bs_bs_banner

The Policy Studies Journal, Vol. 44, No. 1, 2016

Policies that Succeed and Programs that Fail:

Ambiguity, Conflict, and Crisis in Greek Higher
Education†
Nikolaos Zahariadis and Theofanis Exadaktylos

Why do some policies adopted by a wide margin fail to be implemented? Highlighting the role of
policy entrepreneurial strategies within the Multiple Streams Approach (MSA), we examine the
implementation of Greek higher education reform in 2011 to argue that when policies adversely affect
the status quo, successful entrepreneurial strategies of issue-linkage and framing, side payments, and
institutional rule manipulation are more likely to lead to implementation failure under conditions of
crisis, centralized monopoly, and inconsistent political communication. The findings clarify MSA by
specifying the conditions that increase the coupling strategies’ chances of success or failure and
illuminate the role ambiguity and conflict play in policy reform and implementation.
KEY WORDS: multiple streams, coupling strategies, policy entrepreneurs, Greece, higher education

Introduction

Why do some policies adopted by wide margins fail to be implemented? High-

lighting the role of policy entrepreneurial strategies within the Multiple Streams
Approach (MSA) (Kingdon, 1995; Zahariadis, 2014), we explore the implementation
of Greek higher education reform to probe the following hypothesis: when policies
adversely affect the status quo, successful strategies of policy formation (agenda set-
ting and policy adoption) are more likely to lead to implementation failure under
conditions of crisis, centralized monopoly, and inconsistent political communication.
In MSA terms, the mechanisms of failure include decoupling problems from solu-
tions, undermining support in the political stream, and altering estimates of equity
and efficiency in the policy stream. We define failure as inability or unwillingness to
execute the law in letter or spirit within the legally prescribed time frame. Within
our empirical case, the policy has not completely failed, but major aspects of it have
been dramatically modified, delayed, or abandoned.

†
An earlier version of this article was presented at the workshop “Ambiguity and Public Poli-
cy,” University of Alabama at Birmingham, Birmingham, AL, 14–15 November, 2014.

59
0190-292X V
C 2015 Policy Studies Organization

Published by Wiley Periodicals, Inc., 350 Main Street, Malden, MA 02148, USA, and 9600 Garsington Road, Oxford, OX4 2DQ.
60 Policy Studies Journal, 44:1

It is expected policies attracting extensive public and political support should be

implemented with relatively few obstacles. Wide appeal is likely to overcome
impediments to execution because both the public and politicians have come to con-
sensus. We examine one exception: the law reforming Greek higher education in
2011 passed by more than two-thirds majority of votes in Parliament (260 out of 300,
or 87 percent). This feat is remarkable given that it was passed by a greater majority
of votes than any other bill since the restoration of Greek democracy in 1974. It is
also the only time when the two largest parliamentary groups during this period, the
socialist Panhellenic Socialist Movement (PASOK) and the conservative New Democ-
racy, in their entirety voted for a law of any kind. Yet, implementation proved signif-
icantly more difficult.
The importance of our case rests on two reasons. First, higher education reform
was not included in the reforms agreed to in Greece’s bailout packages; yet, it has
received significant public attention and has raised considerable political conflict.
Governing parties and the opposition have used education reform as rallying cry and
“battlefield” to support or oppose the bailout packages and fight broader struggles
regarding the role of the public sector in Greece (Diamantopoulou, 2011; Tsiligiris,
2012). Second, the case enables our research design because it is precisely in circum-
stances of wide political agreement that we expect unproblematic implementation. In
George and Bennett’s (2005) terms, an extreme “most likely” case such as ours is best
positioned to highlight problems because they are least expected. Being exemplar but
temporally limited, the case is not representative of general policy change but of
redistributive reforms under specific conditions. We unpack the paradox of failure by
stressing the contingent nature of implementation strategies (Matland, 1995).
We chose MSA as our theoretical framework for two reasons. First, the case of
reform is precisely the issue that MSA claims to be most applicable. Because reforms
involve collections of problems, solutions, and dramatic shifts in values and institu-
tional structures, there are likely to be high levels of ambiguity, political contestation,
and downright ignorance about cause and effect (March & Olsen, 1983; Rommetveit,
1976). Aberbach and Christensen (2014) use MSA to add that adopted reforms often
disappoint because it is difficult to sustain them, but they do not specify means or
ways to avoid failure. Second, despite its applicability, MSA does not provide
explicit mechanisms linking implementation and policy formation apart from gen-
eral strategic effects of ambiguity (e.g., Baier, March, & Sætren, 1986; Zahariadis,
2008; but see Ridde, 2009). Although the literature (including MSA) typically
explores the role of entrepreneurs in agenda setting or policy formation, we show
coalitions pursue entrepreneurial strategies during implementation that have conse-
quential effects (see also Arnold, 2015).
Situating our analysis within MSA, we first discuss the role of policy entrepre-
neurs in bringing about policy change. Theoretical linkages between entrepreneurial
strategies and implementation are then used to specify conditions of success (or fail-
ure). We do not argue strategies are the only factors affecting reform; we just explore
their effects. Using Pollitt and Bouckaert’s (2011, p. 77) reform typology, we empiri-
cally probe the hypothesis in four areas: financial, personnel, organizational, and per-
formance measurement. The findings clarify MSA by specifying the conditions
Zahariadis/Exadaktylos: Policies that Succeed and Programs that Fail 61

affecting the likelihood of success or failure and illustrate the role ambiguity and
conflict play in linking policy formation to implementation.

Multiple Streams, Implementation, and Policy Entrepreneurship

MSA places policy entrepreneurs and their strategies at the heart of policy
change. It argues policies are made when entrepreneurs couple or join together three
streams—problems, policies, and politics (Zahariadis, 2014). The chances a particular
policy will be adopted increase significantly when all three streams are coupled dur-
ing open policy windows. Windows are opportunities that open in the problem or
politics stream; they define and limit the context within which policy is made. A key
MSA element is ambiguity, which involves contestations over issues, meaning,
causes, and consequences. By encouraging rival interpretations, ambiguity affords
opportunities for policy entrepreneurs to build and sustain coalitions that advocate
or oppose policy change. We assume coalitions are built in the policy formation
phase and explore coupling and decoupling processes during implementation.
We begin where this argument leaves off, namely by linking policy formation
to implementation and by identifying conditions for entrepreneurial coupling
success or failure. We conceptualize a nested policy system (Howlett, McConnell,
& Perl, 2015). Whereas most MSA theorizing assumes a single window and ends
analysis with a policy’s adoption (or not), we view policy outputs as constituting
implementation windows (Ridde, 2009). We estimate two phases (formation and
implementation) with multiple rounds of deliberation, each phase marked by
continuities with previous actions but also adding new actors and/or potentially
new resources (Teisman, 2000). We are not interested in how successful entrepre-
neurial strategies shape policy outputs but how they are used to implement them,
connecting context and coalitions around linked but ambiguous frames of prob-
lems and solutions.

Entrepreneurial Strategies

Studies of the policy process often highlight the ability of policy entrepreneurs to
build coalitions that enact and implement policy changes (e.g., Mintrom, 2000;
Zahariadis, 2003). Entrepreneurs are individual or corporate actors who operate in or
out of government and who are willing to invest resources—time, energy, expertise,
or money—to advocate for major policy change (or prevent change from occurring)
(Kingdon, 1995, p. 122; Mintrom & Norman, 2009, p. 650). Entrepreneurs continu-
ously advocate and broker, display social acuity, define problems, build teams, and
lead by example in support or opposition to policy alternatives at any stage of the
policy process (Mintrom & Norman, 2009, p. 651).
Analysis of entrepreneurial activity is normally divided in two parts: attributes
and strategies. Most of the MSA literature focuses on major attributes—access and
persistence—that increase the chances of entrepreneurial success (Kingdon, 1995).
However, success cannot be attributed solely to individual attributes; it must also be
62 Policy Studies Journal, 44:1

considered as action in context (Schneider, Teske, & Mintrom, 1995). This is not to
downgrade agency but to complement it. While entrepreneurs have the capacity to
choose behavior, explanations of policy outcomes cannot be devoid of the context
(institutions, roles, and resources) that regulates social interaction. Agency and con-
text, therefore, should not be viewed in isolation but as linked through strategy.
Strategies are scripted and socially constrained or enabled patterns of action which
are chosen to go in particular directions as opposed to others. The implication is
success is not simply a matter of choice but also of context. As Ackrill, Kay, and
Zahariadis (2013) contend, employing the same strategy in a different context alters
the outcome. Research has been conducted on entrepreneurial strategies (e.g.,
Mintrom, 2000), but relatively little scholarly attention has been paid to systematically
explaining the conditions under which they may be successful.
Applying Meijerink and Huitema’s (2010) insight, we explore the effects of entre-
preneurial strategies—issue linkages and framing, institutional rule manipulation,
and side payments—on implementation. We chose to apply their argument on entre-
preneurship instead of Mintrom’s (2000) because of its comprehensive and compara-
tive application across government levels. They succinctly categorize strategies but
neither do they identify conditions of success nor do they link strategy to implemen-
tation. Roberts and King (1991) do, but their study is limited not only to a specific
kind of entrepreneur, one who operates outside government, but also to a specific
kind of policy, innovation.
Drawing conclusions from a comparative study of 15 cases of water manage-
ment across different countries, Meijerink and Huitema (2010) differentiate between
five types of entrepreneurial strategies: coalition building, idea development and dis-
semination, venue-shopping, network management, and strategic framing of ideas
and opportunities. Brouwer and Biermann (2011) add that policy entrepreneurs may
pursue more than one strategy at the same time. We adapt Meijerink and Huitema’s
(2010) argument to gain analytical traction in implementation. We simplify by posit-
ing a nested repertoire of strategies where the ultimate objective is to sustain coali-
tions already built during policy formation akin to Levin and Ferman’s (1986)
“fixers.” That is done through the remaining four strategies. Because we deal only
with implementation, the strategies of framing and idea development and dissemi-
nation may be collapsed into a single strategy of issue linkages and framing. We
assume ideas have already been adopted so there is no need to trace their develop-
ment. Furthermore, we collapse the strategies of network management and venue-
shopping into institutional rule manipulation. It encompasses shopping among dif-
ferent venues that contain various networks. Finally, we add side-payments as an
important resource to create and maintain minimum winning coalitions (Riker,
1962).
Issue linkages and framing are discursive strategies used to attract supporters to
new policy proposals, mobilize opponents, and justify policy interventions. Link-
ages between cooperation on one issue and cooperation on another can ensure
that all parties gain by participating. Or they can play a strategic role by expand-
ing the agenda to mobilize opponents and increase policy conflict (Schattschneider,
1960). Problem definition and framing play a critical role in focusing and
Zahariadis/Exadaktylos: Policies that Succeed and Programs that Fail 63

sustaining the attention of coalitions by altering perceived consequences and policy

images (e.g., Rochefort & Cobb, 1994). By framing consequences, policy entrepre-
neurs establish causal links between problems and desirable policy options. Suc-
cessful frames also enable entrepreneurs to supersede institutional constraints and
move across streams with relative ease (Knaggård, 2015). Framing is thus a cou-
pling strategy that joins together problems, solutions, and politics to build narra-
tives among coalition members. Such narratives in turn enable entrepreneurs to
affect coalitional composition, cohesion, size, and sustainability (McBeth, Jones, &
Shanahan, 2014).
Entrepreneurial strategies are bounded by institutional context. As such, entrepre-
neurs routinely manipulate rules to increase the chances of success or failure. While
network management and venue-shopping have been mostly explored in terms of
agenda-setting or decision-making (e.g., Baumgartner & Jones, 1993; Zahariadis,
2003), they have implications for implementation as well. Managing a centralized
network, for example, confers decision power to governing officials, but hostile local
bureaucrats may successfully derail enforcement. Coalitions, therefore, must be not
only horizontally large enough but also vertically integrated across different venues.
Venues are arenas where entrepreneurs try to win the policy game by building stra-
tegic alliances to pool resources and support. The greater the distance between
intended changes and actual implementation, the more institutional ambiguity is cre-
ated and the more intense the conflict will be over rival interpretations in different
venues. Coalitions that are not vertically integrated across different venues, espe-
cially those that include vital stakeholders, are less sustainable over time giving rise
to bureaucratic delay and obstruction. In fact, although strategies may be pursued to
support or block policy, policy entrepreneurs, Meijerink and Huitema (2010) find,
more often than not “create barriers to future change” by institutionalizing their
ideas in more favorable venues.
Side payments are used to create and sustain winning coalitions. They involve
current and future promises to pay that increase value for coalition members. Poli-
cies typically include such provisions to strengthen support and minimize opposi-
tion. Riker (1962, pp. 108–14) lists several types of relevance here. Payments in the
form of monetary or other value often constitute the main incentives to join and sus-
tain winning coalitions. For example, implementing parts of the law might result in
additional funding or open up possibilities to improve the quality of service. Nega-
tive incentives in the form of sanctions sustain coalitions by specifying the conse-
quences of nonparticipation. For example, not enforcing provisions of the law might
result in funding cuts. If governing members of the coalition can be reasonably
expected to play the same role in a number of future policies, they can then credibly
offer promises about subsequent decisions. Members will therefore sustain the coali-
tion in the expectation of future benefits. Finally, payments that cater to the ideology
of coalition members are highly valued for the moral satisfaction of serving the cause
or the political benefit of defeating the “enemy.” The more ambitious the changes
and the greater the ambiguity, the more resources are needed for implementation
and the greater the use of selective side payments to sustain the minimum winning
coalition.
64 Policy Studies Journal, 44:1

Strategy and Implementation Failure

The very fact we conceptualize two phases implies coalitions that support a pol-
icy during the policy formation phase may be different from the ones that implement
it (Aberbach & Christensen, 2014, p. 8). Formation and implementation are linked
through an implementation window, which provides opportunities for renewed con-
testation over policy objectives and outcomes. Ideas and strategies may still be the
same but the activation of new actors in a different context may alter the outcome.
The key difference between the two phases is that agencies and bureaucrats are inti-
mately involved in implementation. In contrast to policy formation where politicians
are the crucial decision makers, in implementation they often have a monitoring
role. Fimister and Hill (1993) claim that leaders often prefer the legitimacy of initiat-
ing reforms while allowing agencies to edit or translate them to deflect blame from
failure. Therefore, the composition of coalitions leads to different objectives and natu-
rally different incentives to join and/or deliver public services. For example, whereas
funding may (or not) be an issue in policy formation, adequate resources to imple-
ment a policy are one of the key ingredients in successfully translating talk into
action (Zahariadis, 2008). When policy is heavily contested, politicians secure success
by structuring access and providing resources (Matland, 1995). In the presence of
ambiguous goals, street-level bureaucrats and more broadly local coalitions acquire
significant discretion in shaping the exact nature, quantity, and quality of public
services (Arnold, 2015; Lipsky, 2010; Matland, 1995, p. 168).
Ambiguity is often essential in public policy because disparate coalitions need to
be built and supporters must declare victory, each perhaps for his/her own reasons.
This ambiguity provides room for interpretation to those who must put laws into prac-
tice, leading to contingent strategies of implementation. When ambiguity is low with
bitter conflict over goals, compliance is contested and outcomes determined by politi-
cal power. When ambiguity is high and conflict equally high, the strength of local coa-
litions shapes the outcome. Matland (1995) labels the former political implementation
and the latter symbolic implementation. Strategy success is therefore likely to vary
because reforms, being redistributive policies, generate conflict and often ambiguity.
Under what conditions will resistance succeed in undermining the implementa-
tion success of coupling strategies? We hypothesize:

When policies adversely affect the status quo, successful entrepreneurial strategies
of issue-linkage and framing, side payments, and institutional rule manipulation
are more likely to lead to implementation failure under conditions of crisis, central-
ized monopoly, and inconsistent political communication.

In MSA terms, the mechanisms linking strategy to failure involve decoupling

problems from solutions, undermining support in the politics stream, and altering
estimates of equity and efficiency in the policy stream.
Government bureaucracies are by nature change-avoiding organizations
(Wilson, 1989). Change involves redistribution of resources, subtraction, or addition
Zahariadis/Exadaktylos: Policies that Succeed and Programs that Fail 65

of responsibilities, creating uncertainty over future outcomes. For example, creating

new universities may generate animosity by existing institutions if new universities
are viewed as taking resources away from established ones. The same is true when
new policies create jurisdictional ambiguities or overlap which in turn lead to the
development of new rules, struggles over interpretation, unstable practices, turf bat-
tles, and potentially new conceptions of political legitimacy and consequences
(Hajer, 2003; Olsen, 2006). Ambiguity is amplified and conflict endogenized espe-
cially when reforms involve collision between internally generated processes of
change and externally imposed ones (Olsen, 2007). Because government bureauc-
racies, especially universities, are complex organizations with numerous and often
ill-defined goals, they are likely to resist implementing new policies that adversely
affect their budgets or structure. Resistance to change needs to be overcome through
successful entrepreneurial strategies that sustain winning coalitions throughout the
policy process (Roberts & King, 1991). The greater the ambiguity of the law, the
greater the discretion over rival interpretations or enforcement (Mahoney & Thelen,
2010) and the stiffer the resistance are likely to be because of potential consequences
(Huber & Shipan, 2002).
Reforms frequently involve framing contests (Boin, ‘t Hart, & McConnell, 2009)
and contradictory demands. The last factor is especially troubling in periods of cri-
sis because policymakers are pushed by internal and external forces to adopt poli-
cies that return some semblance of normalcy and pulled by others to reform in
order to address the conditions that created the crisis in the first place (Zahariadis,
2013). Because crises involve significant turbulence and uncertainty, they create
demands for fundamental changes. When these demands become ambiguous laws,
coalitions (re)form to seek particular interpretations of (including opposition to) the
new rules (Mahoney & Thelen, 2010, p. 11). Conflict increases and resistance to pro-
grammatic change stiffens because autonomous agents and/or clients usually do
not participate in the decision-making process. Therefore, issue linkages that are
successful in creating minimum winning coalitions are likely to inhibit implemen-
tation because they do not help sustain those coalitions. For example, making the
argument that a severe fiscal crisis necessitates dramatic funding cuts stiffens
resistance because such a linkage helps opponents to also put forth broader griev-
ances about the crisis to undermine reforms. Problems and solutions are decoupled
and likely adversely reframed to protract conflict. The end result is spillover across
issues and policy sectors, resulting in coalition breakdown and symbolic efforts to
legitimize responses that are likely to falter under the weight of increased political
conflict. In the presence of ambiguity and conflict, the larger the coalition during
policy formation under crisis conditions, the less sustainable the coalition is likely
to be during implementation because the higher number of affected actors will
increase the number of associated grievances. In implementation, the policy stays
the same but problems are reframed because “actors see their interests tied to a
specific policy definition” (Matland, 1995, p. 168).
Crises prune side payments because of budgetary constraints. While side pay-
ments, such as extra funding or more rights, are needed to maintain the stability of
minimum winning coalitions (Riker, 1962), the activation of new actors during
66 Policy Studies Journal, 44:1

implementation might prove prohibitive during fiscal crises. Reforms may still be
passed, but their likelihood of implementation decreases significantly.
Centralization and enforcement sanctions in case of noncompliance greatly affect
civil servant resistance (Saurugger, 2012). Using MSA to explore implementation,
Ridde (2009, p. 948) finds that in a centralized system “peripheral stakeholders [i.e.,
local bureaucrats], who usually follow central instructions, will have a greater tend-
ency to implement decisions.” But we argue this is not necessarily true. While cen-
tralized monopolies at first glance confer enforcement benefits to elected
representatives because of top-down hierarchy of funding and communication, they
work the other way, too. Centralized monopolies also confer significant benefits to
bureaucrats because of the lack of competition in delivering services and an appro-
priate regulatory framework. If the particular public service is considered to be
socially and economically vital, such as higher education, monopolies operating as
public goods empower administrators and educators to accrue benefits in spite of
small budgets or adverse directives from above by way of trade-offs among conflict-
ing goals (Schoenenberger, 2005). Power differentials privilege policymakers because
they can issue directives and withhold funding if not implemented. However, the
monopolistic nature of delivery implies that significant losses will accrue to both
bureaucrats and politicians if resistance stiffens, i.e., it leads to strikes, legal chal-
lenges, and so on. Assuming no intervention by the courts, the institutional arena
changes from a governmental hierarchy to a more fluid public arena where more
and perhaps irrelevant actors, e.g., the media or unions, might tip the balance against
the government, undermining support within the politics stream. Such nonlegislative
factors shape the probability of expected sanctions (Huber & Shipan, 2002). Central-
ization makes it easier to identify culprits and creates bottlenecks that may stifle
implementation. The more centralized the structure of delivery, the easier it is to
launch legal and other challenges to bring the whole policy to a halt.
Finally, inconsistent political communication affects the intensity and outcome of
reforms (Pollitt & Bouckaert, 2011) because legal challenges and other delay tactics
alter the sustainability and viability of coalitions. Indeed, Sætren (2015) finds leader-
ship commitment and will to be key ingredients of entrepreneurial success in sus-
taining coalitions that promote radical reforms. Inconsistency prompts reframing,
ambiguity, and conflicts because it signals lack of political will (Brinkerhoff, 2000).
As actors become unable to predict outcomes, they are more likely to misrepresent
information in pursuit of non-cooperative behavior (Robbins, 2010, p. 526). Inconsis-
tency also affects perceived payoff. The payoff for joining a coalition is the expected
value of future decisions multiplied by the probability of the coalition functioning
(Gamson, 1961). If a steeper slope discounts the value of decisions because outcomes
are pushed further into the future, coalition support may fall apart. Resources are
redistributed but not in a predictably stable fashion, leading to even deeper utility
discounts for supporters and potential benefits for opponents. In MSA terms, incon-
sistency fosters ambiguity and redistribution, which alter the estimates of equity and
efficiency that sustained the policy’s appeal in the first place (Zahariadis, 2014).
The fact that reforms take a long time to bear fruit saps political motivation and
willingness to build long-term sustainable coalitions (Pollitt, 2008, pp. 16–20, 171).
Zahariadis/Exadaktylos: Policies that Succeed and Programs that Fail 67

The idea is that civil servants can wait out their elected or appointed supervisors
because they have tenure. Because of tenure, they are more likely to face long-term
consequences and are therefore also more likely to resist implementing programs
they perceive might adversely affect them. Besides, delays may also bring politicians
to power who are more sympathetic to bureaucratic grievances. When policies are
decoupled from their supporters in the politics stream, MSA contends, the likelihood
of implementation failure increases. Longer delays lessen the ability to sustain coali-
tions and increase the likelihood that sympathetic policymakers might come to
power and derail the “adverse” reform.

Data and Methods

Implementation success or failure (our dependent variable) is difficult to define

and measure. Analysts use several definitions—efficiency, effectiveness, compli-
ance, or accountability (e.g., Ingram & Schneider, 1990; Matland, 1995; Zahariadis,
2008)—but the short time horizon of our case necessitates an output approach. We
view it dichotomously: either the proposed changes have been implemented within
the specified time period or not. This gives us two values, which we further break
down by including the reasonable outcome of partial change. Partial change refers
to variance either across units (some universities, e.g., implement changes while
others do not) or within unit (a university creates some novel institutional struc-
tures but not others) within the time frame of the law. Failure (no change) exists
when changes are either not or partially implemented beyond the specified time
frame. Success (change) exists when changes are either faithfully executed or when
the majority of units or the majority of proposed changes in an area is implemented
within the time frame of the law.
To gain analytical traction but still retain breadth and variance, we use Pollitt
and Bouckaert’s (2011) typology of reform content—finance, personnel, organiza-
tion, and performance measurement. Our selection has the advantage of different
outcomes, permitting in-case variance and increasing degrees of freedom. We
observe policies enacted by very high margins contain programs that may fail to be
faithfully executed. We examine why by assuming the three entrepreneurial strat-
egies (our independent variables)—issue-linkage and framing, side payments, and
institutional rule manipulation—were used to enact Law 4009/2011 and assess how
their deployment under contextual conditions affects implementation outcomes.
Three conditions are specified. Crisis is defined as sharp fluctuation in national out-
put or employment. Greek economic output shrank by 25 percent in real terms
since 2008 while unemployment peaked at 27 percent in 2013 (Zahariadis, 2013).
Higher education funding decreased by 22 percent in constant terms while faculty
size shrank by 24.6 percent since 2007–08 (Elstat, 2014). Centralization is defined as
monopoly in funding or delivery. Article 16 of the Greek constitution proclaims
higher education to be a state, tuition-free monopoly. Inconsistent political communi-
cation refers to laws, executive decisions, and ministerial proclamations that contra-
dict the letter of the law after its adoption.
68 Policy Studies Journal, 44:1

Table 1. Areas of Reform and Proposals

Area of Reform Proposals On

Organization Establishing university councils

Creating organizational framework of autonomy for universities
Programs of study
Student affairs
National strategy for higher education—Program design agreements
Personnel/Finance Academic staff
Administrative staff
Financial management and access to public/private funding
Performance Assessment and transparency
Evaluation of reforms proposed by the law

Our time period begins in August 2011 when the law was passed and ends at
the time of writing (January 2015). Evidence was gathered from 14 semistructured
interviews and rigorous analysis of legal, university, ministry, newspaper, and other
public documents. Interviewees included members of academic staff, former and
current ministry officials, civil servants, and ordinary voters. Information was dis-
tilled from those sources and tabulated to multiply data sources and identify discrep-
ancies (Yin, 2014). Evidence was then clustered into variables using frequency
distributions and graphs to aid interpretation of results.

Implementing Policy Reform in Greek Higher Education (Law 4009/2011)

We focus attention on Greek Law 4009, which was passed in 2011. Inspired by
Organization for Economic Cooperation and Development (OECD, 2011) advice and
being remarkable in scope and public and political support, it created high expecta-
tions and optimism despite legislative flaws and ambiguities. By decoupling univer-
sity administration from direct political influence; streamlining financial processes to
increase transparency and accountability; extending decision-making autonomy
based on individual institutional needs; and creating a system of evaluation and
accreditation, the legislative framework set up by Law 4009/2011 intended to bring
the quality of Greek universities closer to European and international standards (see
Table 1). Clearly, the sheer range of enacted changes posed serious challenges. By
examining implementation over time across four areas, we can trace the process and
identify sticky areas that have not progressed and areas that were reformed
successfully.
Although passed by an unprecedented majority in Greek Parliament, the Law
immediately faced resistance by coalitions of university unions, some rectors and
professors, and the youth organizations of political parties operating in universities
because it shook up the status quo. Although the composition and intensity of coali-
tions differed across issues, universities, and departments even within the same uni-
versity, opposition grew over time amidst a climate of fear of change, declining
resources, lack of administrative transparency, and perverse tolerance for violence
under the guise of freedom of expression (interviews with four Greek academics
printed in Lakasas, 2015). Some of the law’s clauses were even contested as unconsti-
tutional before the Council of State. The government’s response to implementation
Zahariadis/Exadaktylos: Policies that Succeed and Programs that Fail 69

resistance has been to amend it twice since 2011 to accommodate some of the con-
cerns raised by the stakeholders. Law 4076/2012 (Government Gazette, 159/A/
2012), also known as Arvanitopoulos Law (after the name of the Education Minister
at the time), amended articles dealing with the organizational structure of univer-
sities; the merging, closure, and relocation of schools and departments based on
demand; and election processes of University Councils members. Subsequently, the
ministerial decision 140277/E5 published on 12 September 2014 (Government
Gazette 2435/B/2014)—also known as Loverdos Law (after the name of then educa-
tion minister)—amended the clauses on student transfers between departments at
different universities and on issues relating to student enrollment. Eight additional
laws and one ministerial decision clarified, supplemented, and set up
implementation.
Table 2 shows the planned action by year, size (distance from status quo), impact
on end-receivers, and in bold italics highlights the outcomes. The timeframe for
implementation was mostly not followed. Whereas change took place in several
areas, we observe that many programs were systematically delayed or abandoned,
particularly those calling for structural changes in the operation of Greek univer-
sities. Amendments to the law within the first six months from ratification and the
change of government in June 2012 provided for additional space for resistance
mechanisms and coalitions to develop in spillover fashion, illustrating elements of
symbolic implementation. For example, one change that was later reversed centered
on reconstituting departments as the basic academic units (the law abolished them
and created schools instead). The intent was to rationalize resources and avoid dupli-
cation, but under crisis conditions it was reframed by academics as prelude to layoffs
(interview with faculty from a non-Greek university). Indeed, the quick change dem-
onstrates in line with our expectations that crisis facilitates issue linkages and frames
that undermine implementation. Overall, we find there is no relationship in terms of
distance from the status quo, further reinforcing our argument that variation is an
outcome of entrepreneurial strategies of different coalitions.

Reframing Issues and Manipulating Rules to Affect Organizational Changes

The original law (4009/2011) had a clear framework of implementation which

was extended due to resistance by the end-receivers of reforms. Amendments raised
fresh issues and legal challenges undermining implementation. According to a mem-
ber of the quality assessment team at Aristotle University of Thessaloniki, partial
change is due to new conditions raised by the two amendments of the law. The Hel-
lenic Assurance and Accreditation Agency (HQAA) had been set up and running
since 2008 based on a previous law from 2005 establishing the agency, but it is not at
the heart of making education policy as the law intended (interview with Greek uni-
versity faculty). Interviewed staff from Democritus University of Thrace, whose
department was to incorporate three more departments from other campuses stated
the closure, merging, and relocation of departments has been mostly canceled. Sev-
eral interviewees confirmed that pending court cases regarding student matters
 Table 2. Implementation of Reforms by Year
70

Content/Size/Impact 2011 2012 2013 2014–16

ORGANIZATION  Election procedures  Elect President and Mem-  Merger and closure of PARTIAL CHANGE
Establish University NO CHANGE bers of Council departments  New Councils in place
Councils  Process stalled due to  Elect Chancellors NO CHANGE in many institutions
 Major reform reactions from univer-  Organize Faculties  Scrapped (announced in Septem-
 High impact sity administrations  Elect Deans ber 2014)
 Constitution into a body
PARTIAL CHANGE
 Changes stopped due to
amendments in law
ORGANIZATION  Set up organizational  Set up internal regulations  Publish frameworks in CHANGE
Create Organizational framework  Submit internal regulations Government Gazette  Published in July 2014
Framework of Autonomy CHANGE to the Ministry (enactment)
for Universities  Organizational frame- NO CHANGE  Publish internal regula-
 Major reform work put in place but  Stalled due to the amend- tions in Government
 Medium-high impact not enacted until 2013 ments to the law Gazette (enactment)
PARTIAL CHANGE
 Delayed
ORGANIZATION  Issue vacation time for  Definition of credit sys-
Programs of Study working students tem for similar units of
 Minor reform CHANGE  Procurement for electronic study
 Medium impact  Completed books and virtual learning CHANGE
environments  Completed
CHANGE
 Completed and scheduled
for implementation in
2015–16
ORGANIZATION  Set up Councils for Stu-  Establish upper regis- CHANGE?
Student Matters dent Welfare tration limits and  Issue of eternal students
 Major reform  Set Student Welfare Offi- deregister “eternal” resolved; implementa-
 High impact cers (Ombudsman Offices) students tion will commence in
 Establish Teaching Support NO CHANGE academic year 2015–16
Offices  Frozen for this academic
 Establish Student Support year
Services
 Draft clauses for student
loans
Policy Studies Journal, 44:1
Table 2. cont.

Content/Size/Impact 2011 2012 2013 2014–16

NO CHANGE?
 State of play unknown
ORGANIZATION  Public consultation of Stra-
National Strategy for tegic Plan for Higher
Higher Education—Pro- Education  Ratify Program Design  Evaluate Program
gram Design Agreements  Ratification of Strategic Agreements 2013–14 Design Agreements
 Medium reform Plan NO CHANGE 2013–14 (2015)
 High impact PARTIAL CHANGE  Postponed to 2014 NO CHANGE
 Public consultation took  Ratification of pro-
place but strong reactions grams took place in
postponed Plan September 2014; evalua-
tion delayed
ORGANIZATION  Set up criteria for constitut-
Research Centers of ing Research Centers of
Excellence Excellence
 Major reform  Receive applications to set
 Medium impact up a Research Center of
Excellence
NO CHANGE?
 State of play unknown—
little progress
PERSONNEL/  Procurement to develop  Rationalize promotion
FINANCE and operate electronic sys- scales
Academic Staff tems for managing  Return of Full Profes-
Zahariadis/Exadaktylos: Policies that Succeed and Programs that Fail

 Major reform appointment and promo- sors to teaching

 High impact tion procedures of aca- NO CHANGE
demic staff  Scrapped
 Registry of Assessors
 Procedures to convene
panels for academic staff
assessments
PARTIAL CHANGE
 Partly completed with
some of the regulations
scrapped
71
 72

Table 2. cont.

Content/Size/Impact 2011 2012 2013 2014–16

PERSONNEL/  Salary and rank scales for  Select new University
FINANCE administrative staff Secretariats
Administrative Staff CHANGE CHANGE
 Major reform  Completed albeit with  Administrative restruc-
 High impact some delay in 2013 turing took place with
some delay in 2014
PERFORMANCE  Appoint President of  Appoint HQAA members CHANGE CHANGE
Evaluation and HQAA  Appoint HQAA Director  Initial evaluation of  Accreditation and vali-
Transparency CHANGE General Greek Universities com- dation processes sched-
 Major reform  Completed  Publication of University pleted with some delay uled to be completed on
 Medium-high impact Quality Assurance Internal by 2013–14 time
Structures
 HQAA Charter
CHANGE
 Completed
PERFORMANCE  Convene panel of
Evaluate Reforms experts to assess
included in the law reforms enacted by the
 Major reform law (2015)
 High impact  Submit evaluation
reports (2016)
NO CHANGE
 Plans currently on hold
pending implementation
of the latest amend-
ments in September 2014
Policy Studies Journal, 44:1
Zahariadis/Exadaktylos: Policies that Succeed and Programs that Fail 73

demonstrate the level of resistance to the status quo. More generally, the centralized
nature of Greek higher education (Kyriazis & Asderaki, 2008) assured that legal chal-
lenges caused a chain reaction of delays in implementation. For example, challenges
to the constitutionality of new Councils on the basis of university self-governance
meant elections for new rectors could not proceed because the law states only new
Councils could propose rector candidates. Corroborating previous findings (e.g.,
Kavasakalis, 2013) there was widespread agreement among Greek academics on
reform to escape the state stranglehold in higher education. The irony is opposing
coalitions used the same distaste over centralization to both support the need for
reform and oppose specific changes.
Our expectation that institutional rule manipulation in a centralized environ-
ment would lead to implementation failure is borne by the evidence. Centralization
of Greek higher education ensured failure would spill over across institutions, with
the biggest institutions taking the lead. The interviewee from Democritus University
argued: “the setting up of new Councils at the National University of Athens, Aris-
totle University, and the Athens Polytechnic School were blocked by University
administrations, often helped by student unions affiliated with political parties inside
the Universities. At the same time, when new Councils were set up, student occupa-
tion of university buildings prevented the new bodies from performing their duties.”
It should be noted these three universities are the biggest (in terms of student enroll-
ment) and most prestigious institutions in Greece. In addition, inability to create new
University Charters as envisioned by the Law (several were drafted but none was
adopted by January 2015) has robbed proponents from securing potential benefits
that could empower new internal constituencies.
Success in the implementation phase was not guaranteed despite the law being
adopted by a large majority in Parliament. The new policy brought dramatic changes
to the established state of play in Greek universities, which the government did not
anticipate. For one, the winning coalitions between the two stages of the process were
completely different. Different stakeholders coalesced against implementation, espe-
cially at the microlevel, partly because frequent changes at the helm of the Ministry of
Education (there have been five ministers between January 2012 and January 2015)
provided fodder to opponents with inconsistent messages. An interviewee conjectured
that despite rhetoric to the contrary ministers actually favor centralization and under-
mine their own reforms for political gain. Supporting the statement, a former univer-
sity rector told us Greek politicians routinely undermine enacted changes for fear of
losing political control, illustrating the symbolic elements of implementation. Politi-
cians often lack the political will to mobilize support, sustain effort, or apply credible
sanctions (Brinkerhoff, 2000) partly conditioned by electoral fortune. Besides, as March
and Olsen (1983, p. 290) assert, reforms symbolize the possibility of action and belief in
“that possibility may be of greater [political] significance than the execution of it.”
Seeking to bolster his/her legacy and political fortune, each minister reinter-
preted the law according to partisan criteria (Saiti, 2013, p. 6). For example, Giorgos
Babiniotis, former rector of the National University, called the law “unworkable”
(Mavrogordatos, 2012) and decoupled funding from implementation by ruling
within the first two weeks of his tenure as Minister of Education in 2012 that
74 Policy Studies Journal, 44:1

universities would be eligible for state funding without applying Law 4009
(Marseilles, 2012). Another, Andreas Loverdos (2014) who is also a law school
professor, immediately after becoming minister in June 2014 declared he was
not bound by the education budget because he did not vote for it and went
against government policy by refusing to lay off administrative staff. Corrobo-
rating our hypothesis the partial failure of implementation and considerable
delays were due to inconsistent political messages delivered by those in charge.
Because no Councils were elected by the 2012 deadline except for one institution,
the Ministry relented and gave extensions for the process to take place. Whereas
originally the aim of Councils was to link universities to the broader social environ-
ment, rectors and students alike decoupled the solution (Council) from its problem
(lack of embeddedness in social environment). Instead, they objected to outside,
using terms “political” or “oligarchic,” interference arguing for members to be inter-
nal academics because they knew local conditions (interview with faculty member in
Athens) (Gounari, 2012, p. 282; Mylopoulos, 2012). As a result, the majority of Coun-
cil members are internal; they elect the external minority (Art. 8, par. 5b as amended
by 4076/2012).
Indeed, some professors and rectors viewed opposition as resistance to orders
by foreign powers. Being part of a “broader neoliberal scheme” (e.g., Gouvias, 2012)
to introduce “privatization of the Greek public university” (Gounari, 2012, p. 280),
Ioannis Mylopoulos (2012), then rector of Aristotle University, declared the law to be
“a witch-hunt” detrimental to the academy. In line with expectations, the broader
crisis increased the chances of links among disparate issues, undermining implemen-
tation. Leaving aside the fact that the law did not call for the establishment of private
universities, his argument is important largely for its political implications. The rec-
tor discounted the law’s legitimacy because it was the product of agreement, as he
alleged, between the elites of the three coalition government parties at the time
(PASOK, New Democracy, and the right-wing Popular Orthodox Rally [LAOS]) and
did not adequately reflect the balance of forces in politics. The fact that his essay
appeared in the newspaper of the main opposition party at the time, the Radical Left
(SYRIZA), clearly added a partisan dimension to his allegation. It took three years
for these obstacles to be lifted and most universities have now elected Councils
although the latter are involved in bitter turf battles with faculty senates over the
limit and nature of Council oversight powers (interview with faculty member).

Side Payments and Finance Concerns

As passed, the law called for changes in administration (new rectors elected by
newly constituted Councils) to be implemented by the academic year 2012–13, other-
wise funding would be withdrawn. Under crisis conditions, our hypothesis predicts
the strategy of side payments holding the policy adoption coalition together will
likely fail in implementation. Clearly, the stick of withdrawing side payments that
held the coalition together during adoption proved detrimental during implementa-
tion. Even faculty who supported the spirit of the law objected to what they
Zahariadis/Exadaktylos: Policies that Succeed and Programs that Fail 75

perceived to be coercion because it did not exhibit “sensitivity to the democratic pro-
cess of electing university leadership” (interview with faculty member of the Aris-
totle University). In fact, resistance was so strong that student unions held in some
cases members of the Council captive in their offices, threatening and harassing
them if they took up their new duties as a sign of accepting the legitimacy of the law
(interview with affected member of the University of Macedonia). In addition, the
conference of rectors demanded immediate restoration of public funding to higher
education and decoupling of funding from the assessment exercise envisioned in the
Law (Mylopoulos, 2012). It was additional budget cuts concurrent with the law and
the crisis that fueled suspicions (and opposition) in some quarters that “the ‘reform’
of public education is in reality a massive new profit opportunity for business”
(Gounari, 2012, p. 283).
A thorny financial issue, the case of “eternal” registration of students, has been
met with stiff resistance by student associations, political parties, as well as schools
and departments. The law specified students enrolled in universities before 2003–04
had until September 2014 to complete their studies. As many interviewees con-
curred, opposition by rectors and students centered on the alleged unfairness of
dropping students who could not attend classes due to financial exigencies caused
by the crisis (Marseilles, 2014). Symbolically and conveniently missing was the fact
that the de-enrollment provision was not new but first introduced in a 2007 law. Fig-
ures from the Ministry of Education show the number of inactive (so-called
“eternal”) students has risen steeply since 2004 long before the onset of the 2010 cri-
sis (Elstat, 2014). Implementation clearly altered perceptions of equity in the policy
stream.
Confirming our hypothesis, institutional changes led to stiff resistance, which
the Ministry has attempted to assuage through side payments. Amendment to the
Law by Minister Loverdos states 2014–15 is the last academic year when the system
of eternal registration and nonmerit-based transfer of students is permissible. How-
ever, because previous reform attempts had limited or negative effects, affected
groups did not trust the government to create a “just” framework for reform
(cf. Exadaktylos & Zahariadis, 2014).

Issue Linkages and Problems with Personnel

Confirming our expectation, crisis conditions helped reframe problems and

facilitated issue-linkages with different solutions, increasing the chances of imple-
mentation failure. Regarding changes to promotion and ranks, the broader crisis
helped link disparate issues, decoupling solutions from problems, reframing prob-
lems, intensifying policy conflict, and eventually undermining implementation (Saiti,
2013). The points validate the MSA view that policymaking involves complex inter-
actions of heterogeneous problems and solutions. As reformers are hard-pressed to
sustain political attention, “less central actors come to the forefront” (March & Olsen,
1983, p. 286). An interviewee from the National University of Athens who was
recently appointed as professor abroad emphasized strong reactions to rationalizing
76 Policy Studies Journal, 44:1

salary scales, altering promotion ranks, and forcing full professors to teach more
classes were expressed by established interests inside the universities. He claimed
the threatened legal action at the Council of State would postpone implementation of
these articles too far into the future; so the Ministry decided to scrap it altogether.
“Absenteeism from teaching is a perennial problem,” explained political science Pro-
fessor George Mavrogordatos (2012), “but Greek professors are seldom, if ever, held
accountable.” Another university interviewee informed us he opposed the law
because it did not curb enough the power of department chairs. Still another
deplored what she expected to be a power shift from faculty to administration
(Gounari, 2012, p. 281). It is obvious that coalitions of support or opposition to
radical change are tenuously held together by contextual factors subject to fluid
views of fairness and equity (Sætren, 2015).

Institutional Rule Manipulation and Performance Measurement

Finally and confirming the counterfactual, while accreditation of universities

and validation of programs of study according to HQAA standards has been met
with some resistance, there was little opposition to the general system of evaluating
quality of studies and academic staff. Despite isolated concerns that “‘quality’ and
‘excellence’ serve as the excuse for a corporate standardization of the university, as
well as academic life and thought” (Gounari, 2012, p. 280), when explained to stu-
dents who attempted an intrusion to the process, a quality assurance team member
told us, it was widely accepted as a worthy practice. Assessment pursued the spirit
of broader European reforms to make Greek higher education more transparent,
accountable, and outward looking (Asderaki, 2009). Political coercion worked
because HQAA began work before the crisis erupted in 2010 (Papadimitriou, 2011)
with supportive university coalitions already in place. Besides, the number of imple-
mentation decisions needing to be (re)negotiated was relatively small increasing the
likelihood of implementation success (Pressman & Wildavsky, 1984). The 2011 law
left little room for reframing and reinforced the same message by reiterating previ-
ously made linkages between funding and quality assurance, thereby sustaining the
winning coalition during implementation.

Conclusion

Why do some policies adopted by a wide margin fail to be implemented?

Emphasizing the role of policy entrepreneurial strategies in MSA, we have analyzed
Greek higher education reforms to find when policies adversely affect the status quo
successful entrepreneurial strategies of issue-linkage and framing, side payments,
and institutional rule manipulation are more likely to lead to implementation failure
under conditions of crisis, centralized monopoly, and inconsistent political commu-
nication. Although our study is temporally limited and needs empirical validation
across more cases, the findings have preliminary implications for MSA and the study
of implementation and policy reform.
Zahariadis/Exadaktylos: Policies that Succeed and Programs that Fail 77

We stress an interactive, dynamic approach to MSA that links choices and insti-
tutional arenas to policy outcomes. To do so, we focused on entrepreneurial strat-
egies in implementation pursued by coalitions built during policy adoption. Our
argument has two implications for MSA theorizing. First, the presence of an imple-
mentation window injects more contingency in the policy process and alters the aim
of coupling by seeking to prevent decoupling. As Aberbach and Christensen (2014,
p. 9) inform, “the implementation phase is an opportunity to further new problems
or solutions that may create conflicts or ambiguity.” This does not preclude the open-
ing of “normal” policy windows in the problem or politics streams. Changes in the
political stream, such as ministerial turnover, affect receptivity facilitating decoupling
and increasing the chances of failure. For example, inconsistent political messages
amidst the broader financial crisis generated intense ambiguity and distorted percep-
tions of equity, altering the carefully calibrated balance achieved during policy
formation.
In MSA terms, our claim implies coupling efforts occur not only during policy
formation but also during implementation. In this context, coupling takes a different
direction. Whereas MSA conceptualizes the main task of coupling as the joining of
three streams, we amend the argument to stress the aim is to prevent decoupling in
implementation. MSA assumes independent problem and policy streams, but we
began implementation analysis with closely linked streams and specified conditions
that prompted decoupling and recoupling. Ridde (2009, pp. 947–48) maintains suc-
cessful implementation presupposes the coupling of at least the problem and policy
streams in the presence of a favorable political stream. We confirm and amend his
argument by finding implementation also utilizes the same strategies to actively
decouple the problem from the policy streams. Intense conflict and ambiguity helped
decouple reforms from their problems and encouraged entrepreneurs to reframe
reforms as an assault to the essence of the Greek public sector (interview with uni-
versity faculty member). Decoupling increases the chances of failure, while a
strongly favorable political stream is a necessary but not sufficient condition of suc-
cess, as the Greek case clearly shows.
Second, we take the MSA literature one step further by helping to illuminate the
causal mechanism behind the puzzle of “an idea whose time has come.” Entrepre-
neurship must not only be analyzed in context, as Mintrom and Norman (2009)
rightly point out, but it should also adapt to the different circumstances and actors
that distinguish policy formation from implementation. Entrepreneurs alone are not
enough to bring about reform; we also need to account for context. Seeking to shift
MSA focus from entrepreneurs to entrepreneurship, Ackrill et al.’s (2013) call for
specification of conditions why skilled advocacy or brokering leads to reform in
some circumstances but not others. Otherwise, we are left with “heroic” but idiosyn-
cratic explanations of policy outcomes. Using strategy to infuse context into agency,
we were able to clarify some conditions of entrepreneurial success or failure in
implementation. Crisis, centralized monopoly, and inconsistent political communica-
tion loosen the organic link between formation and implementation. Strategies that
work well in one phase may fall flat in the other. Future MSA research may clarify
further links between conditions and outcomes by conceptualizing entrepreneurship
78 Policy Studies Journal, 44:1

not as a stable characteristic distributed among differentiated policy agents but as

strategies in specific policymaking situations.
The activation of a new set of actors during implementation—university admin-
istration, professors, and students (and through them political parties)—likely
undermines the successful entrepreneurial coupling strategies of issue linking and
framing during policy implementation. Patashnik (2008) maintains the key to suc-
cess lies in reorienting the playing field and giving newcomers a stake in the
reformed status quo. We confirm and refine the argument to add that local coali-
tions make the difference. Bureaucrats can act entrepreneurially to derail implemen-
tation (Arnold, 2015). Political support in Greece never translated into bureaucratic
action (hence the ambiguity) and did not incorporate stakeholders as part of the con-
sultation, so resistance to change was to be expected due to the pending conflict.
Whereas the national coalition built to adopt the law demonstrated the strength of
the government but also brought forward other political forces for accountability
purposes, the government monopolized implementation through bargaining, and
often coercion in top-down fashion (e.g., Van Meter & Van Horn, 1975) without tap-
ping into pockets of reform support or building local coalitions with the universities
themselves.
As a result, the same coupling strategies that worked well when applied to pol-
icy formation hindered implementation because the latter privileged a different set
of actors. We conclude that while the aim of entrepreneurs remains the same the
strategies might possibly differ across stages of the policy process because of differ-
ent contexts. Future research might ask whether implementation success involves
entrepreneurial strategies different from those used in policy formation or similar
strategies but under different conditions from those probed here.
Whereas the literature divides implementation into distinct types with appropri-
ate strategies (e.g., Matland, 1995), we find fluid boundaries and movement across
types. Greek higher education reform had elements of both political (low ambiguity–
high conflict) and symbolic (high ambiguity–high conflict) implementation. In some
instances, coercion failed (e.g., new University Councils in the first year of imple-
mentation or department closures) but in others it succeeded (e.g., HQAA). Amend-
ing Ridde’s (2009, p. 948) finding, in all cases local coalitions proved vital to success
(or failure) even in the presence of strong central (i.e., national political) support.
Implementation success is contingent, but we observe over time a drift from political
toward symbolic implementation. Inability (or unwillingness) to succeed not only
exacerbated policy conflict over time but also signaled opportunities to widen its
reach to the broader environment, reframe problems, and abandon adopted solu-
tions. Reminiscent of the logic of policy drift (Hacker, 2004), the resultant ambiguity
empowered local coalitions to demand further modifications hoping new macrolevel
participants in the politics stream would legitimize them. The observed drift toward
symbolic implementation resembles the institutional drift expounded by historical
institutionalists (Mahoney & Thelen, 2010) in that it captures the dissonance created
by blocked adaptation and the changes demanded by the political landscape. Instead
of policy reform being actively shaped by its political environment, cosmetic changes
are made but the core structure remains the same until the environment changes to
Zahariadis/Exadaktylos: Policies that Succeed and Programs that Fail 79

once again fit more closely to existing policy. The new SYRIZA-led government
(since January 2015) demonstrates this point: it proposed in April 2015 the reversal
of key reforms—relating among others to University Councils, rector elections, and
“eternal” students—and return to the previous status quo (Hope & Barber, 2015).
Finally, our findings help explain why there is so much movement in policy
reform but relatively little movement forward. Brunsson (1989) describes two distinct
spheres to explain organizational behavior: talk (policy formation in our case) and
action (implementation). He shows when organizations face problematic environ-
mental demands they decouple talk from action to contain political conflict and still
produce actionable results. Our findings validate and amend his argument. We, too,
attribute the paradox of wide agreement and little implementation to the interaction
between diverse coalitions during policy formation and implementation. We further
specify, however, that under conditions of crisis, centralization, and inconsistent
political messages, interaction breaks down, ambiguity increases, and conflict spills
over from policy formation to implementation. Spillover effects occur not only in
agenda-setting, as MSA predicts, but also in implementation through the mechanism
of activating new actors whose actions refract the effectiveness of coupling strategies.
Whereas power is given to agents to buffer policy from political friction during the
formation stage, the same power undermines implementation success. Political con-
flict spreads to implementation so that there is ambiguity and perennial change in
the form of new laws passed but very little new happening. MSA shows that failure
is sometimes not the consequence of contentious decisions but the cause of poorly
implemented but widely agreed upon choices.

Nikolaos Zahariadis is professor of political science at the Department of Govern-

ment, University of Alabama at Birmingham, USA.
Theofanis Exadaktylos is lecturer in European politics at the School of Politics,
University of Surrey, UK.

Note

The authors thank Harald Sætren and workshop participants for valuable comments and suggestions.

References

Aberbach, Joel D., and Tom Christensen. 2014. “Why Reforms So Often Disappoint.” American Journal of
Public Administration 44 (1): 3–16.
Ackrill, Robert, Adrian Kay, and Nikolaos Zahariadis. 2013. “Multiple Streams and the EU Policy
Process.” Journal of European Public Policy 20 (6): 871–87.
Arnold, Gwen. 2015. “Street-Level Policy Entrepreneurship.” Public Management Review 17 (3):
307–27.
Asderaki, Foteini. 2009. “The Impact of the Bologna Process on the Development of the Greek Quality
Assurance System.” Quality in Higher Education 15 (2): 105–22.
Baier, Vicki E., James G. March, and Harald Sætren. 1986. “Implementation and Ambiguity.” Scandina-
vian Journal of Management Studies 2 (3–4): 197–212.
80 Policy Studies Journal, 44:1

Baumgartner, Frank R., and Bryan D. Jones. 1993. Agendas and Instability in American Politics. Chicago:
University of Chicago Press.
Boin, Arjen, Paul ‘t Hart, and Allan McConnell. 2009. “Crisis Exploitation: Political and Policy Impacts of
Framing Contests.” Journal of European Public Policy 16 (1): 81–106.
Brinkerhoff, Derick W. 2000. “Assessing Political Will for Anti-corruption Efforts: An Analytic Frame-
work.” Public Administration and Development 20: 239–52.
Brouwer, Stijn, and Frank Biermann. 2011. “Towards Adaptive Management: Examining the Strategies of
Policy Entrepreneurs in Dutch Water Management.” Ecology and Society 16 (4): 5.
Brunsson, Nils. 1989. The Organization of Hypocrisy: Talk, Decisions, and Actions in Organizations. New
York: Wiley.
Diamantopoulou, Anna. 2011. “Anna Diamantopoulou: The Role of Education in Greece’s Economic
Recovery.” 2011 Public Lecture at the London School of Economics [Online]. http://www.lse.ac.uk/
europeanInstitute/research/hellenicObservatory/Events/pubLectures/Diamantopoulou_2011.aspx.
Accessed August 15, 2014.
Elstat. 2014. Tertiary Education (in Greek) [Online]. http://www.statistics.gr/portal/page/portal/ESYE/
PAGE-themes?p_param5A1403. Accessed August 1, 2014.
Exadaktylos, Theofanis, and Nikolaos Zahariadis. 2014. “Quid pro Quo: Political Trust and Policy Imple-
mentation in Greece during the Age of Austerity.” Politics & Policy 42 (1): 160–83.
Fimister, Geoff, and Michael Hill. 1993. “Delegating Implementation Problems: Social Security, Housing
and Community Care in Britain.” In New Agendas in the Study of the Policy Process, ed. Michael Hill.
Hemel Hempstead, UK: Harvester Wheatsheaf, 110–29.
Gamson, William A. 1961. “A Theory of Coalition Formation.” American Sociological Review 26 (3): 373–82.
George, Alexander L., and Andrew Bennett. 2005. Case Studies and Theory Development in the Social Scien-
ces. Cambridge, MA: The MIT Press.
Gounari, Panayota. 2012. “Neoliberalizing Higher Education in Greece: New Laws, Old Free-Market
Tricks.” Power & Education 4 (3): 277–88.
Gouvias, Dionyssios. 2012. “The Post-Modern Rhetoric of Recent Reforms in Greek Higher Education.”
Journal for Critical Education Policy Studies 10 (2): 282–313.
Hacker, Jacob. 2004. “Privatizing Risk without Privatizing the Welfare State: The Hidden Politics of Social
Politics Retrenchment in the United States.” American Political Science Review 98: 243–60.
Hajer, Maarten. 2003. “Policy without Polity? Policy Analysis and the Institutional Void.” Policy Sciences
36: 175–95.
Hope, Kerin, and Tony Barber. 2015. “Faces behind Greece’s Radical Government.” Financial Times (April
21) [Online]. http://www.ft.com/cms/s/0/f00dfbc8-e4d6-11e4-8b61-00144feab7de.html#axzz3Yt9Jrsrx.
Accessed May 1, 2015.
Howlett, Michael, Allan McConnell, and Anthony Perl. 2015. “Streams and Stages: Reconciling Kingdon
and Policy Process Theory.” European Journal of Political Research 54 (3): 419–34.
Huber, John D., and Charles R. Shipan. 2002. Deliberate Discretion? The Institutional Foundations of Bureau-
cratic Autonomy. New York: Cambridge University Press.
Ingram, Helen, and Anne Schneider. 1990. “Improving Implementation through Framing Smarter Stat-
utes.” Journal of Public Policy 10 (1): 67–88.
Kavasakalis, Aggelos. 2013. “Are Common Beliefs Present among Greek Academics during the
Implementation of a Controversial University Policy?” International Research in Education 1 (1):
5–28.
Kingdon, John W. 1995. Agendas, Alternatives, and Public Policies, 2nd ed. New York: Harper Collins.
Knaggård, Åsa. 2015. “The Multiple Streams Framework and the Problem Broker.” European Journal of
Political Research 54 (3): 450–65.
Kyriazis, Athanasios, and Foteini Asderaki. 2008. Higher Education in Greece. Bucharest, Romania:
UNESCO–CEPES.
Lakasas, Apostolos. 2015. “Greek Universities Are Heading toward a Swamp.” Kathimerini (in Greek).
June 21: 15.
Zahariadis/Exadaktylos: Policies that Succeed and Programs that Fail 81

Levin, Martin, and Barbara Ferman. 1986. “The Political Hand: Policy Implementation and Youth
Employment Programs.” Journal of Policy Analysis and Management 5 (2): 311–25.
Lipsky, Michael. 2010. Street-Level Bureaucracy. New York: Russell Sage Foundation.
Loverdos, Andreas. 2014. “Loverdos: I Cannot Implement a Budget I have Voted against.” To Vima (in
Greek) (June 17) [Online]. http://www.tovima.gr/vimafm/interviews/article/?aid5606743. Accessed
October 10, 2014.
Mahoney, James, and Kathleen Thelen. 2010. “A Theory of Gradual Institutional Change.” In Explaining
Institutional Change: Ambiguity, Agency, and Power, ed. James Mahoney, and Kathleen Thelen. Cam-
bridge: Cambridge University Press, 1–37.
March, James G., and Johan P. Olsen. 1983. “Organizing Political Life: What Administrative Reorganiza-
tion Tells Us about Government.” American Political Science Review 77: 281–97.
Marseilles, Makki. 2012. “Universities Face Huge Losses to Reserves after Bond ‘Haircut’.” University World
News (April 2) [Online]. http://www.universityworldnews.com/article.php?story520120402213156178.
Accessed October 10, 2014.
———. 2014. “‘Stagnant’ Students to be Struck off Enrolment Register.” University World News (August 29)
[Online]. http://www.universityworldnews.com/article.php?story520140828170855934. Accessed
November 6, 2014.
Matland, Richard E. 1995. “Synthesizing the Implementation Literature: The Ambiguity-Conflict Model
of Policy Implementation.” Journal of Public Administration Research and Theory 5 (2): 145–74.
Mavrogordatos, George T. 2012. “Styx and Stones.” Times Higher Education (May 24) [Online]. http://
www.timeshighereducation.co.uk/features/styx-and-stones/420077.article. Accessed November 15,
2014.
McBeth, Mark K., Michael D. Jones, and Elizabeth A. Shanahan. 2014. “The Narrative Policy Frame-
work.” In Theories of the Policy Process, 3rd ed., ed. Paul A. Sabatier, and Christopher M. Weible.
Boulder, CO: Westview Press, 225–66.
Meijerink, Sander, and Dave Huitema. 2010. “Policy Entrepreneurs and Change Strategies: Lessons from
Sixteen Case Studies of Water Transitions Around the Globe.” Ecology and Society 15 (2): 21.
Mintrom, Michael. 2000. Policy Entrepreneurs and School Choice. Washington, DC: Georgetown University
Press.
Mintrom, Michael, and Phillipa Norman. 2009. “Policy Entrepreneurship and Policy Change.” Policy
Studies Journal 37: 649–67.
Mylopoulos, Ioannis. 2012. “6 1 1 Changes to Restore Democratic Legitimacy and the Functioning of the
Academy” (in Greek) (July 22) [Online]. http://mylopoulos.gr/node/481. Accessed August 15, 2014.
Olsen, Johan P. 2006. “Maybe it’s Time to Rediscover Bureaucracy.” Journal of Public Administration Theory
and Practice 16 (1): 1–24.
———. 2007. “The Institutional Dynamics of the European University.” In University Dynamics and Euro-
pean Integration, ed. Peter Maassen, and Johan P. Olsen. Dordrecht, the Netherlands: Springer, 25–54.
Organization for Economic Cooperation and Development (OECD). 2011. Successful Performers and Suc-
cessful Reformers in Education: Education Policy Advice for Greece. Paris: OECD.
Papadimitriou, Antigoni. 2011. “Reforms, Leadership and Quality Management in Greek Higher Educa-
tion.” Tertiary Education and Management 17 (4): 355–72.
Patashnik, Eric M. 2008. Reforms at Risk: What Happens after Major Policy Changes are Enacted? Princeton,
NJ: Princeton University Press.
Pollitt, Christopher. 2008. Time, Policy, Management: Governing with the Past. Oxford, UK: Oxford Univer-
sity Press.
Pollitt, Christopher, and Geert Bouckaert. 2011. Public Management Reform: A Comparative Analysis – New Pub-
lic Management, Governance, and the Neo-Weberian State, 3rd ed. Oxford, UK: Oxford University Press.
Pressman, Jeffrey L., and Aaron Wildavsky. 1984. Implementation, 3rd ed. Berkeley: University of Califor-
nia Press.
Ridde, Valery. 2009. “Policy Implementation in an African State: An Extension of Kingdon’s Multiple
Streams Approach.” Public Administration 87 (4): 938–54.
82 Policy Studies Journal, 44:1

Riker, William H. 1962. The Theory of Political Coalitions. New Haven, CT: Yale University Press.
Robbins, Suzanne M. 2010. “Play Nice or Pick a Fight? Cooperation as an Interest Group Strategy at
Implementation.” Policy Studies Journal 38 (3): 515–35.
Roberts, Nancy C., and Paula J. King. 1991. “Policy Entrepreneurs: Their Activity Structure and Function
in the Policy Process.” Journal of Public Administration Research and Theory 1 (2): 147–75.
Rochefort, David A., and Roger W. Cobb, eds. 1994. The Politics of Problem Definition: Shaping the Policy
Agenda. Lawrence, KS: University Press of Kansas.
Rommetveit, Kåre. 1976. “Decision Making under Changing Norms.” In Ambiguity and Choice in Organiza-
tions, ed. James G. March, and Johan P. Olsen. Bergen, Norway: Universitetforlaget, 140–55.
Sætren, Harald. 2015. “Crucial Factors in Implementing Radical Policy Change: A Comparative Longitu-
dinal Study of Nordic Central Agency Relocation Programs.” Journal of Comparative Policy Analysis
17 (2): 103–23.
Saiti, Anna. 2013. “Reforms in Greek Education 1991-2011: Reforms or Something Else?” eJournal of Educa-
tion Policy Spring: 1–10.
Saurugger, Sabine. 2012. “Beyond Non-Compliance with Legal Norms.” In Research Design in European
Studies, ed. Theofanis Exadaktylos, and Claudio Radaelli. New York: Palgrave Macmillan, 105–24.
Schattschneider, Elmer Eric. 1960. The Semisovereign People: A Realist’s View of Democracy in America. New
York: Holt, Rinehart and Winston.
Schneider, Mark, Paul Teske, and Michael Mintrom. 1995. Public Entrepreneurs: Agents for Change in Ameri-
can Government. Princeton, NJ: Princeton University Press.
Schoenenberger, Alain M. 2005. “Are Higher Education and Academic Research a Public Good or a Pub-
lic Responsibility? A Review of the Economic Literature.” In The Public Responsibility for Higher Edu-
cation and Research, ed. Luc Weber, and Sjur Bergan. Strasbourg, France: Council of Europe, 45–94.
Teisman, Geert R. 2000. “Models for Research into Decision-Making Processes: On Phases, Streams and
Decision-Making Rounds.” Public Administration 78 (4): 937–56.
Tsiligiris, Vangelis. 2012. “Grexit: Another Lost Opportunity for Greek Higher Education?” University World
News (June 3) [Online]. http://www.universityworldnews.com/article.php?story520120530125539702.
Accessed October 10, 2014.
Van Meter, Donald S., and Carl E. Van Horn. 1975. “The Policy Implementation Process: A Conceptual
Framework.” Administration and Society 6 (4): 445–88.
Wilson, James Q. 1989. Bureaucracy. New York: Basic Books.
Yin, Robert K. 2014. Case Study Research: Design and Methods, 5th ed. Thousand Oaks, CA: Sage.
Zahariadis, Nikolaos. 2003. Ambiguity and Choice in Public Policy. Washington, DC: Georgetown Univer-
sity Press.
———. 2008. “Europeanization as Program Implementation: Effective and Democratic?” Journal of Com-
parative Policy Analysis 10 (3): 221–38.
———. 2013. “Leading Reform amidst Transboundary Crises: Lessons from Greece.” Public Administra-
tion 91 (3): 648–62.
———. 2014. “Ambiguity and Multiple Streams.” In Theories of the Policy Process, 3rd ed., ed. Paul A.
Sabatier, and Christopher M. Weible. Boulder, CO: Westview Press, 25–58.