The political economy of higher education finance

GARRITZMANNTHE POLITICS OF TUITION FEES AND SUBSIDIES IN OECD COUNTRIES, 1945-2015 OF HIGHER EDUCATION FINANCE THE... The Political Economy of Higher Education Finance The Politics of

JULIAN L GARRITZMANN

THE POLITICS OF TUITION FEES AND SUBSIDIES

IN OECD COUNTRIES, 1945-2015

OF HIGHER EDUCATION FINANCE

THE

Education Finance

The Political

Economy of Higher Education Finance

The Politics of Tuition Fees and Subsidies in

OECD Countries, 1945–2015

ISBN 978-3-319-29912-9 ISBN 978-3-319-29913-6 (eBook) DOI 10.1007/978-3-319-29913-6

Library of Congress Control Number: 2016948423

© The Editor(s) (if applicable) and The Author(s) 2016

This work is subject to copyright All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifi cally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfi lms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed

The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a specifi c statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use The publisher, the authors and the editors are safe to assume that the advice and information

in this book are believed to be true and accurate at the date of publication Neither the lisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made

Printed on acid-free paper

This Palgrave Macmillan imprint is published by Springer Nature

The registered company is Springer International Publishing AG Switzerland

University of Konstanz

Department of Politics & Public Administration

Konstanz , Germany

Many people have helped me to start and fi nish this book First and most, I am deeply indebted to Marius Busemeyer Since we met at the Max Planck Institute in Cologne, Marius has supported me in many respects:

fore-he awakened my interest in education policy and tfore-he welfare state; fore-he mented on numerous versions of my papers; and he offered me convenient research positions with long time horizons (very unusual nowadays) and generous funding (even more unusual), meaning that throughout my time as a graduate student I had the means to attend conferences and workshops and was able to afford a research stay at Harvard University Moreover, Marius created a highly productive work environment at the University of Konstanz; he kept my teaching load low and hardly ever bothered me with organizational matters so that I could focus on my own research Maybe most importantly, Marius always pushed me further with critical comments and tough deadlines, but also left me a lot of academic freedom and accepted that I often had to fi nd my own way, sometimes disregarding good advice (of course, often regretting this later) In short, Marius was the ideal supervisor and, moreover, has become a good friend Christian Breunig has been the perfect second supervisor: always avail-able when I needed advice, but never trying to push me in a direction in which I didn’t want to head; always very clear in his critique and concen-trating on the big, critical questions Christian not only provided substan-tive advice but also pushed me to bring this book into a (hopefully) more easily accessible and shorter format, dropping many of the potentially interesting, but largely unnecessary, meanderings of the book

I am also enormously thankful to Torben Iversen I met Torben during two workshops in Bremen and Konstanz and was very impressed by his analytical precision and style of thinking about politics Right from the start he took a lot of time to have discussions with me and never treated

me as the little graduate student that I was I am very grateful that Torben invited me to Harvard, where he paved my way, connecting me to other exciting scholars Moreover, Torben agreed to serve as a supervisor for my dissertation and took the time to comment extensively on single papers and on the fi nal manuscript His view on my work—though I did not always follow it—has helped me to sharpen my analytical focus and sim-plify the argument as much as possible

Finally, I thank Dirk Leuffen, who agreed—on short notice—to chair

my dissertation defense committee, and I am grateful that he not only took the time to read through my lengthy dissertation but also to com-ment on it from an “outside angle,” which helped me to focus more on the main story

Moreover, I wish to thank our entire team at the University of Konstanz: Aurélien Abrassart, Yvonne Aymar, Margot Beier, Michael Dobbins, Ulrich Glassmann, Susanne Haastert, Susanne Münn, Erik Neimanns, Roula Nezi, Raphaela Schlicht-Schmälzle, and Janis Vossiek I have always enjoyed the critical but constructive discussions in our group I am also grateful for all of the support from our student assistants who saved me

a great deal of time by doing a lot of—sometimes annoying, but sary—work: Dana Behrens, Sophie Fendrich, Maximilian Gahntz, Caspar Kolster, Tobias Tober, Léonie Trick, Marie Zeller, and, most of all, Lina Seitzl, who has been a great support during almost my entire time at the University of Konstanz I also wish to thank Kilian Seng, Peter Selb, and Susumu Shikano for their statistical advice

During my time as a Visiting Fellow at Harvard, I experienced an mously inspiring, creative, and energizing environment Besides Torben Iversen, a number of colleagues commented on my work and stimulated

enor-my thinking in various ways: Daniel Ziblatt, Dan Smith, and Gwyneth McClendon let me participate in and present at the Research Workshop

in Comparative Politics, and had helpful comments Moreover, I am very thankful to Peter Hall and Kathy Thelen, who took time to discuss my ideas with me Furthermore, Jon Fiva, Jeff Frieden, Daniel Koss, John Marshall, Arthur Spirling, and Carina Schmitt commented on several papers Finally, many thanks to Dominik Geering and Olaf van Vliet During our joint lunches almost every day I not only received a lot of

feedback, but also made two great friends that I unfortunately see much too seldom

At the University of Cologne, where I graduated in 2011, I particularly wish to thank André Kaiser, who got me interested in comparative poli-tics in general and in parties and party competition in particular During

my time as a student assistant at his chair, he fundamentally shaped my perspective on political science and inspired me to think about the impor-tance of time in this fi eld I appreciate that I always fi nd the door open (and often a free desk at which I can do some work) when I come “back home.”

Moreover, many people have commented on parts of the book at ous stages: Sakari Ahola, Ben Ansell, Michael Braun, Margarita Gelepithis, Silja Häusermann, Anne-Marie Jeannet, Carsten Jensen, Jens Jungblut, Olli Kangas, Peter Maassen, Paul Marx, Traute Meyer, Stefan Thewissen, Pieter Vanhuysse, Wim van Oorschot, Peter Selb, Daniel Stegmüller, Christine Trampusch, Martina Vukasovic, and Claus Wendt Colleagues have also commented on the paper at various occasions such as at the CES Conferences in Amsterdam and Paris; the APSA Annual Meeting in Washington; the MPSA Conference in Chicago; the ESPAnet Meetings

vari-in Mannheim, Odense, and Oslo; Harvard’s various research workshops; the ECPR General Conference in Montréal; the HEIK seminar at the University of Oslo; the Cologne Center for Comparative Politics; and various occasions at the University of Konstanz Still, I’m pretty certain I have forgotten to mention someone, so I’m certain I owe drinks to some unmentioned but well-deserving friends, here’s a place where you can fi ll

in your name while patting yourself on the back: _ (the next beer is on me!)

Furthermore, I appreciate the generous funding I have received from several bodies, particularly from the German Research Foundation’s Emmy Noether Programme, the University of Konstanz’s Graduate School of Decision Sciences, and the German Excellence Initiative, which helped fi nance my stay at Harvard University

Finally, those who know me at least a little, know that I’m a big fan of Aristotle and Hannah Arendt And those who know Aristotle or Hannah Arendt a little, know about the importance of friendship to connect science with life, philosophy with political science, and society with one’s personal

eudaimonía Thus, a big shout out goes to all of my friends around the world who have helped me not to think about my work To name but a

few (and again I probably have to buy some drinks), Jan Sahm has always

been a true friend (in the Aristotelian sense)—always up for doing foolish things and for talking nonsense for hours, probably the most important thing one can do in today’s “knowledge economy”; moreover, I have had great scientifi c conversations with Leonce Röth, who always tries to chal-lenge everything I say—but (therefore!?) at the same time has been a great friend over the years In addition, Moritz Bassler, Matthias Klöpfer, Kilian Seng, and Jens Winkler were not only excellent musicians to play jazz with but also became close friends in Konstanz Robi Chattopadhyay proved, among other things, that there are fewer people who know how to listen

to good music than how to play it And during the many hours of practice and trips to away games across Germany, my fl oorball team made sure I physically did not have the breath to concentrate on any scholarly work Thanks, everyone!

Last, but actually fi rst, luck had it that I met Susanne B. Haastert ing the stressful wrap-up phase of fi nishing my dissertation I am extremely happy and grateful that ever since I have had such a smart and critical col-league, understanding friend, and loving partner at my side!

Julian L. Garritzmann Konstanz, December 2015

1 The Politics of Higher Education Tuition Fees and 

2 The Four Worlds of Student Finance: A Comparative

Descriptive Overview of Tuition Fees and Subsidies in 

3 Adding “Some Flesh to the Bones”: Illustrative Case

Studies of Four Diverse Cases Over Seven Decades 99

4 What Do Parties Want? Parties’ Positions and Issue

Emphasis on Tuition Fees and Subsidies 209

5 Testing the Time-Sensitive Partisan Theory in Large-n

Analyses: Parties’ Impact on the Tuition-Subsidy

Regimes of 21 Democracies Over Time 237

6 Individual-Level Attitudes Towards Subsidies: How

Positive Feedback-Effects Prevent (Radical) Change

in the Four Worlds of Student Finance 267

7 Conclusion and Outlook 301

high- subsidy (lower panel) regimes, by country

Fig 2.11 Cluster analysis including only signifi cantly discriminating

Fig 3.4 Higher education enrollment levels in Japan (1970–2010) 124

spending over conservative and non-Christian center

institutions compared with the consumer price

fully privately funded (7) higher education, averaged

21 democracies between 1995 and 2010: TSCS,

Table 5.4 Effects of government composition on subsidies in

21 democracies between 1995 and 2010: TSCS,

democracies between 1995 and 2010: TSCS regressions,

21 democracies between 1995 and 2010: TSCS regressions,

attitudes towards subsidies from the “keep signifi cant only”

model (Model 3, Table 6.1): multi-level ordered logit models,

Table 6.A Correlations of dependent and independent variables and

a substantial governmental grant to cover living costs during her studies

We can also imagine Hannah studied in Germany, where no tuition fees would be charged either; presumably, however, she also would not receive any fi nancial student aid Finally, consider that Hannah attended a typical Japanese university where the tuition amounts would resemble those of the USA; in contrast to the USA, however, public grants and subsidized loans would be hardly available

Why do students in some countries pay tremendously high tuition fees while students in other countries study free of charge? Why do some coun-tries offer grants, scholarships, and subsidized loans while such subsidies are non-existent elsewhere? What are the (re-)distributional consequences

of these different tuition-subsidy systems? This book provides answers to these and related questions by analyzing the political economy of higher education tuition fees and subsidies 1

More specifi cally, this book poses and answers three research questions

that build on one another Firstly, how do the tuition-subsidy systems differ

across the advanced democracies? As a systematic comparative descriptive

The Politics of Higher Education Tuition

Fees and Subsidies

overview on the variety of tuition-subsidy regimes across the advanced democracies is still absent in the literature, I compiled a huge compara-tive dataset on the tuition-subsidy systems in 33 advanced democracies (OECD countries, i.e the Organisation for Economic Co-operation and Development) to provide such an overview Chapter 2 summarizes these data, which comprise information on more than 70 characteristics of the respective tuition-subsidy systems The main takeaway from the descrip-

tive overview is that the advanced democracies fall into Four Worlds of

Student Finance In some countries (mainly continental Europe) tuition

fees are low, but fi nancial student aid is also largely non-existent: the

low-tuition–low-subsidy regime A second group (comprising Nordic Europe)

is characterized by the absence of tuition fees but very generous public subsidies: the low-tuition–high- subsidy regime The Anglo-Saxon coun-

tries form a third group, the high- tuition–high-subsidy regime , where

most students are charged considerable tuition amounts but also often receive public grants or publicly subsidized student loans Finally, we also

fi nd the—prima facie potentially most surprising—combination of high tuition fees accompanied by sparse public subsidies in Japan, Korea, and

other Asian countries, as well as some Latin American countries: the

high-tuition–low-subsidy regime Figure  1.1 depicts the Four Worlds of Student Finance with two exemplary variables (cf Chap 2 for details)

The second—and major—aim of this book is to explain the variety of

the Four Worlds of Student Finance Why do countries’ higher education

fi nance systems differ so considerably? This question is particularly puzzling,

because when one goes back to the immediate post-World War II (WWII) period, all of these countries’ tuition-subsidy systems looked very much alike 2 : systematic public subsidies were non-existent in all countries and tuition was comparably low (Chap 3 ; cf also Eicher 1998 ; Nakata and Mosk 1987 ) Moreover, enrollment levels were also very low, as barely

5 % of each age cohort enrolled in higher education (Trow 1972 ; Windolf

1997 ) In other words, the higher education systems of countries such

as Germany, Sweden, the UK, the USA, Japan, or France were all highly similar in the mid-twentieth century The main question of this book,

therefore, is how can we explain the origin of the Four Worlds of Student

Finance? How and why have countries developed from a low-tuition–low- subsidy regime in the immediate post-WWII decade to today’s four highly distinct regimes?

The main argument of this book is—challenging a literature that so far has almost exclusively focused on structural socio-economic explanations

(see Sect.  1.2 )—that the main reason for countries’ diverse developments lies

in politics More specifi cally, I use partisan hypothesis (Hibbs 1977 ; Alt

1985 ; Tufte 1978 ) as the starting point, arguing that the partisan position of government is crucial for the development of the advanced democracies’ tuition-subsidy regimes Yet, as I will demonstrate, partisan theory alone cannot explain the full variety of the Four Worlds of Student Finance, rather only two of the four regimes: the constant predominance

com-of conservative parties can explain the high-tuition–low- subsidy regime (e.g., Japan), while the stable predominance of left-wing parties explains the low-tuition–high-subsidy cluster (e.g., Sweden) Yet, in order to under-stand the origin of the other two regimes (low-tuition–low-subsidy as, for example, in Germany, and high-tuition–high-subsidy as, for example, in the USA), we need to go beyond partisan hypothesis in its “simple” form Drawing on key contributions in the recent historical institutionalist literature (see Pierson 1993 , 2004 ), I extend simple partisan hypothesis

by arguing that we need to take time and timing seriously, particularly the

Fig 1.1 The Four Worlds of Student Finance ( Source : Author’s compilation, cf

sequencing and duration of parties in offi ce I develop a Time-Sensitive

Partisan Theory and claim that only a “time-sensitive” analysis of the

parti-san composition of government can explain why countries’ tuition-subsidy tems have developed from a highly similar starting point in the mid-twentieth century to today’s Four Worlds of Student Finance Section  1.3 develops the argument more fully and derives several empirically testable implica-tions that the empirical chapters of this book (Chaps 3 4 5 , and 6 ) then engage in testing, combining a broad variety of quantitative and quali-tative analyses in a multi-method approach (Lieberman 2005 ; Rohlfi ng

sys-2008 ) The main theoretical contribution of the book is thus the refi ment of a major theory of comparative politics that can also be applied to other policy areas (as outlined in Chap 7 )

The third and fi nal question this book seeks to answer arises from the fact that during roughly the last 20  years, hardly any country has wit-nessed radical policy change That is, no country changed from one tuition- subsidy regime to another 3 In fact, it seems to be the case that the advanced democracies’ tuition-subsidy systems were largely formed dur-ing the fi rst four post-war decades, while any subsequent developments have mainly continued along countries’ respective regime paths: tuition kept increasing in high-tuition countries, while low-tuition countries have remained tuition-free; likewise, high-subsidy countries have kept subsidies

at a high level or extended these even further, while low-subsidy countries have hardly made attempts to install any serious public fi nancial support (cf Chap 2 for empirical data) Thus, the question arises, why is it the case

that countries do not change their tuition-subsidy systems anymore? Why are the Four Worlds of Student Finance so immune to radical policy change?

The argument I bring forward to explain this empirical puzzle is grounded in recent work on “positive feedback-effects” and “path depen-dencies” (see Pierson 1993 , 2000a , 2004 ; Campbell 2012 ; Kumlin and Stadelmann-Steffen 2014 ) In Sect.  1.3 , I claim theoretically—and dem-onstrate empirically in Chap 6 —that radical policy change has become

increasingly diffi cult for parties, because the existing tuition-subsidy systems

generate their own support by shaping people’s attitudes, bringing them in

line with the status quo via positive feedback-effects: people support those characteristics of the higher education systems that they themselves have experienced Consequently, radical change from the respective regime paths becomes increasingly electorally costly for political parties, thereby limiting governments’ leeway in departing from the respective regime paths

In less abstract terms, I argue that individuals who have (not) paid tuition fees for their own education will also expect others (not) to pay for their own education, thereby reinforcing the high-tuition (low-tuition) path Similarly, individuals who have (not) received generous subsidies will support that other students also (do not) receive these benefi ts, thereby strengthening the high-subsidy (low-subsidy) path I show in Sect.  1.3 that this follows from both a rational choice perspective and from sociological approaches Taken together, I argue that positive feedback-effects become stronger over time as the group affected by higher education increases and

as the topic increasingly gains political salience This sets high incentives for governments to follow further along the embarked upon paths because departure from the respective tuition-subsidy paths becomes electorally more costly Consequently, the Four Worlds of Student Finance are solidi-

Chapters 4 , 5 , and 6 continue the analyses in a quantitative large-n ting Chapter 4 provides the empirical groundwork by studying parties’ positions on higher education fi nance in order to test whether parties hold distinct positions on higher education fi nance in the fi rst place, whether these vary along ideological lines, and whether parties alter their positions over time due to positive feedback-effects The analyses utilize data from

set-a recent expert survey on pset-arty positions (Rohrschneider set-and Whitefi eld

2012 ) and my own codings of all manifestos of the British parties over four decades The fi ndings offer strong support for the Time-Sensitive Partisan Theory and provide the starting point for Chap 5 , which applies cross- sectional and time-series–cross-sectional (TSCS) regressions to all publicly available data on tuition-subsidy systems, testing whether the Time- Sensitive Partisan Theory also holds in a large-n setting over time

The quantitative results underpin the qualitative fi ndings and put them

on a broader basis by demonstrating that the partisan composition, and particularly the sequencing and duration of parties in offi ce can explain the Four Worlds of Student Finance

The fi nal empirical chapter (Chap 6 ) applies multi-level regressions to public opinion survey data for 21 countries over two decades to probe the arguments about positive feedback-effects, i.e., that the existing tuition- subsidy regimes feed back on individual preferences, making the systems increasingly immune to radical policy change Again, the empirical data are supportive, showing that voters of left-wing and right-wing parties hold different preferences on higher education policies and that positive feedback-effects seem to be at work, as respondents in countries with generous (expanding) subsidy systems are more supportive of expanding these subsidies even further

In summary, by bringing together arguments from party politics, fare state analysis, political economy, and public opinion research, routed

wel-in a meta-theoretical combwel-ination of rational choice and historical wel-tutionalism, and by testing these arguments in a multi-method design drawing on a large variety of empirical material, this book offers an expla-nation of the political economy of tuition fees and subsidies The remain-der of this chapter discusses the existing literature (Sect.  1.2 ), introduces

insti-my Time-Sensitive Partisan Theory in greater length (Sect.  1.3 ), sketches some important socio-economic consequences of the tuition-subsidy sys-tems to underpin the socio-economic and political relevance of tuition fees and subsidies (Sect.  1.4 ), and presents an outline of the book (Sect.  1.5 )

It should be emphasized that while I believe it makes sense to read the chapters in the order of presentation, each chapter also stands alone, so readers can jump back and forth or skip over chapters if they wish

1.2 LITERATURE REVIEW: EXISTING EXPLANATIONS

OF TUITION FEES AND SUBSIDIES After a long period of neglect in political science, education systems and education policies have recently experienced a massive increase in scholarly attention—often under the headings of “skill formation” or “human capi-tal investment” (Busemeyer and Trampusch 2011 ; Iversen and Stephens

2008 ; Jakobi et  al 2009 ) Core contributions to the fi eld have sized the fundamental importance of patterns of skill formation: most

empha-prominently, the Varieties of Capitalism (VoC) approach (Hall and Soskice

2001 ) places skill formation at the core of politico-economic systems by demonstrating that production regimes and welfare states, patterns of inequality and redistribution, and effects of structural changes such as globalization or skill-biased technological change cannot be understood without taking into account the design of the education systems (Estevez- Abe et al 2001 ; Iversen 2005 )

Scholars in the VoC tradition have, however, placed particular emphasis

on vocational education and training (VET) (Busemeyer 2009b , 2012 ; Busemeyer and Trampusch 2012a , b ; Crouch et  al 2004 ; Estevez-Abe

et  al 2001 ; Thelen 2004 ); higher education features much less

promi-nently in this literature (cf also Ansell 2008 , 2010 ) The same holds true for the recently evolving “social investment” literature (see Bonoli

2007 ; Esping-Andersen 2002 ; Hemerijck 2013 ; Morel et  al 2012 ), which places education and care systems at the core of the analytical and

political agenda: here, the focus lies almost entirely on pre-primary tion, further education, and re-skilling via Active Labor Market Policies (ALMPs)— higher education, in contrast, is largely neglected in the social

educa-investment literature Thus, despite the novel prominence of education systems and policies in political science, we still know surprisingly little about one key element of skill formation systems, namely higher educa-tion More specifi cally, we know even less about higher education tuition fees and subsidies

Why do students in some countries pay tremendous tuition fees while others study free of charge? Why do students in some countries receive generous fi nancial student aid while students in other countries remain empty-handed? In the following sections I fi rst present an overview of existing explanations of tuition fees and then turn to the literature on sub-sidies A “unifi ed explanation” of tuition fees and subsidies as intertwined phenomena has been hitherto non-existent 5

1.2.1 What Explains Tuition Fees? From Structural to Politico-

Economic Accounts

The existing explanations of tuition fees can be grouped into two

catego-ries The fi rst group seeks to explain tuition fees by focusing on structural

economic factors Within this group I distinguish the “Functionalists,”

“Ineffi ciency Critics,” “Baumolists,” and “Cost Sharers” and discuss each

of these in turn The second group takes political factors into account and

is discussed subsequently

1.2.1.1 Structural Economic Explanations of Tuition Fees

The fi rst explanation of tuition fees—and the one most visible in the literature—argues that they are a necessary corollary of fi scal austerity

(Johnstone 2003 , 2009 , 2011 ; Johnstone and Marcucci 2010 ; Jongbloed

2004 ; Marcucci and Johnstone 2007 ; Sanyal and Johnstone 2011 ; Vossensteyn 2009 ): a decline in public revenues is argued to lead to rising

fi nancial diffi culties for governments, resulting in a “sheer need for other than governmental revenue” (Johnstone 2003 : 353, 2009 ) Other struc-tural factors are argued to intensify this pressure, such as growing public demand for higher education, mounting per-student costs, a shift towards the knowledge economy, and intensifi ed globalization-induced economic competition (Johnstone and Marcucci 2010 ) In order to cope with these pressures, policy-makers are assumed to introduce or raise tuition fees to obtain additional revenues As this argument focuses on a single structural factor (fi scal austerity) and assumes a deterministic relationship, I label this the “Functionalistic Explanation.”

The Functionalist Explanation struggles with at least three facts Firstly,

it cannot explain the variation of the Four Worlds of Student Finance

It fails to explain, for example, why Finland, Sweden, or Germany are still tuition-free although their economies have been affected by severe downturns Secondly, temporal variation cannot be explained by the

Functionalists, i.e., the timing of the introduction of fees For example, why had Japan introduced fees by the mid-twentieth century (despite a grow-ing economy), while other countries have not pursued any such measures despite severe fi scal constraints? Thirdly, the Functionalist Explanation lacks a micro-mechanism explaining how austerity is believed to transmit into policies, because governments are treated as a mere transmission belt that transfers socio-economic pressure automatically into policies In fact, Johnstone and Marcucci explicitly claim that “we believe that the factors most directly affecting the fi nancing of higher education [i.e fi scal aus-terity] … are beyond politics and ideologies” (Johnstone and Marcucci

2010 : 24f) One does not have to be a political scientist, I believe, to fi nd this model at least somewhat simplistic

A second group of studies attributes the origin of tuition fees to the presumed ineffi ciency of higher education institutions: the “Ineffi ciency Critics.” The Ineffi ciency Critics (e.g., Vedder 2004 ; Brandon 2010 ) fea-ture less prominently in the scientifi c literature, but are very popular among

a more general audience and therefore deserve a short discussion 6 They

claim that tuition fees exist because higher education institutions are

inef-fi ciently organized, unnecessarily wasting money (e.g., on unproductive staff, luxurious student housing, marketing, etc.), which again is believed

to be a consequence of “lacking market-imposed discipline to economize” (ibid) In this vein, Vedder ( 2004 ) and colleagues consequently recom-mend more competition between higher education institutions as a means

of decreasing tuition fees

Again, almost the same criticism applies to the Ineffi ciency Critics as

to the Functionalists Firstly, this approach fails to explain the existing country variation, unless it is willing to assume that higher education institutions in, for example, Germany or Denmark are much more effi -cient than institutions in the USA or Canada—which does not seem to

be a very plausible assumption Moreover, the Ineffi ciency Critics cannot explain the timing of tuition policies: why, for example, did policy-makers

in Germany institute tuition fees in 2005 but abolished them again soon after? It does not seem convincing to attribute this to higher education governance structures Finally, other scholars (e.g., Ehrenberg 2000 ) have questioned the direction of causality, arguing that fi erce competition between higher education institutions leads to an “arms race” between colleges, increasing tuition even further

A third structural explanation can be termed the “Baumolist Approach.” The Baumolists (e.g., Archibald and Feldman 2011 ) apply Baumol and Bowen’s ( 1966 ) “cost disease” argument—that labor productivity does not increase uniformly across industries but lags behind particularly in the non-tradable, labor-intense service sector—to higher education They claim that higher education is not special in its rise in (private) costs; rather, higher education is just like any other high-skilled labor-intense industry (e.g., dentistry, legal services): as the productivity of higher edu-cation can hardly increase and as educated personnel become increasingly expensive, the costs of higher education necessarily increase The growing costs are then argued to be passed to consumers via higher fees That is, according to the Baumolists, tuition fees can simply be explained by the fact that higher education is a typical high-skilled labor-intense service industry (Archibald and Feldman 2011 )

While the argument about a “cost disease” in higher education is tainly convincing, the second conclusion—that this necessarily translates into higher tuition levels—does not automatically follow, as other policy

cer-solutions (e.g., larger public spending) are also possible Accordingly, like

the Functionalists and the Ineffi ciency Critics, the Baumolists struggle

to cope with cross-national and cross-temporal variation: for example, is there any reason to believe that the “cost disease” is more pressing in the USA than in Sweden, or why is it that Sweden has not installed tuition fees but tuition in the USA has skyrocketed?

A fi nal structural explanation points at socio-cultural changes in the perception of higher education; this explanation argues that higher educa-

tion is increasingly not seen as a public good anymore (benefi tting ety in large), but as a private good (benefi tting the students themselves)

soci-Consequently—they argue—students are demanded to contribute to their own education, as they are likely to benefi t from higher wages, better work conditions, longer life expectancies, and so on (Chevaillier and Eicher

2002 ; Barr 2004 ; Johnstone 2004 , 2009a ; Hauptman 2006 ; Jongbloed

2004 ; Lepori et al 2007 ; Salmi and Hauptman 2006 ; Vossensteyn 2004 ,

2009 ) Johnstone ( 1986 ) coined and publicized the euphemistic term

“cost sharing” for this reasoning This discourse has become socially and politically extremely relevant because the World Bank and other inter-national institutions (such as the European Union [EU] Commission) have picked up these arguments and engaged intensively in policy advice for developed and developing countries to establish “cost-sharing” poli-cies (e.g., Jongbloed et  al 2008 ) Yet, the Cost Sharers’ argument has the same pitfalls as the other three structural explanations: neither cross- national nor temporal variation can be explained with this approach, as it does not seem plausible that this discourse was prominent in Japan and the USA by the 1950s but has never reached the Scandinavian or Continental European countries 7

In summary, while it is certainly true that fi scal austerity challenges ernments’ budgets, higher education institutions could be more effi cient

gov-fi nancially, any high-skill labor-intense industry is increasingly expensive, and in times of fi scal austerity policy-makers might consider tuition fees

as a potential additional fund, the four structural approaches can neither explain the cross-country variation of the Four Worlds of Student Finance nor make sense of the timing of the introduction of, rises in, or abolish-ment of tuition fees Moreover, as these explanations focus on structural factors alone and essentially disregard (political) actors, they all lack plau-sible mechanisms of how the structural forces become policies Therefore,

I now turn to more elaborate approaches that take political factors into account

1.2.1.2 Existing Political Economy Explanations of Tuition Fees

A second group of explanations of the existence of tuition fees assigns

a crucial role to political factors and/or the interaction of political and economic factors Wolf and Zohlnhöfer ( 2009 ) and Wolf ( 2009 ) were among the fi rst to focus on political determinants of private (higher) edu-cation spending However, they do not advance any particular theoretical argument, but rather test whether the “established theories of compara-tive public policy” (Wolf and Zohlnhöfer 2009 : 242) have explanatory power regarding private education spending Consequently, their analysis incorporates almost 20 potential politico-economic determinants (ibid) Empirically, Wolf and Zohlnhöfer apply cross-sectional ordinary least squares regressions to spending data from 26 countries (21 for higher

education) in 2002 Regarding higher education, they fi nd that many

fac-tors play a role: the level of private education spending is higher the lower the level of public disbursement, the higher the degree of federalism, the lower the share of Catholics in the country, and the higher the levels of tertiary enrollment Interestingly, and at odds with the argument I pro-mote in this book, Wolf and Zohlnhöfer do not fi nd any robust effects of governing parties

Wolf and Zohlnhöfer’s contribution is a good starting point as it combines economic and political factors Yet, a couple of shortcomings

remain unaddressed Firstly, the authors utilize all types of private (higher)

education spending as the dependent variable, thereby not only ing money transferred from private households to (higher) education institutions (via tuition fees) but also contributions of “fi rms, founda-tions and other private entities” (ibid.: 231) However, as donations from companies and foundations might follow entirely different dynamics than tuition fees (e.g., Busemeyer and Trampusch 2012b ), the analysis—strictly speaking—does not tell us much about tuition fees Secondly, as Wolf and Zohlnhöfer only use data for a single year that is (2002), it is impossible

includ-to draw causal conclusions from the results This is particularly worrisome because the explanatory variables “refer to the same year as those for the dependent variable” (ibid.: 238, Table 1), i.e., they are not lagged as is otherwise customary in the spending literature Thus, many crucial ques-tions remain unanswered

Förster ( 2012 ) proposes a second politico-economic explanation of tuition fees: he argues that the creation of the European Higher Education Area has led to competition among EU member states, which might

have led to policy convergence towards a common tuition-fee model because policy-makers are assumed to carefully observe and potentially follow other countries’ (higher education) policies (Förster 2012 : 107f) Empirically, Förster investigates tuition fee policies in three—somewhat arbitrarily selected—cases (England, Austria, Germany) in the period after

the Bologna reforms (1999–2009) Not surprisingly, he fi nds no

conver-gence regarding tuition fees; rather, these countries drifted further apart

as England increased fees, while fees remained low or were abolished in Germany and Austria (Förster 2012 : 200ff.; see also Sect 3.8 for the German case and Sects 3.10.2 and 4.4 for the British case)

Förster ( 2012 : 204) ascribes this non-convergence of tuition policies

to different “logics of appropriateness” (March and Olsen 2008 ) in ferent welfare regimes (for similar arguments see Dolenec 2006 ; Pechar and Andres 2011 : 32) While tuition fees seem to fi t well into the liberal

dif-welfare model (as they foster market mechanisms and individual

respon-sibility), Förster believes they do not fi t into conservative welfare states,

where “other sentiments apply about the appropriate burden to be placed

on individuals” (ibid.: 203, Author’s translation) Moreover, he claims that tuition fees are a more confl ictual political topic in conservative wel-fare states, while he believes it is non-confl ictual in liberal welfare states (Förster 2012 : 204) Accordingly, the partisan composition of govern-ment is argued to matter for tuition policies in Germany and Austria but not in England

Förster deserves credit for his attempt to bring party politics and fare states into the analysis of tuition fees Yet, several questions put his argument and empirical analysis into doubt Firstly, it remains unclear why welfare regimes and party politics should be important in some cir-cumstances but not in others For example, why should parties matter in Germany and Austria but not in England? Related to this, it is empiri-cally simply not true that partisan confl ict over tuition is absent in liberal welfare states: Förster overlooks—because he focuses only on the very recent decade—the fact that the topic was strongly debated between par-ties before this period (my analysis of party programs in Britain in Sect 4.4 underpins this empirically; cf also Sect 3.9 on the USA) and that

wel-Labour and the Conservatives still disagree on the type of tuition fees

(e.g., ex ante up-front tuition vs ex post income-contingent) Secondly,

as Förster only investigates cases from two welfare state regimes (liberal and conservative), it remains unclear how one would explain the full

universe of cases, i.e., the Four Worlds of Student Finance, with this approach Finally, I argue and demonstrate here that the focus on the recent 10  years (1999–2009) is misleading because it misses the fact that the Four Worlds of Student Finance have been shaped during much earlier decades This makes Förster’s argument particularly questionable, because—as I show—the welfare state regimes and the tuition-subsidy regimes were essentially formed at the same time by similar actors (cf also Busemeyer 2015 for vocational and secondary education), making

it diffi cult to use one (the welfare regime) as an explanation of the other (the education system)

A third political economy explanation was proposed by Busemeyer and Iversen ( 2014 ), who argued that electoral systems shape countries’ respective relationships between public and private education spending The logical backbone of this argument is that electoral institutions are believed to affect the likelihood of certain political coalitions (Iversen and Soskice 2006 ) Busemeyer and Iversen ( 2014 ) claim that the resulting coalitions can determine whether countries develop mainly publicly or privately fi nanced education systems Proportional representation (PR) is held to coalitions of low- and middle-income groups more likely (Iversen and Soskice 2006 ), and it is argued that these coalitions establish a pub-

licly fi nanced education system because they can impose the lion’s share

of costs on the highest income group Majoritarian voting systems, in contrast, are assumed to produce coalitions of middle- and high-income groups (Iversen and Soskice 2006 ), which are theorized to establish a

mainly privately funded education system because the danger exists that

the center-left party deviates from the median voter’s interest, also ing higher costs on the middle class Therefore, the middle class opts for

impos-a coimpos-alition with the higher-income group impos-and prefers low timpos-axes impos-and vate provision of education In summary, this theory states that tuition fees should be found in majoritarian systems, while PR systems should be tuition-free Empirically, Busemeyer and Iversen applied TSCS regressions

pri-to spending data for 19 countries between 1997 and 2008 They report strong support for their argument, as the electoral system variable shows a robust effect: countries with majoritarian voting systems have signifi cantly lower proportions of public vis-à-vis private spending

To date, Busemeyer and Iversen’s analysis is the most theoretically and empirically advanced explanation of private (higher) education spending Nevertheless, a few crucial questions remain unanswered (partly because

they analyze total education spending and not specifi cally higher tion spending) Firstly, the argument does not offer an explanation for

educa-variation within electoral systems: for example, it should not be neglected

that large differences exist between the Nordic higher education fi nance system (characterized by low tuition but high subsidies) and that in the Continental European countries (combining low tuition with low subsi-dies; see Chap 2 ), despite the fact that both use PR systems Moreover,

some countries that do use majoritarian systems have not installed any

considerable tuition fees (e.g., France), while some PR countries have been highly tuition dependent (e.g., Japan, Chile, or South Korea) These crucial cases cannot be explained by Busemeyer and Iversen’s model Secondly, despite the model’s parsimonious elegance, the clear-cut dichot-omy between majoritarian systems, on one hand, and PR systems, on the other, seems somewhat stylized as many countries actually use much more complex systems, leading to more complex coalition dynamics Busemeyer and Iversen even acknowledge this point, as they add a third type (modi-

fi ed PR/mixed systems) in the empirical analysis, but do not explicate whether and how this would affect the theoretical logic of the model or its results Thirdly, Busemeyer and Iversen apply TSCS regressions despite the fact that—as they admit ( 2014 : 17)—electoral systems very rarely change over time Thus, a simple cross-sectional or difference-in-difference design would have been much more convincing Using TSCS is problematic in this case because it remains unclear whether the TSCS results are simply driven by the fact that the variables are time-invariant and, therefore, the number of cases is artifi cially infl ated The authors try to address this by presenting several model specifi cations, but none of these can really cir-cumvent this weakness

In summary, while both economists and political scientists have oped several explanations of tuition fees/private education spending and point towards several potential factors, no previous approach could make sense of the variation of the Four Worlds of Student Finance and explain the timing of their historical origin

devel-1.2.2 What Explains Public Subsidies? Public Policy Theories and “Simple” and “Conditional” Partisan Hypotheses

The previous section presented an overview on the existing literature on private education spending (i.e., tuition fees) Before presenting my own argument in Sect.  1.3 , I summarize existing studies on public subsidies (i.e., fi nancial aid to higher education students) This can, however, be ful-

fi lled quickly because, to my knowledge, hitherto no publication has aimed

to explain the variation in public fi nancial student aid across countries or

time Some country-specifi c explanations exist, particularly on the USA and

variation among its states (Doyle 2012 ; Hearn et al 1996 ; Lowry 2001 ), but comparative work across countries or time is lacking—presumably because the issue has been perceived as too specifi c, not salient enough, less interesting, or less (re-)distributive than other spending areas As I argue in Sects.  1.3 and 1.4 , however, subsidies are a crucial component of higher education systems In fact, I argue that one can neither understand the emergence and sustainability of the different higher education systems nor the political economy of higher education without considering public subsidies as a major component of the explanation

While research on the determinants of public subsidies is lacking, a by

now broad strand in the literature investigates determinants of public

edu-cation expenditure in general As this literature can serve as a good starting

point to develop hypotheses on public spending on subsidies, I discuss the results of this related literature here

Several factors have been found to infl uence public education ing: it is higher in more affl uent economies, in countries with a higher (demographic) demand for education, and in countries with larger female labor force participation rates, and it is lower when public debt increases (Busemeyer 2006 , 2009a ) Besides these economic factors, by far the most

spend-attention in the literature has been paid to the infl uence of political

par-ties on public education spending, based on the famous “ partisan esis ” (Hibbs 1977 ; Alt 1985 ; Castles 1982 ; Chappell and Keech 1986 ; Hicks and Swank 1992 ; Schmidt 1996 ; Tufte 1978 ) Yet, the debate on party effects on public education spending is still far from settled (cf Busemeyer et al 2013 ; Garritzmann and Seng 2015 ) Following partisan hypothesis in its simple, original form, early studies used the left–right continuum to deduce hypotheses about party preferences towards public education spending (Verner 1979 ; Castles 1982 , 1989 , 1998 ; Hega and Hokenmaier 2002 ) These studies argued that left parties spend more on

hypoth-education because hypoth-education systems could foster equality of ties, upward mobility, and redistribution Additionally, Boix ( 1997 , 1998 ) claimed that the left’s traditional “demand-side politics” are not feasible in

opportuni-an economically globalized world opportuni-anymore; therefore, he argues, left ties increasingly focus on “supply-side policies,” especially on investment

par-in human capital formation Busemeyer ( 2009a ) offers empirical support for this claim, demonstrating that social democrats in globalizing econo-mies indeed spend more on higher education

Subsequent publications argued that the left–right “super-issue” allows

fi rst hypotheses to be derived, but does not cover party differences on education policies adequately Schmidt ( 1996 ), for example, claims that neither all left nor all right parties hold the same position regarding edu-cation policies Focusing especially on Christian democratic (CD) parties, scholars have demonstrated that the patterns of party preferences over education are more complex than simple left–right dichotomies sug-gest Despite being right-wing in many terms, Christian democrats were

found—just as for social spending (Esping-Andersen 1990 ; Huber et al

1993 ; Kalyvas and van Kersbergen 2010 )—to be important proponents

of education spending (Busemeyer 2007a , 2015 ; Iversen and Stephens

2008 ) Furthermore, liberal parties do not fi t neatly into the clear-cut

left–right picture, as they can be of a left-leaning or a right-leaning type (Smith 1988 ; Franzmann 2012 ) and hold very different education policy positions depending on which branch of liberal parties they belong to (see Busemeyer et al 2013 and Chap 4 ) Moreover, social democrats seem to

be more obliged to spend on education than other left parties ( Busemeyer

2006 , 2007b , 2009a )

Consequently, in order to account for these differences within the

broader left–right camps, these recent contributions proposed using party

families as a more fi ne-grained distinction ( Busemeyer 2006 , 2007b , 2009a ; Potrafke 2011a ; Schmidt 2002 , 2007 ) Empirical studies have underlined these claims, demonstrating that parties within the broader left–right camps indeed behave differently: on the right side of the politi-cal spectrum, liberals and Christian democrats increased public education spending—at least in times of economic prosperity ( Busemeyer 2006 )—while conservative parties are negatively associated with public education spending (Schmidt 2007 : 172; Busemeyer (2006 , 2009a ) On the left side, social democrats were found to support education more than parties of the far left (Ansell 2010 ; Busemeyer 2006 , 2007b , 2009a ; Jakobi 2011 ) Complicating things further, some studies have argued that the effects

of parties are not uniform across all education sectors because different

(re-)distributional patterns are at work In particular, higher education has

been argued to follow a different (re-)distributional logic: as children from higher-income families have greater chances of participating in higher edu-cation (Breen and Jonsson 2005 ; Becker and Hecken 2009 ; Blossfeld and Shavit 1993 ), public expenditure on higher education institutions is fi nan-cially regressive because it redistributes tax money from the general public

to the rich (Fernandez and Rogerson 1995 ), as emphasized by Karl Marx

in 1875: “When [not only lower but also] higher education institutions

in some countries are ‘free-of-charge’, this only means covering the rich offspring’s education costs with public money” (Marx 1973 [1890/91], Author’s translation)

Accordingly, some scholars argue that we should expect right parties

to spend more on education, at least when it comes to higher tion (Fernandez and Rogerson 1995 ) Indeed, Rauh et al ( 2011 ), Wolf ( 2006 ), and Potrafke ( 2011b ) report support for this claim—at least in

educa-comparisons on the German Länder —as Christian Democrats and (to a

lesser degree) liberal government participation is associated with larger higher education spending, while left-wing governments are found to lead

to lower tertiary education spending

A fi nal, more nuanced perspective is offered by Ansell ( 2008 , 2010 ), who took partisan hypothesis one step further and proposed a “condi-tional party hypothesis” (Ansell 2010 : 19), which is essentially a synthesis

of the previous accounts Ansell reasons that parties’ positions and ferred higher education policies are not constant over time, but rather contingent on the status quo of the higher education system, particularly

pre-the enrollment level When higher education enrollment is low , right

par-ties want to increase public spending on higher education because access

to higher education is conditioned by income and mainly to the benefi t of the constituency of right parties For the same reasons, left parties oppose more spending and enrollment expansion when enrollment levels are low

When enrollment levels are high , however, Ansell reasons that after passing

a certain threshold, party positions reverse: now, left parties are theorized

to favor further expansion, because left party voters suddenly see a able chance of receiving higher education In contrast, right parties are argued to oppose further expansion and additional spending when enroll-ment surpasses a certain threshold

To date, Ansell offers the most elaborate political economy analysis

of higher education Nonetheless, a number of crucial questions remain open Firstly, perhaps the most important element of Ansell’s argument, namely the “fl ipping point” leading parties to alter their positions, remains theoretically ambiguous and empirically impossible to determine (as Ansell admits) Moreover, if parties are as strategic as Ansell assumes, then

we should expect that enrollment levels never reach that fl ipping point, because right parties could (if they govern) put a cap on further expansion

before it begins benefi tting the left’s constituency (as indeed happened for example in Japan, see Sect 3.7) Secondly—and more fundamentally as

I show below—Ansell essentially treats enrollment levels as exogenously given I argue in the next section, however, that parties can and empirically have had large effects on enrollment levels, mainly by offering (or abstain-ing from offering) fi nancial student aid In other words, while Ansell treats enrollment as exogenous, it is endogenous in my explanation In fact, I

posit that because subsidies are disregarded in the existing literature, the

essential cornerstone of the story is missing , leaving us with the puzzling

inves-have investigated public spending on higher education Moreover, no study

has investigated determinants of public expenditure on higher education

subsidies , i.e., student fi nancial aid However, as I argue in Sect.  1.3 and demonstrate empirically in the rest of this book, this neglect is the corner-stone of our understanding of the political economy of higher education

fi nance that is lacking I posit that we can only understand the political economy of higher education once we distinguish between public higher

education spending on institutions , on one hand, and spending to students

in the form of subsidies , on the other By doing so, we can also resolve the

unsettled debate of whether left and/or right parties favor higher tion spending

educa-1.3 THE ARGUMENT OF THIS BOOK

The literature review (Sect.  1.2 ) demonstrated that—despite the topic’s high theoretical and societal relevance (cf Sect.  1.4 )—very few studies have aimed to explain tuition fees or subsidies Moreover, although it is often claimed that tuition fees and subsidies are intertwined phenomena,

no analysis to date has attempted to offer a comprehensive model ing sense of both fi nancial dimensions, i.e., the Four Worlds of Student Finance In the following sections, I propose a model that explains both the historical origin and the more recent sustainability of the Four Worlds of Student Finance Subsequently, I derive crucial empirically testable impli-cations, which the remainder of this book then investigates empirically

mak-1.3.1 What Explains the Origin of the Four Worlds of Student

Finance?

1.3.1.1 Partisan Hypothesis as a Starting Point

The starting point of my theory—as in some of the core contributions in the skill formation literature (Ansell 2008 , 2010 ; Busemeyer 2009a , 2015 ; Iversen and Stephens 2008 )—is partisan hypothesis (Hibbs 1977 ; Alt 1985 ; Castles 1982 ; Chappell and Keech 1986 ; Hicks and Swank 1992 ; Schmidt

1996 ; Tufte 1978 ) In its most simple form, partisan hypothesis states that different voter constituencies have different preferences that parties aggregate into party positions and governing parties transform into poli-cies favoring their constituencies Parties are thus considered to be policy

and vote maximizers (Müller and Strøm 1999 : Chap 1) Consequently, expectations can be derived from partisan hypothesis from two perspec-tives: on one hand, one can begin in a “top-down” approach with parties’ ideologies and derive hypotheses about their behavior from their ideologi-cal position; on the other hand, one can begin in a “bottom-up” approach with voters’ preferences to derive hypotheses on party positions In this book, I develop hypotheses on higher education policies from both of these perspectives and demonstrate that they lead to the same conclusions Despite the simplicity of partisan hypothesis, debate in the skill for-mation literature is still ongoing regarding the effects of parties on higher education As summarized earlier (cf also Busemeyer et al 2013 ; Garritzmann and Seng 2015 ), some argue that left-wing parties favor

more spending on higher education and promote further expansion (e.g., Boix 1997 , 1998 ; Busemeyer 2009a ), while others argue that right-wing

parties aim to expand higher education spending and enrollment levels (Fernandez and Rogerson 1995 ; Jensen 2011 ; Potrafke 2011b ; Rauh

et  al 2011 ; Wolf 2006 ) Finally, Ansell ( 2008 , 2010 ) posits that party effects are conditional on the enrollment level Thus, although partisan hypothesis in principle offers a straightforward and parsimonious theoreti-cal framework, the application to (higher) education in the existing litera-ture has run into contradictory expectations and fi ndings

I posit that these contradictions can be resolved once we distinguish between public expenditure on higher education institutions , on one

hand, and public expenditure on fi nancial student aid (i.e., subsidies), on

the other hand Drawing this simple distinction—which has so far been neglected in the literature—allows deduction of much clearer and non-

contradictory expectations: Left-wing parties are commonly assumed to

aim at equality of opportunities and social upward mobility in order to achieve socially, economically, and politically more equal and just soci-eties 8 I argue that (means-tested) fi nancial student aid is one means of achieving equality of opportunities and upward mobility, because they set incentives for students from lower socio-economic or educational back-grounds to begin and continue studying (Sect.  1.4 below provides evi-dence that subsidies indeed have these effects)

Thus, I expect that left parties will seek to establish public subsidies and expand their generosity More specifi cally, left parties will seek to widen the share of subsidy recipients, increase the amounts received, and improve the repayment conditions, e.g., by offering grants/scholarships (i.e., subsidies without repayment necessity) rather than loans (which the students have to pay back) This proposition should hold irrespective

of country and time contexts and also irrespective of the status quo of the education system That is, while Ansell might be right that parties’ preferences fl ip conditionally on the enrollment level when we consider

public expenditure on institutions (because the makeup of the student body changes over time), it seems plausible regarding subsidies that left

parties pursue their introduction and expansion because they are likely to foster equality of opportunities, serve their ideological goals, and benefi t their constituency in any case Thus, from the ideological standpoint of left parties we can deduce that they do engage in establishing generous subsidies

The same conclusion follows when we begin “bottom-up” with left parties’ constituencies and their preferences: left parties are generally assumed to receive votes particularly from lower socio-economic strata (SES) As lower SES voters benefi t disproportionately much more from subsidies but pay disproportionally little when the tax system is progres-sive, they will favor generous subsidies as a means of achieving access

to higher education, in order to gain higher wages, better work tions, and the like (Chap 6 shows that this is indeed the case) Left parties, as agents of these lower socio-economic groups, can conse-quently be anticipated to install and expand far-reaching subsidies when

condi-in government

Right-wing parties, in contrast, can be expected to oppose public

sub-sidies for four reasons Firstly, their constituencies, generally assumed to stem mainly from higher SES, pay disproportionally much more for and benefi t disproportionally little from this progressive (redistributive) spend-

ing Secondly, as access to higher education is de facto stratifi ed by tal income in all countries (Breen and Jonsson 2005 ; Becker and Hecken

paren-2009 ; Blossfeld and Shavit 1993 ), the right’s constituency of income groups might fear that the exclusiveness of “their good” higher education might decrease as the share of students increases Put differently, right parties and their constituencies prefer elite higher education to mass higher education Thirdly, right parties and their constituencies can be

higher-expected to consider the quality of higher education to be more important

than broad access and equality of opportunities Thus, they might fear that the quality of higher education decreases when more students study (due

to subsidies) Finally, right parties, at least of a conservative and Christian Democratic type (though this is less true for liberal parties), ideologically place much emphasis on the family and its responsibility in supporting its children (Esping-Andersen 1990 ; Heywood 2007 ; Iversen and Stephens

2008 ; Kalyvas and Van Kersbergen 2010 ; van Kersbergen 1995 , 1999 ) It follows from their interpretation of the subsidiarity principle that mainly families—and not the state—should be responsible for enabling their chil-dren to attend higher education Moreover, as right parties conceive of society as “naturally hierarchical” (Heywood 2007 ), right parties are ideo-logically not as opposed to unequal access conditions as left parties Thus, for these ideological and constituency-based reasons, an ideal-typical pure right-wing government can be expected to refrain from installing any kind

of public subsidies, preferring to spend tax money directly on higher cation institutions or reducing taxes in the fi rst place

In summary, we can expect that left-wing parties always seek to expand subsidies, while right-wing parties prefer lenient student support systems What expectations can we deduce from partisan hypothesis regarding

parties’ positions on tuition fees ? For very similar reasons as those just

discussed for subsidies, I expect that left parties will oppose tuition fees while right parties will favor them (when they see the quality of higher education threatened due to increasing student numbers) For left parties and their constituencies, tuition fees are another threshold to entering higher education A by now broad sociological and psychological litera-ture has demonstrated repeatedly that children from lower SES tend to

be more debt- and risk-averse than children from high SES backgrounds, i.e., they over-estimate the costs and under-estimate the benefi ts of higher education, consequently enrolling at lower rates, in cheaper (worse) insti-tutions, or in shorter study programs (see Archer et al 2003 ; Becker and

Hecken 2009 ; Callender and Jackson 2005 , 2008 ; Conner and Dewson

2001 ; Forsyth and Furlong 2000 ; Hesketh 1999 ) So, even if broad sidy systems exist, (non-trivial) tuition fees will have a negative effect

sub-on the enrollment decisisub-ons of potential students from lower SES grounds (Sect.  1.4 summarizes empirical studies confi rming this relation-

back-ship) Consequently, in order to generate equality of opportunities, left

parties will oppose tuition fees so that (fi nancial) deterrence effects are as

small as possible

Children from higher-income groups, in contrast, are less responsive

to tuition fees because the fi nancial contribution that their families can provide is larger; consequently, the higher the students’ fi nancial back-grounds, the lower the discouraging effect of tuition fees Accordingly, the constituency of right parties can be expected to be less concerned with the effects of tuition on participation rates, but more concerned with

the quality of higher education: right parties prioritize high-quality (elite)

higher education over (potentially lower-quality) mass enrollment and equality of opportunities Thus, when fearing that the quality of higher education might decrease (due to increasing enrollment levels), right par-ties will look for additional revenue sources and use tuition fees as one means to generate additional funds Taking this information together,

right governments can be expected to install and raise tuition fees — at least

when concerned that the quality of higher education might decrease — while left governments can be expected to keep tuition at a minimum

To summarize, we can deduce from partisan hypothesis from both a

“top-down” ideological perspective and a “bottom-up” constituency spective that left parties will favor generous subsidies and oppose tuition fees while right parties conversely oppose subsidies and favor tuition fees,

per-at least when fearing thper-at the quality of higher educper-ation might decrease (due to rising student numbers)

How do these party positions transfer into policies? We should expect that countries that have constantly been governed by left-wing govern-ments should have departed from the historical low-tuition–low-subsidy starting point in the immediate post-war years by keeping tuition low and extending the generosity of the subsidies over time Thus, this fi rst path of constantly dominant left governments can explain the low-tuition–high- subsidy regime Vice versa, countries with constantly strong right-wing parties can be expected to increase tuition fees but leave their subsidy system meager, resulting in considerable private contributions and little state support This second path of right-wing dominance thus explains

the high-tuition–low-subsidy regime Chapter 3 illustrates these two paths empirically by showing how and why predominant left governments in Finland and predominant right governments in Japan led to the respective regime paths, while Chap 5 provides quantitative support for this rea-soning in a large-n setting What remains unexplained thus far, however, are cases of “off-diagonal” low-tuition–low-subsidy (e.g., Germany) and high-tuition–high-subsidy (e.g., the USA) regimes In order to under-stand these regimes, we need to go beyond the simple form of partisan hypothesis

1.3.1.2 From “Simple” Partisan Hypothesis Towards

a “Time-Sensitive” Partisan Theory

Partisan hypothesis applies well to countries that are governed purely (or at least predominantly) by either left-wing or right-wing majorities Yet it is much more diffi cult to derive expectations for cases where either cross- ideological grand coalitions governed or where left and right par-ties altered whilst in offi ce What, for example, should our expectation be regarding the makeup of the tuition-subsidy system of a country that was governed half of the time by left-wing parties and half of the time by right-wing parties? Applying partisan theory in its simple form, we should see that each party establishes “half of its preferred policies”; thus, we should

fi nd an “average system” of medium-high tuition fees and medium-high subsidies Yet, this neither seems theoretically convincing nor does it fi t empirically, as no OECD country shows such a regime How, then, can

we deal with these “mixed” cases, which actually form the vast majority of cases across the OECD (cf Chap 2 )? Put differently, how can we explain the two remaining regimes, namely the high-tuition–high- subsidy system and the low-tuition–low-subsidy cluster?

I argue that in order to offer a convincing explanation of the Four Worlds of Student Finance we need to extend the simple partisan hypoth-

esis by taking time and timing seriously, moving towards a Time-Sensitive

Partisan Theory To illustrate, consider an analogy proposed by Pierson

( 2004 ):

Imagine that your friend invites you to the trendiest new restaurant in town, charmingly named ‘The Modern Social Scientist’ … When you arrive, the chef … proceeds to elaborate her culinary approach: good cooking, she says, amounts to having the perfect ingredients, perfectly measured Traditional cooks have stressed how important the cooking process itself is, including