic Deficit? ,
Transformations of the State, pages 1–29. Palgrave Macmillan UK, London.
Tansey, O.(2007). ProcessTracingandEliteInterviewing: ACaseforNon-probabilitySam-
pling. PS: Political Science & Politics, 40(4):765–772. Publisher: Cambridge University
Press.
Tent (2022a). Amazon Commitment.
Tent (2022b). Cisco Commitment.
Tent (2022c). Coursera Commitment.
141
Tent (2022d). LinkedIn Commitment.
Tesner, S. and Kell, G. (2000). The United Nations and Business: a Partnership Recovered.
St. Martin’s Press, New York. OCLC: 1004390669.
Tian, Y., Hung, C., and Frumkin, P. (2020). An experimental test of the impact of corpo-
rate social responsibility on consumers’ purchasing behavior: The mediation role of trust.
Corporate Social Responsibility and Environmental Management, 27(6):2972–2982.
Tichenor, D. J. (2002). Dividing lines: the politics of immigration control in America.
Princeton studies in American politics. Princeton University Press, Princeton, N.J.
Tomz, M. (2007). Reputation and international cooperation: sovereign debt across three
centuries. Princeton Univ. Press, Princeton, NJ.
Torres, A., Bijmolt, T. H., Trib´ o, J. A., and Verhoef, P. (2012). Generating global brand
equity through corporate social responsibility to key stakeholders. International Journal
of Research in Marketing, 29(1):13–24.
UNDP (2013). POLICY ON DUE DILIGENCE AND PARTNERSHIPS WITH THE PRI-
VATE SECTOR. Technical report, UNDP, New York.
UNDP (2016). GUIDELINES FOR PRIVATE SECTOR PARTNER RISK ASSESSMENT
TOOL. Technical report, UNDP, New York.
UNHCR (2016). Corporate Guidelines| UNHCR Code of Conduct.
UNHCR (2020). Consequences of Underfunding in 2020. Technical report, UNHCR The
Refugee Agency, Geneva, Switzerland.
UNHCR(2022). UNHCR-PrivatesectordonatesoverUS$200milliontoUNHCR’sUkraine
emergency response.
United Nations Industrial Development Organization, U. (2002). UNIDO Business Partner-
ships for Industrial Development - Partnership Guide.
Urban Institute(2018). WhatisthePrivateSector’sRoleinRespondingtotheInternational
Refugee Crisis?
Wagner, T., Lutz, R. J., and Weitz, B. A. (2009). Corporate Hypocrisy: Overcoming the
ThreatofInconsistentCorporateSocialResponsibilityPerceptions. Journal of Marketing,
73(6):77–91.
Wainwright, O. (2017). Why Ikea’s flatpack refugee shelter won design of the year. The
Guardian.
Wang, L. and Juslin, H. (2012). Values and Corporate Social Responsibility Perceptions of
Chinese University Students. Journal of Academic Ethics, 10(1):57–82.
Weiss, T. G. (2013). Humanitarian Business. Polity, Cambridge. OCLC: ocn812687264.
142
White House, . (2016a). Fact Sheet: White House Announces Commitments to the Call to
Action for Private Sector Engagement on the Global Refugee Crisis.
White House, . (2016b). FACT SHEET: White House Launches a Call to Action for Private
Sector Engagement on the Global Refugee Crisis| whitehouse.gov.
Wilson, R. K. and Sell, J. (1997). “Liar, Liar... ”: Cheap Talk and Reputation in Repeated
Public Goods Settings. Journal of Conflict Resolution , 41(5):695–717.
Witte, J. M. and Reinicke, W. H. (2005). Business Unusual: Facilitating United Nations
Reform through Partnerships.
Yamane, T. and Kaneko, S. (2021). What Motivates Stakeholders to Demand Corporate
Social Responsibility: A Survey Experiment. Sustainability, 13(15):8313.
Yoon, Y., G¨ urhan-Canli, Z., and Schwarz, N. (2006). The Effect of Corporate Social Re-
sponsibility (CSR) Activities on Companies With Bad Reputations. Journal of Consumer
Psychology, 16(4):377–390.
Zasuwa, G. (2018). Do consumers really care about organisational motives behind CSR?
The moderating role of trust in the company. Social Responsibility Journal, 15(8):977–
991. Publisher: Emerald Publishing Limited.
Zyck, S. and Armstrong, J. (2014). Humanitarian crises, emergency preparedness and re-
sponse: The role of business and the private sector: Jordan case study.
Zyck, S. and Kent, R. (2014). Humanitarian crises, emergency preparedness and response:
The role of business and the private sector: Final report.
143
Appendices
A Appendix to Chapter 2
A.1 Treatment Text Survey
The following figures present the treatment texts per treatment group.
Figure 5.1: Refugee Treatment Text in Survey
144
Figure 5.2: BLM Treatment Text in Survey
145
Figure 5.3: Climate Change Treatment Text in Survey
146
A.2 Summary Statistics for Treatments
Statistic N Mean St. Dev.
Women 517 0.511 0.541
Age under 35 517 0.350 0.477
White 517 0.714 0.452
Income under 40K 517 0.306 0.461
Bachelors or Higher 517 0.555 0.497
Democrat 517 0.646 0.479
Region West 517 0.199 0.400
Table 5.1: Summary Statistics for Refugee CSR Treatment
Statistic N Mean St. Dev.
Women 519 0.545 0.529
Age under 35 519 0.326 0.469
White 519 0.726 0.446
Income under 40K 519 0.331 0.471
Bachelors or Higher 519 0.574 0.495
Democrat 519 0.582 0.494
Region West 519 0.218 0.413
Table 5.2: Summary Statistics for CSR for the BLM movement
Statistic N Mean St. Dev.
Women 519 0.486 0.523
Age under 35 519 0.316 0.465
White 519 0.721 0.449
Income under 40K 519 0.316 0.465
Bachelors or Higher 519 0.595 0.491
Democrat 519 0.582 0.494
Region West 519 0.229 0.421
Table 5.3: Summary Statistics for CSR for the fight against climate change
147
A.3 Difference in Means Test: Overall and Different Groups
Overall Difference in Means Treatment Groups
The following table shows the ttests for the diff-in-means for the climate vs. refugees
support levels. As we can see, difference is not statically significant.
Test statistic df P value
Alternative
hypothesis
mean of x mean of y
-1.31 1033 0.1904 two.sided 7.888 8.141
Welch Two Sample t-test: Mean for Support Refugee Group
and Mean for Support Climate Group
Table 5.4: T-test: Diff-in-Means for Refugee and Climate Group (Level of Support)
Next,thedifferenceinmeansfortheBLMandrefugeegroupsispresented. Thedifference
here is statistically significant.
Test statistic df P value
Alternative
hypothesis
mean of x mean of y
7.819 985.9 1.363e-14 *** two.sided 7.888 6.21
Welch Two Sample t-test: Mean for Support Refugee Group
and Mean for Support BLM Group
Table 5.5: T-test: Diff-in-Means for Refugee and BLM Group (Level of Support)
Finally, the diff-in-means also reveals the statistically significant support levels for the
BLM and Climate Treatment groups:
Test statistic df P value
Alternative
hypothesis
mean of x mean of y
8.858 1002 3.643e-18 * * * two.sided 8.141 6.21
Welch Two Sample t-test: Mean for Support Climate Group
and Mean for Support BLM Group
Table 5.6: T-test: Diff-in-Means for Climate and BLM Group (Level of Support)
Difference in Means by Gender, Political Party, Age, Personal Responsibility,
and Skepticism
The following table presents the t-test for support levels by gender. The difference
between men and women is statistically significant.
148
Test statistic df P value
Alternative
hypothesis
mean of x mean of y
-5.121 475.1 4.417e-07 * * * two.sided 7.178 8.549
Welch Two Sample t-test: Mean for Support from Men
and Mean for Support from Women
Table 5.7: T-test: Diff-in-Means for Women vs. Men (Level of Support)
The following tables show the t-tests between political party identification. All results
are statistically significant.
Test statistic df P value
Alternative
hypothesis
mean of x mean of y
-5.408 79.57 6.501e-07 * * * two.sided 6.812 8.931
Welch Two Sample t-test: Mean for Support from Independents/others
and Mean for Support from Democrats
Table 5.8: T-test: Diff-in-Means for Independents vs. Democrats (Level of Support)
Test statistic df P value
Alternative
hypothesis
mean of x mean of y
2.696 139.1 0.007896 * * two.sided 6.812 5.538
Welch Two Sample t-test: Mean for Support from Independents/others
and Mean for Support from Republicans
Table 5.9: T-test: Diff-in-Means for Democrats vs. Republicans (Level of Support)
Test statistic df P value
Alternative
hypothesis
mean of x mean of y
-10.5 166.5 4.054e-20 * * * two.sided 5.538 8.931
Welch Two Sample t-test: Mean for Support from Republicans
and Mean for Support from Democrats
Table 5.10: T-test: Diff-in-Means for Independents vs. Republicans (Level of Support)
The following figure displays the results of a multiple comparison test between the vari-
ous age brackets (p-values with and without Bonferroni correction). The adjusted p-values
indicate that there is no statistically significant difference between the age groups.
149
Figure 5.4: Multiple Comparison with Bonferroni Correction: Age Brackets
The two table below presents the t-test for low vs. high attention to social issues in the
media as well as for low vs. high attention to companies’ social engagements. We see that
the differences are statistically significant.
Test statistic df P value
Alternative
hypothesis
mean of x mean of y
4.878 446.1 1.494e-06 * * * two.sided 8.484 7.178
Welch Two Sample t-test: Mean for Support for High Attention to Social
Issues in News and Mean for Low Attention to Social Issues in News
Table5.11: T-test: Diff-in-MeansforHighvs. LowAttentiontoSocialIssuesinNews(Level
of Support)
150
Test statistic df P value
Alternative
hypothesis
mean of x mean of y
11.53 251.9 5.4e-25 * * * two.sided 10.09 7.328
Welch Two Sample t-test: Mean for Support for High Attention to Companies’
Social Engagement and Mean for Low Attention to Companies’ Social
Engagement
Table 5.12: T-test: Diff-in-Means for High vs. Low Attention to Brands’ Social Engagement
(Level of Support)
Thetablebelowreportsthedifferencesbetweenlowvs. highskepticismaboutcompanies’
motives as to why they get involved in refugee protection and empowerment. As we can see,
the differences in means are statistically significant.
Test statistic df P value
Alternative
hypothesis
mean of x mean of y
-8.909 480.8 1.07e-17 * * * two.sided 6.806 9.024
Welch Two Sample t-test: Mean for Support for High skepticism
about companies’ motives and Mean for Low skepticism about
companies’ motives
Table 5.13: T-test: Diff-in-Means for High vs. Low Skepticism about Companies’ Motives
(Level of Support)
A.4 Multicollinearity Checks for Regression Models
Figure 5.5 reports multicollinearity between all independent variables in the regression mod-
els. As can be seen, multicollinearity is low to moderate, with all variables indicating no
more than a negative or positive correlation of 0.36. The binary/categorical variables were
recoded as follows for the correlation matrix: Gender (Male = 1, Female = 0), Bachelor or
higher (Yes = 1, No = 0), Political Identification (Republican = 1, Independent/other =
151
2, Democrat = 3), Under 35 (Yes = 1, No = 0), White (Yes = 1, No = 0), Know about
refugee crisis (Yes = 1 , No = 0), Attention to brands’ social engagement (Yes = 1, No = 0),
Attention to social issues in news (Yes = 1, No = 0), Skepticism about companies’ motives
(Yes = 1, No = 0).
Figure 5.5: Correlation Plot IVs
152
A.5 Additional Robustness Check: Logistic Regression
As an additional robustness check for my original ordered and OLS regression model, I
recode the DV into a binary variable (Support Yes/No), and run a logistic regression. As
the coefficient plot suggests, all results hold and are in the expected directions.
Figure 5.6: Logistic Regression
153
A.6 AdditionalFigurestoShowAnswerDistributionforHappiness
About CSR activities
This section exemplifies the answer distribution by different demographic factors for hap-
piness about CSR engagement to support and empower refugees (this is one facet of the
support variable).
Figure 5.7: Level of Happiness by Age
154
Figure 5.8: Level of Happiness by Education
Figure 5.9: Level of Happiness by Political Party
155
B Appendix to Chapter 3
B.1 Glimpse into Datasets
The following tables 5.15 and 5.14 present snapshots of the datasets that were used to create
the PPP network. The meta dataset contains all programs and program characteristics and
illuminates which actors are involved. From this dataset I can derive the edgelist dataset:
it is a 2-mode edgelist, which shows all connections between programs and actors. Put
differently, each row presents a program-actor observation.
Initiative Industry Engagement Actors Geo Focus Start
Development
of tablet-based educational content
Education Sharing capabilities BrainPOP, UNHCR Kenya, Malaysia 2013
Digital education for refugee
children
Technology Sharing capabilities BRCK Limited, Norwegian Refugee Council Kenya 2016
Refugee First Response Center Technology Sharing capabilities
Cisco, Lebanon’s Ministry of Health,
Beyond Association
Lebanon 2015
Citi Foundation Pathways to
Progress - Rescuing Futures
Financial Services Enabling employment Citi Foundation, International Rescue Committee Jordan, Nigeria 2017
Support for nonprofits serving
refugees
Retail Humanitarian assistance
Cotopaxi, International Rescue Committee,
Nothing But Nets
Middle East,
Sub-Saharan Africa,
Latin America, Europe
2014
Coursera for Refugees Education Sharing capabilities
Coursera, RefuSHE, Xavier Project,
Talent Beyond Boundaries,
Jesuit Refugee Services,
The Power of International Eduation, Upwardly
Global, Bilingual Education Institute,
Lutheran Familiy Services, kiron,
Syrian Youth Assembly, EQI,
US State Department, HIAS, Save the Children,
inzone, Tent Partnership for Refugees
Global 2016
Solar energy for refugees Energy Building a business d.light, Crossroads Foundation Uganda 2016
Bringing renewable energy to
Kakuma refugee camp
Energy Sharing capabilities
Energias de Portugal, UNHCR,
HELPIN
Kenya 2009
Cash assistance delivery Financial Services Extending services
Equity Bank, World Food Programme,
UNHCR
Kenya, Rwanda, Uganda 2012
Partnership with UNHCR for
sustainable and clean
energy for refugees
Energy Sharing capabilities
Eurelectric, Energias de Portugal, Engie,
Iberdrola, Enel,
A2A, UNHCR
Kenya 2017
Table 5.14: Snapshot Meta Data of PPP Programs
156
programs actors actor id program id
Alianza Shire acciona.org Foundation 100 300
Alianza Shire Signify 101 300
Alianza Shire Iberdrola 102 300
Alianza Shire
Technical University of Madrid’s Innovation & Technology for
Development Center (idtUPM)
103 300
Alianza Shire Spanish Agency for International Development Cooperation (AECID) 104 300
Alianza Shire European Union 105 300
Alianza Shire UNHCR 106 300
Social integration in Sanliurfa - Turkiye adidas 107 301
Social integration in Sanliurfa - Turkiye Turkey’s Ministry of Labor 108 301
Social integration in Sanliurfa - Turkiye Ministry of Youth and Sports 109 301
Social integration in Sanliurfa - Turkiye Gen¸ c Hayat 110 301
Shiriki Hub ARED 111 302
Shiriki Hub Rwanda Red Cross 112 302
Open Homes project Airbnb 113 303
Open Homes project Help Refugees 114 303
Open Homes project International Rescue Committee 115 303
Mobile communication services for
refugees
Airtel Uganda 116 304
Mobile communication services for
refugees
UNHCR 106 304
Humanitarian cash transfers Airtel Uganda 116 305
Humanitarian cash transfers Mercy Corps 117 305
Table 5.15: Snapshot 2-Mode Edgelist
B.2 2-Mode Network
The figure below displays the 2-mode refugee protection network consisting of programs
and actors. More specifically, the visualization shows how actors are connected through
programs. This 2-mode network was created using the the above discussed edgelist dataset.
From this network, I extracted an affiliation network that portrays actor-actor connections
based on shared programs (as displayed in chapter 3).
157
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121 122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177 178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275 276
277 278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368 369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404 405
406
407
Actor
Program
Figure 5.10: 2-Mode Refugee Protection Network
B.3 Correlation
InanefforttoensurethatmulticollinearityofmyIVsisnotamajorproblem,IrunPearson’s
Chi-squared tests and Cramer’s V for categorical variables. Here are the results for 1)
Geographic Region - Actor Type, 2) UNHCR Connection - Actor Type, and 3) Geographic
Region - UNHCR Connection. Overall, the correlation between the different variables is
rather weak.
business education foundation IGO NGO state
North 53 4 15 7 62 11
South 23 1 0 1 15 4
Table 5.16: Contingency Geographic Region - Actor Type
158
Pearson’s Chi-squared test::
χ 2
= 8.0228, df = 5, p-value = 0.155
Cramer’s V:
0.202318
Given that the p-value > 0.05, we fail to reject the null (independence). As we cannot
confidently rule out the null, it is sensible to assume that there is no correlation between
geographic region and actor type. Although, there appears to be no correlation, I report the
results of the Cramer’s V; even if there was correlation, it is weak.
Next is the correlation between the UNHCR connection and actor type variables.
business education foundation IGO NGO state
No Connection UNHCR 37 4 10 5 59 9
Connection UNHCR 39 1 5 3 19 6
Table 5.17: Contingency UNHCR Connection - Actor Type
Pearson’s Chi-squared test:
χ 2
= 12.786, df = 5, p-value = 0.02547
Cramer’s V:
0.2547617
Given that the p-value < 0.05, we can reject the null (independence). Being connected
to the UNHCR and Actor Type appear to be correlated. However, the results from the
Cramer’s V illuminate a rather weak positive correlation between the variables.
Finally, correlation between geographic region and being connected to the UNHCR are
reported below.
No Connection UNHCR Connection UNHCR
North 97 55
South 28 17
Table 5.18: Contingency UNHCR Connection - Geographic Region
Pearson’s Chi-squared test:
χ 2
= 0.00035281, df = 1, p-value = 0.985
159
Cramer’s V:
0.001338258
Given that the p-value < 0.05, we fail to reject the null (independence). We are not
confidenttoruleoutthenull,andthus,thereappearstobenocorrelationbetweengeographic
region and UNHCR connection. Turning to the Cramer’s V for an understanding of what
the magnitude of the correlation (if there was one) is, we also see that the correlation is
basically 0.
B.4 Goodness of Fit Model 4 and MCMC diagnostics
It is important to check the goodness-of-fit (GOF) of an ergm. The GOF of an ergm is
evaluated by simulating networks using the fitted parameters of our model and calcualting
a variety of structural measure from these graphs (i.e. degree distribution, edge-wide shared
partners, etc.). We can the compare the estimates of these simulated networks with those in
yourobservednetwork. Infigure5.11, weseetheboxplotsofthesimulatedcountsoverlayed
with our observed graph statistics. Overall, our model statistics fall nicely within those of
the simulated model statistics. Turning to figure 5.12, we see that our degree distribution
(solid black line) falls mostly within those of the simulated degree distributions (boxplots).
Finally, figure 5.13 illuminates edge-wide shared partners. While the gof of this metric is not
ideal, it is quite common to have divergent estimates for ESP between the observed and the
simulated networks. It is quite complicated for a model to capture the overall complexity
of real world networks. In order to get a better fit, I have run the ergm with different
dependence terms and various decay parameters. In future versions of this paper, I will
introduce more complex dependence terms to improve the fit for ESP. It could be the case
that future model version slightly improve model fit for ESP, but not significantly.
160
edges nodecov.conn_UNHCR gwdeg.fixed.0.5
0.0 0.2 0.4 0.6 0.8 1.0
model statistics
simulated quantiles
Figure 5.11: GOF: Model
0 6 13 21 29 37 45 53 61 69
0.00 0.05 0.10 0.15 0.20
degree
proportion of nodes
Figure 5.12: GOF: Degree
0 2 4 6 8 10 12 14 16 18
0.0 0.2 0.4 0.6 0.8
edge−wise shared partners
proportion of edges
Figure 5.13: GOF: ESP
161
In addition to model fit, it is necessary to evaluate the quality of the MCMC simulation
thatproducedthemodelestimatestoensurethatthemodelisreliable. Thefollowinggraphs
depict the sample statistics. Overall, the trace plots and density plots look exactly what we
would hope to see. The trace plot illuminates the difference between the sample statistics
and the observed network for every step of the simulation. In our case, they show evidence
of mixing (random variation at each step), centered around zero.
The density plots are all normally distributed, centered around zero (i.e. no difference
from the observed network). The sawthorne pattern in some of the plots is normal for
discrete variables that have a a small range.
162
163
B.5 Odd Ratios for Model 2 and 3
Tables 5.19 and 5.20 report odd ratios that correspond with the regression results of model 2
and3. SimilartotheoddratiosinTable3.2,educationalinstitutionsandIGOsareabout1.7
- 3 times more likely to form ties compared to businesses. Being connected to the UNHCR
increases the odds of forming ties by 13 percent. Finally, the similarity added in 5.19 aligns
with the reported odd ratios in model 4. Actors in the same geographic region partner up
more than those across regions (for Northern actors: 1.9 times increased odds; Southern
actors: 1.77 times higher odds).
164
Odds ratio 2.5 % 97.5 %
Edges 0.022 0.017 0.027
Education 1.709 1.257 2.322
Foundation 1.276 1.056 1.541
IGO 2.994 2.410 3.720
NGO 1.503 1.329 1.700
State 1.193 0.989 1.438
Conn.UNHCR 1.129 1.022 1.248
GWDegree 0.081 0.052 0.127
Table 5.19: Main Effects (Model 2) Odd Ratios
Odds ratio 2.5 % 97.5 %
Edges 0.011 0.008 0.015
Education 2.015 1.475 2.752
Foundation 1.303 1.070 1.588
IGO 3.330 2.673 4.147
NGO 1.378 1.221 1.555
State 1.356 1.110 1.657
Conn.NHCR 1.139 1.023 1.267
Match.Geo.North 1.920 1.594 2.312
Match.Geo.South 1.771 1.093 2.871
Match.ActorType 1.721 1.417 2.091
GWDegree 0.120 0.073 0.199
Table 5.20: Similarity (Model 3) Odd Ratios
165
C Appendix to Chapter 4
C.1 Sample Interview Questions
Partner Choice/Partnership Formation and Initiation
• Whatmadeyourorganization/businessdecidetobecomeengagedinsupporting/protecting
refugees and partnering up with other protection actors?
• Could you tell me a bit more about who decides to get involved who decides that a
refugee program is necessary, and who decides what kind of refugee program will be
constructed?
• Could you describe the steps/process that leads to the creation of a collaborative
refugee protection/support program?
• Who initiates these partnerships? Do you reach out to [public/private] actors or vice
versa?
• Which (type of) actors would you never partner up with? / Are there any partnership
dealbreakers, and if yes, which ones?
• Why do you partner up with the [for-profit sector/non-profit sector]?
Performance and Evaluation of Joint Programs
• In your experience, what factors make a program (more or less) successful?
• How do you evaluate/know if a program is or is not successful?
• How do you monitor the progress of these collaborative projects/programs? (Who is
in charge; and who reports to whom?)
• How do you report the results of the joint programs?
166
• When you find that a program/project isn’t performing as expected, what steps do
you take to address the shortcomings?
• Do you disagree with your partners when it comes to program tasks and implementa-
tion? If yes, what are the sources for the disagreement, and how do you resolve such
disagreements?
Structure of Partnerships/Programs
• How are the duties and responsibilities within the program distributed?
• How do you communicate with your partners, and how often?
• Do you have staff or a team devoted to refugee protection programs/initiatives?
• Is there a central/main actor who is the leader in the protection program, and if yes
who?
• Do you work with refugees or refugee-led organizations to develop the programs?
C.2 Example Recruitment Email for Interviews
The following text is a basic template that was used to email study participants to inquire
about an interview. The template was adjusted and personalized based on the recipient.
For example, it would mention the company/organization/department name and its exper-
tise/involvement in the refugee cause (with potential specifics about a program/project).
Email/Social Media (English)
[To Whom It May Concern/Name of email recipient],
I am a Ph.D. candidate in Political Science and International Relations at the University
of Southern California. For my dissertation research, I explore how public and private
actors cooperate to provide protection and services to refugees. Given your expertise and
167
engagement in refugee protection, I would like to ask you for an interview. The interview
will take approximately 30-60 minutes to complete.
If you are interested in participating in this study, please reply to this email. Your input
is highly valuable and important to the success of my research. If you have any questions
regarding the study or would like more detailed information, please feel free to contact me.
Thank you for your time and your consideration.
168
Abstract (if available)
Abstract
The international community faces a dilemma: while numbers of refugees are rapidly increasing, states and organizations have yet to find adequate solutions to address this crisis. A new trend is the active participation of the for-profit private sector consisting of businesses and multi-national corporations. While the business community has long been engaged in inter- national investment and philanthropic activities, their active corporation with public actors for the purpose of creating and executing refugee protection and empowerment programs, is rather novel. This dissertation project sets out to explore the emergence of cooperation between businesses, governments, and organizations: a public-private network that develops collaborative solutions to protect, support, and empower asylum seekers and refugees. It investigates why and how these public and private actors cooperate to tackle the global refugee crisis. Through a multi-method approach that combines social network analysis, content analysis of stakeholder interviews and public reports, and a survey experiment, it exposes motivations, factors, and overall dynamics that are at play when various actors find and form partnerships and evaluate joint projects and programs. It deconstructs the complex partnership formation and evaluation process and finds that factors such as geography, type of industry, goals, visibility, and reputation influence actors’ ability to join this network and cooperate. While the partnership creation and program evaluation processes are arduous, the resulting public-private partnerships often create innovative solutions for refugee situations and have the potential to revolutionize humanitarian program/project evaluation metrics. This dissertation concludes by offering a discussion on the future of cooperation in the refugee protection and empowerment space and laying out a research agenda beyond this project.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The IO contestation project: a multi-method approach towards understanding the ways states contest international organizations.
PDF
Pathways to nuclear weapon reversal: exploring mechanisms and understanding non-proliferation policy
PDF
Deciding modalities of global health governance: What facilitates or hinders public-private partnerships?
PDF
Goldilocks’ signal for security cooperation in East Asia: China’s rise, hedging, and joint military exercises
PDF
Cosmopolitan logics and limits: networked discourse, affect, and identity in responses to the Syrian refugee crisis
PDF
Selling virtue: how human rights NGOs and their donors work together to create a better world ... for themselves
PDF
Forced migration as a cause and consequence of conflict
PDF
Things fall apart: the unraveling of international institutions through withdrawal
PDF
Taking orders from Brussels? external actors and regional organizations in the developing world
PDF
By word, not deed: advancing a theory of rhetorical compliance in international law
PDF
Organized intraethnic conflict and cooperation in divided stateless nations: the competitive dynamics of rival ethnonationalist factions
PDF
Signaling identity: how race and gender shape what representatives say online
PDF
A theory of status-quo terrorism: democracies in conflict and their proclivity to outsource repression
PDF
The strategy of firms towards political elites and consumers in polarized western democracies
PDF
Enterprising citizenship: Mexican immigrant empowerment and public-private partnerships in Santa Ana, California
PDF
Legitimizing self-determination: advancing the sovereignty of separatist movements
PDF
Creating destruction: the political economy of zombie firms
PDF
Circuit breakers: how policy entrepreneurs interrupted the electric flow with Peru’s first renewable energy legislation for the grid
PDF
Making the global local: Sino-foreign cooperation universities, interdisciplinary study and 21st-century skills
PDF
The allocation and accessibility of aid: the role of recipient government agency in dictating donor decisions
Asset Metadata
Creator
Neumeier
(),
Stefanie Anna
(author)
Core Title
Refugee protection and empowerment joint venture: understanding how and why public and private actors cooperate to tackle the global refugee crisis
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Political Science and International Relations
Degree Conferral Date
2023-05
Publication Date
04/20/2023
Defense Date
02/06/2023
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
mixed-methods,OAI-PMH Harvest,public-private cooperation,Refugees
Format
theses
(aat)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Sandholtz , Wayne (
committee chair
), Garry, Hannah (
committee member
), James , Patrick (
committee member
)
Creator Email
sneumeie@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-oUC113056781
Unique identifier
UC113056781
Identifier
etd-NeumeierSt-11669.pdf (filename)
Legacy Identifier
etd-NeumeierSt-11669
Document Type
Dissertation
Format
theses (aat)
Rights
Neumeier, Stefanie Anna
Internet Media Type
application/pdf
Type
texts
Source
20230420-usctheses-batch-1026
(batch),
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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.
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
Repository Email
cisadmin@lib.usc.edu
Tags
mixed-methods
public-private cooperation