Governance and State Capacity in
Post-communist states
Lars Johannsen and Ole Nørgaard
Department of Political Science
University of Aarhus
Denmark
Paper prepared for Workshop no. 2, ‘Political transformation in Soviet successor
states: the politics of CIS states in comparative and theoretical perspective’
ECPR Joint Sessions of Workshops,
Institute of Political Studies, Grenoble
First draft, April 2001
1. Governance and State Capacity in Post-Communist Countries
Anti-state attitudes came to epitomize the gist of public policy in Central and
Eastern Europe during the first years after the systemic changes as anticipated. The
collective experience of political repression, economic inefficiency and the
administrative volatility of the totalitarian state prepared a fertile political ground for
anti-statist attitudes. Politicians and institutional decision-makers alike expected that
the institutions of democracy and the market would be quite easy to construct on the
ashes of communism. Good governance, from this perspective, was identical to good
formal institutions, because the collapse of communism had allegedly produced an
institutional vacuum so that new institutions could be made from scratch. Individual
and collective actors were then supposed to adapt to the new rules of the game, at
that time defined as the rules of the free market. This perspective, based on the assumptions of public or rational choice, soon proved inadequate. Similar institutions
in different counties produced dissimilar outcomes, implying that the local context
affected the performance of new institutions. Moreover, the social costs of transition
in conjunction with institutional and administrative instability soon gave rise to
demands for state intervention and institutional and administrative reform.
These demands coincided with a shift towards more state oriented policies
that arose in the major financial institutions from the mid-1990s, (The World Bank,
1997) a shift which in itself was partly provoked by the innovative perspectives on
the role of the state that had been developed by historians, sociologists and development economists since the mid-1980s (Leftwich, 1997; Weiss, 1998; Evans, 1995;
Evans et al., 1979; Krasner, 1984; Skocpol, 1979). Different perspectives accompanied this renewed focus on the state in post-communist countries. One such perspective focus on the declining ability of the state to provide basic services for its
citizens caused by financial constraints and inefficient tax collection (Schleifer and
Vishney, 1999; Schleifer and Treisman, 2000; Solniki, 1998). Another perspective
examines the capacity of incumbent state institutions and officials to adapt to the
changed technical and administrative demands generated by the systemic transformation (Ágh, 1997; Nunberg, 1999 & 2000; UNDP, 1997; Horváth, 2000;
Galligan and Smirlov, 1997). A third perspective on state capacity deals with the
ability of state institutions to respond to and integrate political demands, mediate
conflicts and maintain popular legitimacy while maintaining their autonomy and
capacity to govern. The point where society, politics and institutions meet designates
what can be termed ‘governance’. ‘Good governance’ requires a plus sum reciprocity between state and society that produces popular trust and support for institutions
while asserting the primacy of national objectives over interests and demands any
particular group. Using this perspective as a starting point, the main argument of this
paper is that the character of the state-institutional matrix is what ultimately induces
virtuous or vicious circles of good or poor governance.
1
The paper addresses the issues of governance and state capacity in Central
and Eastern Europe. It asks why some countries better than others have been able to
achieve good governance. Section two introduces the analytical framework and
explains the concepts of governance and capacity. Based on statistical analyses of
survey data, section three maps the relationship between perceived performance;
trust in political and administrative institutions and state-society relationship in the
post-communist countries. Section four searches for causal models in a comparative
case study of Latvia and Lithuania, two countries that epitomize many of the
governance problems besetting post-communism, demonstrating the effects of
historical contingencies on institutions and state-society relations. Section five
summarizes the findings and proposes an agenda for further research.
2. The analytical framework and explanations of governance and capacity
The term ’governance’ describes a state’s ability to steer and manage. Good governance concerns a state’s ability to respond to and integrate political demands, mediate
conflicts and maintain popular legitimacy while maintaining autonomy and capacity
to govern. In this dynamic view, good governance is the result of a positive
reciprocity between the state and society. In this respect, good governance is more
than hierarchies and administrative and political institutions, although these remain
important because they structure the functions of the state (Pierre and Peters, 2000,
p. 22). Thus, good governance is found in the interactions between structures.
We propose that good or less good governance is the end result of virtuous or
vicious circles comprising perceived performance, trust in and legitimacy of political
and administrative institutions and the way in which state and society interacts. This
proposition is visualized in Figure 1.
Perceived performance is to be understood as popular perceptions of the
government and its ability to meet and respond to public expectations. Popular
perceptions as such express an evaluation of governance in the country in question.
On the one hand, dissatisfied citizens who evaluate performance negatively do not
trust political and administrative institutions to meet and respond to their demands
and bestow the regime with little legitimacy. Satisfied citizens, on the other hand,
trust political and administrative institutions and provide performance-based support
for the regime. Good governance presupposes democratic participation and democratic participation presuppose trust in institutions, thus generating virtuous or vicious
circles of governance. Democratic participation, however, is more than mere electoral participation. It also encompasses the participation of civil society organizations
in the formulation and implementation of policies. A civil society autonomous from
the state was not tolerated in the totalitarian communist state,1 and much of the
interest in the reforms during the transition in the 1990s was focused on the way in
which the state exerted influence on the society and the economy.
2
Figure 1: Virtuous and Vicious circles.
Virtuous circle
Perceived
Performance
Satisfaction
and
dissatisfaction
State/society
interaction
Trust in/
Legitimacy of
political and
administrative
institutions
Participation
or
Capture
Good
Governance
Poor
Governance
Vicious circle
The state, however, must also maintain some degree of autonomy from interest
groups seeking influence if it is to pursue the public interest and keep up national
priorities (Kjær, Frølund Thomsen and Hersted Hansen, 2000). In contrast, the state
is ‘captured’ if decisions are made to appease specific interests as opposed to the
public interests aggregated and mediated through democratic processes. 2 State
capture differs from other more conventional forms of political influence because it
implies that specific groups and firms have total control over the policy process. The
policy elites have become captives of the groups and firms they are supposed to
control – rather than conceptual captives of the requirements and rationality of their
bureaucratic positions. State capture thus undermines a transparent and legitimate
democratic policy process by reducing the access of rival interests to the state,
thereby setting in motion, as we propose here, vicious circles that degenerate into
poor governance.
Thus, while Figure 1 summarizes the hypothesized relationship between the
parameters of the present study, it does not explain why one country enters the
virtuous circle and another the vicious circle. This paper argues that the conflict
between new and old institutions explains why countries take one or the other path
of governance. Traditional institutions are those that are inscribed into popular
attitudes, values and expectations. New institutions are formal arrangements (laws,
decrees, organizations, court decisions etc.) introduced in the course of reforms. To
understand what determines the outcome of this conflict between new and old
institutions we apply the perspectives of two traditions within (new) institutional
theory: historical and empirical institutionalism.
3
Historical institutionalism sees institutions as historically determined rules,
norms, values and expectations that, at later historical stages, may independently
affect outcomes when actors become wedged between old and new standards of
behaviour (Thelen and Steinmo, 1992; Peters, 1999; Nørgaard, 2000). In the present
context, we can distinguish between four tiers of historically determined institutions:
first, pre-communist institutions whose mentalities or societal structures outlasted
the communist epoch. They may consist of general mentalities or values that render
societies more or less permeable to new ideas and values. This tier of historically
determined institutions is specific to each country or region and may influence
subsequent developments by providing focal points for national identities. Second,
communist institutions are common heritage in all post-communist countries. Yet,
while all communist states were ‘totalitarian’ in the sense that they ideally strove for
a total transformation of society, they were in reality quite different on a number
dimensions. In particular, they differed with respect to the state control over civil
society. Third, while it is obvious that violent revolutions and ensuing political and
social turmoil impair the capacities of any state, the long-term effects of radical
ruptures may be different. We may find cases where radical ruptures have helped
disarm traditional elites, thus providing a political opening for new forces. The
fourth tier is the actual policy and institutional choices that were made after reforms
were launched. In this context, the distinction between public policy and institutions
may be blurred, because institutional changes form a critical part of reform policy.
Rather than looking for the origin of institutions, the ambition of ‘empirical
institutionalism’ (Peters, 1999; Weaver and Rockman, 1993) is to examine the
political effects of various institutional arrangements. Research adhering to this
tradition focuses on the observable effects of institutions in different contexts or in
comparative perspectives. One line of research has focused especially on the political and economic consequences of presidential versus parliamentary systems in new
democracies. Neither general comparative studies nor studies focused specifically on
the subset of post-communist countries have, however, provided clear answers as to
which system of government produces the best outcome in terms of political and
economic development. In addressing these issues, Nørgaard (2000a, b) detects a
slight negative relation between economic reform and presidentialism, eventually
resulting in slower growth. Frye (1997) also finds a negative correlation between
presidentialism and economic reforms. Linz’s two celebrated articles (1990a; 1990b)
ague that presidentialism is more prone to resulting in breakdowns of democracy
than is parliamentarism. In his discussions of the post-communist cases, however,
Johannsen (2000) finds that only some forms of presidentialism, for instance
presidentialism combined with a highly fragmented party system, can be associated
with a slower or non-existent development of democracy compared to parliamentarism. Going one step further back in the causal chain, Johannsen (2000) finds that
the choice of government is strategic, reflecting the distribution of the actual and
expected resources of the political actors who seek to protect future influence in the
4
design of institutions. Frye (1999) sees the gradual expansion of executive power in
post-communist systems as a function of the inability of weak institutions to protect
the gains of transformation winners, thus providing these particular interests with
strong incentives to present their demands directly to politicians and state. This
argument is in line with Hellman, Jones and Kaufman (2000), who report that the
old elites already have influence through traditional networks, and the new elites
therefore have to resort to illicit methods to obtain power. For Easter (1997),
however, the choice is more specifically related to capture. In his analysis of the
choices made in the post-communist countries, Easter finds that presidentialism is
the preferred choice of surviving communist elites seeking to preserve their power
and to secure access to the state for traditional sectoral elites.
3. Virtuous and Vicious circles3
Low levels of trust in institutions engender low levels of support for the regime
(Rose, Mishler and Haerpher, 1997, p. 30). Democratic participation presupposes
that you trust that the institution of which you are a part will provide the benefits or
goods that pertain to its functional raison d'être. And if societal participation is a
precondition for good governance, trust in institutions becomes a precondition for
democratic participation. As an element in the virtuous and vicious circles of good
(or poor) governance, it is thus important to examine what fosters trust. Trust in
institutions is expressed in two different ways: as evaluations of the past and present
performance of institutions, and as trust in the future performance of institutions.
Starting from these definitions we explore two dimensions of popular trust in
institutions in post-communist countries: 1) satisfaction with the performance of the
institutional arrangements that have evolved in the process of transition, and 2) trust
in the political, administrative and traditional institutions. Finally, we examine state
capture as the third component of the virtuous and vicious circles introduced in
section 2.
3.1 Performance gaps
As measured by the Central and Eastern EuroBarometer4 a majority of East European populations believe that their country is heading in the wrong direction (Figure
2) and demonstrate dissatisfaction with development of democracy (Figure 3).
Figures 2 and 3 reveal the gap between popular expectations and the overall institutional performance following the systemic changes. Both figures show the general
trends in three regions and one outlier: the CIS countries (indicator), 5 the CEE
countries, and the Baltic States and Albania.
The performance evaluations in Figures 2 and 3 illustrate three points. First,
there is a high degree of correspondence between the curves in the two figures,
indicating that no major distinction is made between the general performance of the
system and the outcome of democratic institutions.6 However, the interpretation of
the ‘satisfaction-with–democratic-development’-question runs in two directions. At
5
the aggregated level Johannsen (2000; pp. 34-5; 205-6) and Nørgaard (2000a, pp.
202-5) find that rather than measuring democratic development in the liberal sense,
the satisfaction variable expresses an evaluation of the government’s ability to
deliver desired outcomes on a broadly interpreted welfare dimension. At the respondent level Berglund (1999) and Klingerman and Hoffbert (1998), in contrast, have
argued that individuals’ perception of the human rights situation is a better predictor
for satisfaction with democracy than individual assessments of economic prospects.
However, a closer examination of available data shows that it is a mixture of the
two, and that Berglund fails to consider the respondents’ retrospective economic
assessments. Regression analysis at the respondent level including this item
indicates that both evaluations of past and future economic prospects and evaluations of the government’s record in respecting human rights enter into the equation. 7
Figure 2. The overall direction of the country. (right =1, wrong = 2) 1990 – 1996.
Mean.
2
1,9
1,8
1,7
1,6
1,5
1,4
1,3
1,2
1,1
1
1990
1991
CEE (excl. Baltic)
1992
1993
CIS indicator
1994
1995
Baltic States
1996
Albania
Legend: CEE = Bulgaria, the Czech Republic, the Slovak Republic, Hungary, Poland, Romania, and
Slovenia. CIS indicator includes Armenia, Belarus, EuroRussia, and Ukraine. Baltic states include
Estonia, Latvia, and Lithuania. Interpolations for 1991 for Armenia, Belarus, Ukraine, the Slovak
Republic, and Slovenia. Interpolations for 1993 for Albania, Armenia, Belarus, The Czech Republic,
Estonia, EuroRussia, Hungary, Latvia, Lithuania, Romania, Ukraine, The Slovak Republic, and
Slovenia. Croatia, Georgia, Kazakhstan, and Moldova omitted from analysis because of too few or too
distant observations to make interpolation.
The question: ‘In general, do you feel things in (your country) are going in the right or in the wrong
direction?’ Responses were scored 1=right direction, 2=wrong direction, 3= don’t know. Respondents
answering’ don’t know’ were excluded from the analysis.
Source: Central and Eastern EuroBarometer 1990-1996 surveys. Danish Data Archive.
On the one hand, the data indicate the growing salience of economic welfare issues
and economic voting in most post-communist countries. The post-communist countries have thus already become ‘normal’ societies in the sense that citizens pursue
their own individual interests and judge institutions according to their performance
either on aggregated indicators or on the basis of individual utility (Nørgaard 2000a;
Fidimurc, 2000). On the other hand, Berglund’s (p. 20) conclusion, ‘there is no
6
evidence that human rights loses salience as time go by and freedom is taken for
granted’ is important as an indicator that present performance is also judged against
the past record of communist rule. This interpretation also helps to explain the
surprising position of Albania. Contemporary Albania may not be a pleasant place to
live and not even be a democracy, but in the minds of its people, it nevertheless
stands out as a better place to live in terms of human rights than it was before the
changes. In the specific context of the Albanian transition the shadow of the past still
has a relatively massive influence on people’s evaluation of performance. 8 For all
years, Albanian majorities express that the country is heading in the right direction,
the 1.2 mean value expressing that 85 percent of the population believe so. Furthermore, Albania constitutes two of the three cases throughout the period in which a
majority have expressed satisfaction with democratic development, respectively 61.6
and 76.2 percent in 1995 and 1996 doing so.9
Figure 3: Satisfaction with the development of Democracy. (Very = 1, fairly = 2, not
very = 3, not satisfied at all = 4). 1990-1996. Mean.
3,5
3
2,5
2
1,5
1990
1991
CEE (excl. Baltic)
1992
1993
CIS indicator
1994
1995
Baltic States
1996
Albania
Legend: See Figure 2.
The question: ‘On the whole, are you very satisfied, fairly satisfied, not very satisfied or not at all
satisfied with the way democracy is developing in (your country)?’ Responses were scored 1= very
satisfied, 2=fairly satisfied, 3=not very satisfied, 4=not at all satisfied, 5=don’t know. Respondents
answering ‘don’t know’ were excluded from analysis.
Source: Central and Eastern EuroBarometer 1990-1996 surveys. Danish Data Archive.
Second, the two figures also support the political idea that the Baltic States are in
fact a part of Central Europe. The similarity in the development of satisfaction
variables in the CEE countries and the Baltic States supports this argument. Only the
1991 data reveal a major difference between the Baltic States and the Central and
East European Countries following the restoration of Baltic independence. Third, the
stable difference between the CIS countries on the one hand and the CEE countries
and the Baltic States on the other suggests, irrespective of economic development
7
and political fortunes, a conflict between traditional and modern institutions (the
market and pluralist institutions). The institutional legacies (the ‘collective memory’
of the market and pluralist ideas) in the CEE and the Baltic States have rendered the
new institutions less alien in these countries than in the CIS, where especially the
communitarian traditions, ‘sobornost’ in Russian (Danilovich, 2001 forthcoming),
provide a major hindrance for liberal institutions. The stable level of satisfaction
variables also indicates that we are in fact dealing with structural variables.
3.2. Trust in political, administrative and traditional institutions
During communism, traditional institutions did enjoy greater legitimacy and popular
support than intermediate institutions in the sphere of politics and administration.
Surprisingly, this pattern has been replicated under post-communism, where public
trust in intermediate institutions remains modest. For example, Steen (1994) reports
that the Balts place more trust in the symbolic institutions than in the new democratic institutions. These findings are confirmed on a larger scale by the New Democracy Barometer IV and New Baltic Barometer III,10 as reported in Table 1.
When we traced mean trust in specific institutions for each country the
institutions were sub-divided into three groups: the political institutions of political
parties, parliament, government and president; the administrative and judicial institutions, i.e. courts, public officials and police, and finally, the church and the army
represent traditional national institutions.11
While there is variation between the countries in the level of trust placed in
specific institutions, there are nevertheless three overall patterns in Table 1. First,
mean level of trust falls below the natural midpoint, signifying a large degree of
mistrust in political and administrative institutions. Second, more trust is generally
placed in the traditional rather than in the political and administrative institutions.
Among the latter two, less trust is placed in the political institutions. Political parties
inspire the highest degree of mistrust. By comparison, trust in the president is
generally higher than for the other political instructions. In 8 of the 13 cases, trust in
the president is at or above the natural midpoint. This pattern probably reflects not
only a personified office, but also the differences among the presidencies in terms of
executive power and symbolic functions.
Compared with trust in the traditional institutions, the church and the army
are trusted at a rate above or at the natural midpoint in 9 and 10 cases, respectively.
The low degree of trust in political and administrative institutions implies that state
institutions are vertically insulated in both the political and the administrative
sphere. There are too few cases to provide a solid regional perspective, but Table 1
indicates the same divide which occurred concerning attitudes to macro institutional
development: Trust in political and administrative institutions seems somewhat
higher in the CEE than in the two CIS countries, whereas hardly any difference can
be detected with regard to traditional institutions. However, and we need more
country cases if we are to reach firm conclusions.
8
Table 1: Trust in Political, Administrative and Traditional Institutions.
Bulgaria Mean
Std. dev
Croatia Mean
Std. dev
Czech R. Mean
Std. dev
Hungary Mean
Std. dev
Poland Mean
Std. dev
Romania Mean
Std. dev
Slovakia Mean
Std. dev
Slovenia Mean
Std. dev
Estonia Mean
Std. dev
Latvia
Mean
Std. dev
Lithuania Mean
Std. dev
Belarus Mean
Std. dev
Ukraine Mean
Std. dev
Part.
34.5
28.8
37.9
26.2
42.8
22.6
27.4
24.2
27.2
23.4
37.0
29.4
34.9
24.6
32.7
25.1
28.2
22.9
22.4
25.9
29.7
25.7
22.8
23.5
22.4
24.9
Political inst.
Gov. Parl.
44.2 37.8
34.5 32.3
49.4 46.7
29.6 29.4
53.5 40.5
26.1 24.5
35.3 37.8
26.7 27.2
39.9 41.9
23.9 23.8
38.7 38.3
30.9 30.5
41.9 40.5
30.7 28.3
53.9 43.4
28.6 27.0
48.4 45.7
25.0 24.3
41.4 34.7
28.7 27.9
39.3 36.3
27.6 26.5
32.9 30.3
25.8 25.5
29.4 26.0
27.6 26.6
Pres.
46.9
34.3
50.0
33.4
50.0
23.6
50.4
26.0
43.0
28.1
55.0
31.8
42.4
25.5
63.0
29.9
57.8
27.6
52.0
29.3
50.5
30.0
46.1
34.9
42.3
32.7
Administrative inst.
Courts Police P. Off.
30.5 31.3 31.0
30.3 30.4 26.9
48.9 50.8 44.8
28.7 29.9 26.3
47.0 45.6 44.3
27.7 23.4 22.0
54.5 53.8 46.7
25.8 25.9 24.7
50.0 49.6 44.5
26.3 25.7 23.6
48.9 50.2 40.9
31.8 32.2 29.6
46.3 44.3 40.9
25.7 24.9 22.7
53.0 50.1 51.7
29.8 29.2 26.9
50.6 46.2 36.2
25.9 26.3 25.2
46.1 41.4 36.0
28.4 28.1 26.9
35.5 35.4 31.6
27.6 28.4 25.3
35.3 31.4 18.2
26.3 27.6 20.7
36.8 32.2 39.6
29.6 30.3 28.3
Trad. inst.
Church Army
46.1
64.6
35.5
31.6
51.4
70.1
32.5
29.6
41.1
47.6
29.7
25.0
48.6
53.5
30.4
25.2
55.8
60.6
30.4
24.9
78.2
76.5
28.3
26.9
49.0
57.6
31.9
25.9
38.0
56.2
32.5
30.0
58.8
50.0
27.9
26.1
62.4
42.1
31.0
29.7
64.8
41.7
29.8
29.1
56.7
55.6
32.7
30.5
60.6
55.2
32.6
31.7
Legend: High value equals high degree of trust. The question reads: ‘there are many different
institutions in this country, for example, government, courts, police, civil servants. Please show me on
this scale how great is your personal trust in each of these institutions that I read to you:’
Notes: The standard error of the mean less than 1.5 for all cases. In NDB IV and NBB III, 7 and 4point scales, respectively, were used. Both have been standardized into 100-point scales. For the
Baltic States ‘Cabinet of Ministers’ is used as proxy for government.
Source: NBD IV and NBB III.
Howard (2000) demonstrates that the average post-communist citizen has ’significantly and consistently lower levels of membership and participation than citizens of
most other democratic countries’ (p. 41), and that the differences between the postcommunist societies can be better characterized as ‘difference in degree’ rather than
‘differences in kind’. Furthermore, Misher and Rose (1998, p. 25) find that even if
social capital in the form of interpersonal trust ‘has substantial effect on individuals’
9
trust in political institutions, the cumulative effects of performance evaluations on
political trust are even greater’. These findings are consistent with the pattern of low
trust in political and administrative institutions reported in Table 1 and the hypothesized relationship between democratic participation and trust depicted in Figure 1.
In a historical perspective, the pattern of vertical insulation shown in Table 1
is the outcome of three historical legacies or institutional path dependencies. First,
they obviously reflect not only the repressive nature of communism, but also the
patterns institutionalized by the orthodox communist top-down style of government
where state institutions were seen as conveyor belts rather than forums for public
engagement and societal participation. Second, in some countries, and in particular
in the ’anti-political politics’ associated with Havel (1988), they reflect a pattern
institutionalized during the resistance to the incumbent communist systems, in that
meaningful political engagement was by definition being outside and against the
state institutions. Third, these patterns were in many countries reinforced by the neoliberal and retrenchment-of-the state policies of the early 1990s.
Societal participation unfortunately also has a dark side when participation
turns into penetration and capture of state institutions.
3.3 The capture of the state
Societal engagement and participation in political and administrative life are prerequisites for good governance and participation requires trust in the institutions. Trust
encourages participation, which in turn improves performance, which again increases trust. This is the spiral argument of the previous sections. The mirror image
of this virtuous circle is a situation where low levels of trust cause low participation,
which leads to a decline in performance, and eventually less trust in institutions, in
short, a vicious circle. The vicious circle can be initiated by the capture of state
institutions. ‘Capture’ is understood as groups, firms or institutions that come to
dominate political and administrative institutions, the state thus ceasing to represent
national concerns. Captured states endow particular groups, firms or institutions
with special advantages and benefits, thus bypassing normal democratic procedures.
The World Bank Business Environment and Enterprise Performance Survey
(BEEPS) is a useful indicator of state capture in post-communist countries. The survey uses a set of questions to measure the ‘extent to which firms make illicit or nontransparent private payments to public officials in order to influence the formation of
laws, rules, regulations or decrees by state institutions’ (Hellman, Jones and
Kaufman, 2000, pp. 5-6).12 The resulting Index of State Capture is illustrated in
Figure 4, and the bivariate correlations between state capture, performance
evaluations, and patterns of trust in political, administrative and traditional
institutions is reported in Table 2.
10
Figure 4: Degree of State capture in CEE and CIS. 1999.
UZB
AVG CIS
UKR
RUS
KGZ
MDA
KAZ
GEO
AZE
BLR
ARM
AVG CEE
SVN
SVK
ROM
POL
LVA
LTU
HUN
CZE
EST
HRV
BGR
ALB
0,45
0,40
0,35
0,30
0,25
0,20
0,15
0,10
0,05
0,00
Source: BEEPS surveys by The World Bank (1999). For discussions about design, see
Pradhan (2000), Hellman, Jones and Kaufmann (2000), Hellman, Jones, Kaufmann and
Schankerman (2000).
Figure 4 gives rise to three observations. First, it appears that there is no connection
between economic reforms and capture. Typical star reformers such as Estonia,
Hungary and Slovenia experience a low degree of capture, but so do Belarus and
Uzbekistan. It is more likely, however that this pattern illustrates the costs of an
incomplete transition. It seems possible to ward off new interests seeking to capture
the state by successfully implementing new institutions or by keeping the old
structures. This may seem paradoxical in, for example, the Latvian case because that
country has returned to pre-communist institutional structures. These structures
designed in the interwar period, however, and perhaps no longer ‘fit’ a country that
has seen massive demographic, economic and social changes during the communist
period. This case will be pursued in further detail in section 4. Second, the wars that
accompanied the break-up of Yugoslavia and the Soviet Union, resulting in war
economies and political systems, seem not to influence state capture. War torn
Croatia and Azerbaijan exhibit high degrees of state capture, while equally war torn
Armenia does not. Third, although no overall geographic pattern is detectable, the
CIS countries, on average, are captured to a greater extent than are the CEE
countries (incl. the Baltic States). This corresponds to the patterns observed in
Figures 2 and 3 above.
11
State
capture
N
Satisfaction
N
Direction
N
Adm. Inst.
N
Pol. Inst.
N
Army
N
Church
N
1.000
22
.250
18
.247
18
-.272
13
-.269
13
.264
13
.261
13
.250
18
1.000
19
.896 ***
19
-.607 **
13
-.522 *
13
-.188
13
-.275
13
.247
18
.896 ***
19
1.000
20
-.415
13
-.563 **
13
-.335
13
-.241
13
-.272
13
-.607 **
13
-.415
13
1.000
13
.570 **
13
.196
13
-.228
13
-.269
13
-.522 *
13
-.563 **
13
.570 **
13
1.000
13
.166
13
-.394
13
.264
13
-.188
13
-.335
13
.196
13
.166
13
1.000
13
.155
13
Church
Army
Pol. Institutions
Adm. institutions
Direction
Satisfaction
State Capture
Table 2. Bivariate Correlations (Pearson) between performance evaluations, trust in
institutions and state capture.
.261
13
-.275
13
-.241
13
-.228
13
-.394
13
.155
13
1.000
13
Legend: *** Correlation is significant at the 0.01 level (2-tailed), ** Correlation is significant at the
0.05 level (2-tailed), * Correlation is significant at the 0.10 level (2-tailed). Aggregated means used
for satisfaction and direction. The rescaled means from Table 1 used for Army and Church. For
Political and Administrative Institutions, aggregated and average means used for the component
institutions from Table 1.
Source: BEEPS surveys by The World Bank (1999), NDB IV, NBB III, Central and Eastern
EuroBarometer 1996.
But why are states captured? Is it a question of institutional design or a
reflection of broader patterns of historical legacy?
On the institutional aspect, we tested whether presidential systems are more
prone to capture than are parliamentary systems. To measure the strength of a
presidency we applied the degree of formal authority (Index of Presidential
Authority - IPA) and compared this index with the degree of state capture, as shown
in Figure 5. While there is a clear division into more presidential and more
parliamentary systems, there is no apparent effect of either presidentialism or
parliamentarism. Hence, Figure 5 illustrates that the argument that governance and
especially capture are related to the form of government is not valid. If the design of
a presidency contains a deliberate strategy to capture the state and not only to hedge
influence in the institutional design, then Figure 5 demonstrates that this strategy
does not always succeed.
12
Figure 5: State capture and the Index of Presidential Authority.
,5
Arz
,4
,3
Geo
Mol
Ukr
Rus
Bul Kyr
Lat
Cr
Slova
Rom
State Capture
,2
Al
Cz
,1
Lit
Est
Hun
0,0
10
Uz
Slove
20
Kaz
Pol
ArmBel
30
40
50
60
70
I n d e x o f P r e sid e n tial Au th o r ity
Source: Capture from World Bank BEEPS Survey (2000). For the coding and a detailed
description of the Index of Presidential Authority, see Johannsen (2000) or Nørgaard
(2000a). Index of Presidential Authority for Azerbaijan, Armenia, and Croatia from
Johannsen and Nørgaard (2001 forthcoming).
3.4 Virtuous and vicious circles revisited
Virtuous and circles produce good or poor governance in post-communist countries.
Performance breeds trust and participation, which again enhance performance. The
capture of state institutions by outside actors seeking advantages at the expense of
the community at large contributes to our understanding of why circles turn either
virtuous or vicious. Once it is in place, the vicious circle become difficult to escape
because state capture contributes to deteriorating trust and performance. Capture,
perceived performance and trust form a cluster of factors, which explains the
presence or absence of good governance. However, we have been unable to confirm
whether the specific institutional design in the form of presidentialism versus
parliamentarism makes states more prone to capture.
This is not to imply that institutional explanations are unimportant, but it does
indicate that the impact of institutions varies with the local context. It is impossible
to test this hypothesis for all countries in the post-communist world in the context of
the present paper, and we have therefore chosen to focus on two countries that, while
apparently sharing the communist history and legacy of forced industrialization,
have distinctly different pre-communist identities and post-communist developments.
4. History and capacity: The cases of Latvia and Lithuania 13
Cross-country comparative statistical analyses demonstrated the links between capture, trust and performance in the virtuous and vicious circles, producing good or
poor governance. We also showed that institutional factors (presidentialism) did not
affect the propensity of states to become captured by specific interests. However, if
13
we want to examine the strength of the second line of explanation introduced in
section 1, the broader historically formed institutions, we must move from the
general comparative level to the focused case study level.
Both Latvia and Lithuania countries can be described as intermediate cases
between Central Europe and the CIS countries, because they were closely integrated
into the Soviet orbit of government and control on the one hand, while on the other
they were independent states until the Second World War. This common historical
background should, it would seem, dispose for similar capacities to generate good or
poor governance, as they were also exposed to rather similar benign international
environments following the restoration of independence in 1991. This is, however,
not the case. As indicated by the World Bank Survey (see Figures 4 and 9) Latvia
has been more than twice as exposed to state capture as Lithuania.
This picture was confirmed in a recent survey of state capacities in Latvia and
Lithuania. The two studies used a subset of persons, who had been ministers in
selected (‘critical’) ministries since restoration of independence, as informants about
the workings of central state institutions and relations to civil society actors and
organizations.14 First, the survey showed that greater ministry officials in Lithuania
had a 20 percent greater propensity for collaborating with societal institutions (Fig.
6). Second, as shown in Figure 7, the survey showed a markedly higher frequency of
permanent patterns of collaboration with external institutions and actors. Finally, the
survey also showed that Lithuanian ministers had much greater confidence that the
active participation of civil and economic society actors would improve the
implementation of state policies (Fig. 8).
Figures 6, 7 and 8 show that collaboration between central state institutions
and civil society organizations is more commonplace in Lithuania than in Latvia, as
is the frequency of working contacts. This pattern is apparently carried by the much
greater confidence of Lithuanian top officials in the positive effect of state-society
interaction.
Figure 6: Working relationships of Ministerial officials.
100
90
80
70
60
50
40
Percent
30
20
Lithuania
10
0
La tvia
Common
Not Common
Working re lationship to orga niza tions c ommon or not?
14
Figure 7: Close external relationships
50
40
30
Percent
20
Lithuania
10
Latvia
0
Yes, most of the tim
No
Yes, on imp. issues
Note: Missing values have been omitted from this display
Figure 8: Private participation improves implementation.
60
50
40
30
Percent
20
Lithuania
10
0
Strongly agree
La tvia
Disa gre e
Agre e
Strongly dis agre e
No te: Mis s in g valu es h av e bee n o mit ted
These observations on the capacity of states with very similar starting points are
further specified if we return to the World Bank survey of state capture. This survey
allows us to differentiate between capture of different sectors of the political and
economic system, as illustrated in figure 9.
According to Figure 9, two countries that we would ex ante expect to have
very similar state capacities as measured in trust, state capture and performance
evaluations, turned out to perform very differently. How do we explain this
discrepancy? First, available data show that the logic of the virtuous/vicious circle
argument presented in section 2 holds for these two countries: while both states are
characterized by very low rates of trust, the state suffering most captured (Latvia)
has the least satisfied constituency, while the state with the highest approval ratings
(Lithuania) is characterized by the highest interaction and trust between societal
organizations and central state actors.15 Hence, the critical variable in the instigation
of virtuous and vicious circles, and eventually in good and poor governance,
15
becomes the predisposition for positive state-society interaction versus the
predisposition for state capture.
Figure 9: Institutional capture in Latvia and Lithuania compared.
0,50
0,45
0,40
0,35
0,30
0,25
0,20
0,15
0,10
0,05
0,00
Parliamentary
legislation
Presidential
decrees
Central Bank Criminal courts Political party
finance
Latvia
Commercial
courts
State Capture
Index
Lithuania
Source: BEEPS surveys by The World Bank (1999).
We now know that the explanation why countries end up in the first or second
alternative is not to be found in institutional design (at least at the macro level). We
believe that it is necessary to turn to long-term institutional trajectories instead
(Thelen and Steinmo (1992), the historical institutions celebrated by historical
institutionalism, if we want to explain when countries come to produce a positive
synergy between state and society and when state capture results. In section one we
distinguished between four tiers of explanations. The first three tiers reflect the
different institutions that were formed by history in the two countries. The last tier
deals with policy choices made after independence.
The first tier addresses the consequences of the different modes in which the
incumbent communist systems were extricated. Following independence in 1991,
Latvia by tacit consensus among the political parties representing ethnic Latvians
chose to reinstate the 1922 parliamentary system with a largely ceremonial presidency and a president appointed by the parliament. In comparison, following a protracted constitutional struggle Lithuania adopted a semi-presidential system in 1992
(Nørgaard and Johannsen, 1999, pp. 60-66). Where the long and bitter constitutional
struggle in Lithuania gave all contesting political actors a stake in the new institutions typified by the complex constitutional system, the Latvian constitution excluded about 30 percent of the population (largely Russian speaking immigrants)
from political influence. Today these groups may have to turn to illicit methods in
order to gain political influence.16 The degree of penetration of the political system
16
(political parties, president and parliament) reflects not only the way in which
excluded interests seek influence, as discussed above, but also that the political
parties in Latvia are weaker than their counterparts in Lithuania (Nørgaard and
Johannsen, 1999, pp. 94-96).17 This also serves to illustrate that the political and
civil society at the meso level is more developed in Lithuania than in Latvia.
Second, the communist epoch was very different in the two countries and
produced different patterns for state-society relations. Lithuania was, for a number
of reasons, able to maintain an ethos of national communism, where national
politicians were able to control the political agenda and defend national priorities. In
Latvia, in contrast, the state apparatus was controlled from the early 1950s onwards
by aliens, mostly expatriate Latvians from Russia. The consequence was that while
the state in Lithuania maintained legitimacy during most of the communist epoch,
this was not the case in Latvia, where it became a symbol of foreign dominance, and
hence endowed with little legitimacy and trust. The communist epoch with its forced
industrialization has also generated differences between the countries. Although
already industrialized in the pre-war period, further industrialization during communism left the Latvian economy highly concentrated in certain sectors, in particular
the country’s position on east-west trade and energy transit routes. This gave rise to
what may be termed a rentier state, providing huge opportunities for extracting rents
and therefore an obvious target for capture (Pradhan, 2000, p. 70; Nørgaard and
Hersted Hansen, 2000). Also the third institutional tier related to pre-communist
symbols reinforces the same patterns of state-society relations in the two countries.
Lithuania has always had a strong national self-consciousness, reaching back to her
Middle Age status as a major European power. Even today, this distant history
bestows the state and its institutions with an aura of past grandeur that spills over
into popular trust, which may protect the state against being exploited by special
groups. Latvia, in contrast, has always struggled with a week national identity
caused by quite fundamental regional differences in culture, religion and economic
development. Large ethnic minorities (Germans and Russians) with major economic
and political positions have also made it difficult to cultivate a genuine sense of a
Latvian identity, with ensuing negative attitudes towards a state, which was never
perceived as a national institution symbolizing shared values. Finally, also the policy
and institutional choices made after independence in the two countries have affected
the level of state capture and state capacities. Especially the case of the (non
captured) Central Bank in Latvia is informative, because it shows that policy choices
and institutional arrangements can prevail over a predisposition towards capture
penetrating other state institutions (Nørgaard and Johannsen, 1999, p. 129-131).
This short comparative case study illustrates that although we were unable to
identify any direct effect of macro (institutions) on governance, institutions may
nonetheless have an effect. The consequences of institutional choices vary between
contexts, however, here conceptualized as the three tiers of institutions formed by
the history of each country. The case of the central banks in the two countries also
17
underscores this point. Special interests have apparently captured most state
institutions in Latvia, but evidently not the Central Bank. Although history conditions, politics decide. Appropriate political and institutional choices can change the
course and predispositions embedded in the historically formed institutions.
5. Conclusion: institutions and governance in post-communist countries
The ambition of this paper was to improve our understanding of what produces good
and poor governance in post-communist countries. Starting form a definition of
good governance as a style of government which integrates social and political
demands while upholding national goals and priorities, it was hypothesized that
countries could enter virtuous or vicious circles of good or poor governance. In the
virtuous version, a government is able to inspire popular trust in state institutions,
which is a precondition for active societal engagement and the participation that
provides the government with the information and resources necessary to meet the
expectations of the population. In the vicious version, particular interests captured
state institutions. This results in decreasing popular trust in institutions, and
eventually declining performance and bad governance. To understand what caused
good and poor governance we applied the insights gained from historical and
empirical institutionalism. When exposed to comparative statistical analyses the data
provide evidence that two patterns link institutions to governance. Perceived performance and trust in institutions are positively correlated and both negatively
associated with state capture. Further, if state capture is seen as the trigger of the
virtuous and vicious circles of governance, the comparative case study showed that
long-term historical trajectories (historically determined institutions) provide sound
explanations of why the two countries performed so differently, regardless of their
apparently similar starting positions. However, the case study also demonstrated that
one institution, the Central Bank of Latvia, diverges from the overall picture and has
escaped the capture that characterizes most other state institutions in Latvia. This
case illustrate that even in countries conducive to capture, it is possible to design
new institutions that avoid the disposition of the historically determined context.
This observation suggests that we must combine the perspectives of empirical
and historical institutionalism if we are to understand the performance of particular
institutional configurations. In such a combined perspective, the performance of
particular institutional arrangements should be assessed in conjunction with each
country’s specific institutional endowments. New institutions can only enjoy trust
and legitimacy if they are coincide with the informal institutions of society,
eventually boosting participation and improving good governance. While this observation remains tentative and has to be developed in further empirical research, it
does correspond with the observation made long ago in the structuralist tradition in
development economics. Recent insights from Central Europe lend further credit to
this argument.18 Future research should therefore concentrate on how to conceptualize the channels that link the past with present institutional endowments and on
18
further elaborating and defining the parameters involved in the fusion of new and
old institutions in specific country contexts.
Notes
1. Using Diamond’s (1999, p. 221) definition of civil society.
2. This definition is developed from the discussion in Pradhan (2000), p. 3.
3. Thanks are due to Tom Y.K. Nielsen for assistance with the compilation, coding and
comparison of the data presented in this section.
4. Weights were used wherever possible to make the sample representative for each
individual country.
5. We prefer the term indicator because only Armenia, Belarus, EuroRussia and Ukraine
are represented in the mean.
6. Using the 1996 survey, a significant mean difference in satisfaction with development of
democracy can be demonstrated between those who perceive that their country is going in
the right direction and those who perceive it to be going in the wrong direction. Satisfaction
with development of democracy: Mean Right direction = 2.47, Mean wrong direction = 3.34.
Mean difference = 0.87. Significance = 0.000. N = 17.980. See also Table 2 for the correlations on aggregated level.
7. Without repeating Berglund’s recoding of the data but instead using the original finer
scale, regressions analysis for all countries in the 1996 Central and Eastern EuroBarometer
shows an overall adjusted R2 of 28.8 percent with standardized beta values for (Q7) human
rights, (Q3) retrospective economic assessment, (Q4) prospective economic assessment at
0.33, 0,21 and 0.17 respectively.
8. Analysis for Albania in the 1996 Central and Eastern EuroBarometer shows an overall
adjusted R2 of 49.5 percent with standardized beta values for (Q7) human rights, (Q3)
retrospective economic assessment, (Q4) prospective economic assessment at 0.42, 0,33
and 0.10 respectively.
9. Romania constitutes the third example. In 1996, 54.6 percent expressed that they were
very or fairly satisfied with developments.
10. NDB and NBB surveys are provided as part of the CITNET program. For more information about the surveys, see http://www.cspp.strath.ac.uk/
11. Repeated factor analysis demonstrates that three types of institution can be distinguished. In NDB IV political parties, government and Parliament all load on the first
component. Courts, police and public officials load on the second, while the church and the
army as the traditional institutions load on the third component. Presumably because the
President carries symbolic as well as executive functions, the question of trust in the
President loads equally (poorly) on both the political institution component and the
symbolic component. In NBB III, the President, Cabinet of Ministers (here used as proxy
for government) and parliament load on the first component. Police, Courts, Officials and
the local army load on the second, and the church on the third. In NBB III, the question of
trust in political parties loads poorly on all components.
12. The battery involves 1) the sale of parliamentary votes on laws to private interests; 2)
the sale of presidential decrees to private interests; 3) Central Bank mishandling of funds;
19
4) the sale of court decisions in criminal cases; 5) the sale of court decisions in commercial
cases; 6) illicit contributions by private interests to political parties and election campaigns.
13. The analysis and discussion in this section are based on Nørgaard and Johannsen (1999).
14. The Latvian survey (face-to-face interviews) was conducted in January 2000 and
complemented with about 10 in depth interviews with centrally placed civil servants. The
Lithuanian survey was conducted by Ole Hersted Hansen in November 2000, and is now
being supplemented by in-depth interviews. The interviews included 51 (Latvia) and 53
(Lithuania) former ministers with the following portfolios: A) Ministries with general
(domestic) responsibilities – Finance and Economics. B) Those with more specific (but
central) domestic tasks – Welfare and the Interior (there are, of course, other ministries in
this category). C) Foreign Affairs. For a more detailed presentation of the two studies, see
the research reports on www.demstar.dk (Ole Hersted Hansen, 2001) is forthcoming.
15. See table 1 on trust variables. For satisfaction variables, see Nørgaard and Johannsen,
1999, p. 99.
16. This pattern corresponds to the distinction between capture and influence as discussed
by Hellman, Jones and Kaufmann (2000, pp. 5-6, 12).
17. For a comparison of the organizational strength of Latvian and Lithuanian political
parties, see Smith-Sivertsen (1996, 1997).
18. Ágh, (2001 forthcoming).
20
References
Ágh, Attila (1997) ‘Actors of Systemic Change: The Political Context of Public Sector
Reform in Central Europe’, Discussion Paper No. 19, European Centre for Comparative Government and Public Policy.
Ágh, Attila (2001 forthcoming), ‘Understanding Politics in Hungary’, DEMSTAR: Understanding Politics Series 1, Department of Political Science, University of Aarhus.
Berglund, Sten (1999), The Challenge of Human Rights in Central and Eastern Europe,
unpublished paper, Department of Social Science, University of Örebro.
Business Environment and Enterprise Performance Survey – BEEPS (1999), World Bank
Institute: http://www.worldbank.org/wbi/governance/datasets.htm#beeps
Central and Eastern EuroBarometer, 1990-1996.
Danilovich, Alexander, ‘Understanding Politics in Belarus’, DEMSTAR Understanding
Politics Series, 2, 2001 (forthcoming), www.demstar.dk
Easter, Gerald M. (1997), ‘Preference for Presidentialism: Postcommunist Regime change
in Russia and the NIS’, Word Politics, no. 49.
Evans, Peter (1995), Embedded Autonomy, Princeton N.J.: Princeton University Press.
Evans, Peter R., Dietrich R. Ruesmeyer, and Theda Skocpol (eds.) (1979), Bringing the
State back In. Cambridge, Cambridge University Press.
Fidmurc, Jan, (2000), ‘Political Support for reforms: Economic Voting in transition
countries’, European Economic Review, no. 44, pp. 1491-1513.
Frye, Timothy (1997), ‘A Politics of Institutional Choice: Post-Communist Presidencies’,
Comparative Political Studies, Vol. 30, No. 5, pp. 523-552.
Frye, Timothy (1999), ‘Changes in Post-Communist Presidential Power: A Political
Economy Explanation’, paper presented at the conference: Constitutional Design 2000,
Center for Continuing Education, University of Notre Dame, December 9-10, 1999.
Galligan, Denis and Daniel Smilov (eds) (1999), Administrative Law in Central and
Eastern Europe, 1996-1998, Budapest: Central European University.
Havel, Václav (1988), ‘Anti-Political Politics’, pp.381-398 in Keane (ed.), Civil Society
and the State: New European Perspectives, London: Verso.
Hellman, Joel S., Geraint Jones and Daniel Kaufmann (2000),’Seize the State, Seize the
Day: State Capture, Corruption and Influence in Transition’, Policy Research Working
Paper 2444, Washington D.C.: The World Bank Institute.
Hellman, Joel S., Geraint Jones, Daniel Kaufmann and Mark Schankerman (2000),
’Measuring Governance, Corruption, and State Capture: How Firms and Bureaucrats
Shape the Business Environment in Transition Economies’, Policy Research Working
Paper 2312, Washington D.C.: The World Bank Institute.
Horváth, Tamás (ed.) (2000) Decentralization: Experiments and Reforms – Local Governments in Central and Eastern Europe, Budapest: Local Government and Public Service
Reform Initiative (Login).
Howard, Marc Morjé (2000), ‘Free not to participate: The Weakness of Civil Society in
Post-Communist Europe’, Studies in Public Policy 325, Centre for the Study of Public
Policy, University of Strathclyde.
Johannsen, Lars (2000), The Constitution and Democracy: The Choice and Consequence of
the Constitution in Post-Communist Countries, Aarhus: Politica
21
Johannsen, Lars and Ole Nørgaard (2001 forthcoming), ‘The Index of Presidentialism: A
Database of Presidential Authority in the 2nd and 3rd world’, DEMSTAR Research
Paper, Department of Political Science, University of Aarhus.
Kjær, Anne Mette, Jens Peter Frølund Thomsen and Ole Hersted Hansen (2000), ’Conceptualising State Capacity’, DEMSTAR draft paper, Department of Political Science,
University of Aarhus
Kilngerman, Hans-Dieter and Richard I. Hoffbert: ‘Remembering the Bad Old Days:
Human Rights, Economic Conditions and Democratic Performance in Transitional
Regimes’, Discussion paper FS III 98-102, Wisshenschaftszentrum für Socialforschung (WZB), 1998.
Krasner, Stephen D. (1984), ‘Approaches to the State: Alternative conceptions and
historical dynamics’, Comparative Politics Vol. 16, pp. 223-46.
Leftwich, Adrian (1995) ‘Bringing Politics Back in: Towards a Model of the Developmental State’, The Journal of Developmental Studies, Vol. 31, no. 3, pp. 400-427.
Linz, Juan J. (1990a),‘The Perils of Presidentialism’, Journal of Democracy (Winter 1990),
pp. 51-69.
Linz, Juan J. (1990b) ‘The Virtues of Parliamentarism’, Journal of Democracy (Fall 1990),
pp. 84-91.
Mishler, William and Richard Rose (1998), ‘Trust in Untrustworthy Institutions: Culture
and Institutional Performance in Post-Communist Societies’, Studies in Public Policy
310, Centre for the Study of Public Policy, University of Strathclyde.
New Baltic Barometer, III.
New Democracies Barometer, IV.
Nørgaard, Ole (2000a), Economic Institutions and Democratic Reform, Cheltenham &
Northampton, MA: Edward Elgar.
Nørgaard, Ole (2000b), ‘Politische Institutionen und wirtschaftlicher Wandel in jungen
Demokratien am Beispiel ehemals kommunistischer Länder’, pp. 196-223 in van Deth,
Jan Koenig, Thomas (Hrsg.), Europaeische Politikwissenschaft - ein Blick in die
Werkstatt, Mannheimer Jahrbuch für Europaeische Sozialforschung, Band 4, Campus:
Frankfurt/M.
Nørgaard, Ole and Lars Johannsen with Mette Skak and Rene H. Sørensen (1999), The
Baltic States after Independence, Cheltenham: Edward Elgar.
Nørgaard, Ole and Ole Hersted Hansen with Ilze Ostrovska and Luise Pape Møller (2000),
‘State of the State in Latvia’, DEMSTAR Research Report 1, Department of Political
Scince, University of Aarhus: http://www.demstar.dk/
Nørgaard, Ole (forthcoming), ‘Democracy, Democratization and Institutional Theory’,
DEMSTAR Research Paper, http://www.demstar.dk/
Nunberg, Barbara (1999) The State After Communism: Administrative Transitions in
Central and Eastern Europe, Washington D.C.: The World Bank.
Nunberg, Barbara (2000) Ready for Europe: Public Administration Reform and European
Union Accession in Central and Eastern Europe, Washington D. C.: The World Bank.
Hansen, Ole Hersted (2001, forthcoming), ‘State of the State in Lithuania’, DEMSTAR
Research Report, www.demstar.dk
Peters, B. Guy (1999), Institutional Theory In Social Science: The ‘New Institutionalism’,
London: Pinter.
22
Pierre, Jon and B. Guy Peters (2000), Governance, Politics and the State, New York: St.
Martin’s Press.
Pradhan, Sanjay (2000), Anticorruption in Transition: A Contribution to the Policy Debate,
Washington, D.C.: The World Bank.
Putnam, Robert D. (1993), Making Democracy Work: Civic Traditions in Modern Italy,
Princeton, NJ: Princeton University Press.
Rose, Richard, William Mishler and Christian Haerpfer (1997), ‘Getting Real: Social
Capital in Post-Communist Societies’, Studies in Public Policy 278, Centre for the
Study of Public Policy, University of Strathclyde.
Schleifer, Andrei and Robert W. Vishny (1999), The Grabbing Hand. Government
Pathologies and their Cures, Cambridge, Mass.: Harvard University Press.
Andrei Shleifer, Andrei and Daniel Treisman (2000), Without a Map, The MIT Press.
Skocpol, Theda (1979), ‘Bringing the State back in: strategies of analyses in current
research’, pp. 3-37, in Evans, Ruesch and Skocpol (eds.) Bringing the State back In,
Cambridge, Cambridge University Press.
Smith-Sivertsen, Hermann (1996) ‘Mot elitepartier eller massepartier?’, Nordisk Østforum,
Vol. 10 (3), pp. 5-22.
Smith-Sivertsen, Hermann (1997), ’Is the smallness of the party organisations of contemporary Latvia and Lithuania a danger for the democratic consolidation in these
countries?’, unpublished paper for the Oslo Summer School in Comparative Social
Science Studies, 21 p.
Solnick, Steven L. (1998), Stealing the state. Control and Collapse in Soviet Institutions,
Cambridge, Mass.: Harvard University Press.
Steen, Anton (1994), ‘Confidence in Institutions in Post-Communist Countries: The Case
of the Baltic States’, Working Paper 12, Oslo: Department of Political Science,
University of Oslo.
Thelen, K. and S. Steinmo (1992), ‘Historical Institutionalism in Comparative Politics’, pp.
1-32 in K. Steinmo, K. Thelen and F. Longstreth (eds.), Structuring Politics: Historical
Institutionalism in Comparative Analyses, Cambridge: Cambridge University Press.
UNDP Regional Report (1997) The Shrinking State: Governance and Human Development
in Eastern Europe and the Commonwealth of Independent States, New York: UNDP.
Weaver, R. Kent and Bert A. Rockman (1993), ‘Assessing the Effects of Institutions’, pp.
1-41 in Do Institutions Matter?, The Brooking Institutions, Washington D.C.
Weiss, Linda (1998), The Myth of the Powerless State. Governing the Economy in a global
Era, Oxford: Polity Press.
World Development Report (1997) The State in a Changing World, Oxford: Oxford
University Press, published for The World Bank
Young, Iris Marion (1996), ‘Political Theory: An Overview’, pp. 479-502 in Goodin and
Klingeman (eds.), A New Handbook of Political Science, Oxford and New York:
Oxford University Press.
23