E c o n o m i c &
S o c i a l A f f a i r s
DESA Working Paper No. 55 ST/ESA/2007/DWP/55
August 2007
Corruption and Democracy
Michael T. Rock
Michael T. Rock is Harvey Wexler Professor of Economics in the Department of Economics, Bryn Mawr College, 101 N. Merion Ave, Bryn Mawr, PA, USA 19010. E-mail: mrock@brynmawr.edu. Comments should be addresd by e-mail to the author.Abstract
What is the impact of democracy on corruption? In most models, analysts assume a negative rela-tionship, with more democracy leading to less corruption. But recent theoretical developments and ca evidence support an inverted U relationship between corruption and democracy. By drawing on a panel data t covering a large number of countries between 1996 and 2003, substantial empiri-cal su
pport is found for an inverted U relationship between democracy and corruption. The turning point in corruption occurs rather early in the life of new democracies and at rather low per capita incomes.
JEL Classification: O12, D72, D73, H11, H77, K42
英国签证费用Keywords: corruption, electoral democracy, consolidated democracy, rule of law, government effectiveness
UN/DESA Working Papers are preliminary documents circulated in a limited number of copies and posted on the DESA website at /esa/desa/papers to stimulate discussion and critical comment. The views and opinions expresd herein are tho of the
author and do not necessarily reflect tho of the United Nations Secretariat. The designations and terminology employed may not conform to United Nations practice and do not imply the expression of any opinion whatsoever on the part of the Organization.
Typetter: Leah McDavid United Nations
Department of Economic and Social Affairs
2 United Nations Plaza, Room DC2-1428 New York, N.Y. 10017, USA
Tel: (1-212) 963-4761 • Fax: (1-212) 963-4444 e-mail:
/esa/desa/papers
Contents
Corruption and Democracy: What We Know (2)
Data and Hypothesis Tests (4)
脚手架搭设Data (4)
Hypothesis tests (10)
Conclusions (16)
References (17)
Corruption and Democracy
Michael T. Rock
What is the impact of democracy or its polar opposite, autocracy, on corruption? Existing econometric evi-dence is mixed. Ades and Di Tella (1999: 987) and Fisman and Gatti (2002: 336-338) fail to find any posi-tive association between the political and/or civil rights associated with democracy and corruption. In fact, as Ades and Di Tella (1998: 987) state, “If anything, the lack of political rights ems to be associated with less corruption”. On the other hand, Goel and Nelson (2005: 127 and 130) find that corruption declines with the degree of civil liberties associated with democracy, Chowdhury (2004: 96, 98) finds that corruption declines with Vanhanen’s (1992) democracy index, while T riesman (2000: 417) finds that the duration of de-mocracy, defined as the number of uninterrupted years in which a country is democratic, reduces corruption.
The results hardly inspire confidence. They are also inconsistent with a growing body of ca evidence which suggests that corruption ris, at least initially, in newly democratizing countries, before falling as democracies become consolidated. Mohatdi and Roe (2003: 445), among others, comment on this phenomenon in Russia, T urkey and Latin America. Knowledgeable obrvers in Indonesia (McLeod 2005, Robison and Hadiz 2004, Rock 2003) and Thailand (Ca 2002, Hicken 2001, Rock 2000, Pasuk and Baker 1998, Ammar 1997) agree that corruption ro in both countries following democratization as the collap of centralized networks of corruption gave way to more cor
rosive and decentralized corruption free for alls. What accounts for this apparent ri in corruption following democratization and is there broader evidence to suggest that it subquently falls with the consolidation of democracy, yielding an inverted U pattern?
Mohatdi and Roe (2003) provide theoretical support for why corruption might follow an inverted U relationship with democracy. Ca evidence prented below for Indonesia and Thailand, which is broadly consistent with their theoretical argument, suggests why this might be so, at least, in the economies.
This still leaves open the question, is there robust empirical support for an inverted U relationship between democracy and corruption as depicted in figure 1, which plots the relationship between corruption and what Schneider and Schmitter (2004: 84) label the consolidation of democracy?1 If so, which aspects of democracy matter: electoral competition,2 rule of law,3 effective governance4 or the behaviour, attitudes and
1 Schneider and Schmitter (2004: 67-68, 84) develop an empirical index of the consolidation of democracy for
25 countries between 1974 and 2000. This index is bad on the ratings of country experts for 11 ite
ms. Their
measure has an electoral bias, but it also includes a number of elements that go beyond free, fair, regular, and
contested elections such as agreement on rules governing association formation and behaviour, territorial division
of competencies, and rule of ownership and access to the media (Schneider and (Schmitter 2004: 68). Figure
1 plots the relationship between this measure of consolidated democracy (CoD) and a World Bank measure of
corruption taken from (Kaufmann, et al. 2007). The World Bank measure has been rescaled so that an increa
in the variable implies more corruption. The predicted value for corruption in figure 1 is bad on the following
simple OLS regression equation. CORRWB = -9.73 + 6.76 Log (CoD) – 1.07 Log (CoD)2 . Both regression
coefficients are statistically significant at the .01 level (t = 3.16 and 3.63) as is the equation F statistic (F = 16.48).
Adjusted R2 is .56.
2 Schneider and Schmitter label this Dahl’s procedural minimum (2004: 63), while Karl (1986: 9-36) labels it an
‘electoralist fallacy’.
3 Linz and Stephan (1996: 10) identify the rule of law as one of five interacting arenas that must be in place for
democracy to be consolidated. Diamond (1999: 111-112) also includes it as an important element in the consolidation
of democracy.
4 Both Linz and Stephan (1996: 10-11) and Diamond (1999: 93-96) stress the importance of effective government. As
Linz and Stephan (1996: 11) state, “Modern democracy…needs the effective capacity (of government) to command,
regulate, and extract. For this it needs a functioning state and a state bureaucracy considered uable by the new
王熙凤性格特点
democratic government.
2D E S A W o r k i n g P a p e r N o.55
norms of political actors?5 The questions are answered by using a panel data t for 1996-2003 to dem-onstrate that after controlling for the other factors affecting corruption, corruption follows an inverted U relationship with more process oriented definitions of democracy, but not with the electoral aspect of democ-racy. The argument proceeds in three steps. The next ction asmbles the theoretical and ca evidence in support of the inverted U hypothesis. The following ction tests for this empirical relationship in a panel of data that controls both for other variables affecting corruption and for endogeneity. The final ction clos by drawing implications.
Corruption and Democracy: What We Know
Both theory and ca evidence provide compelling support for a democratization breeds corruption hypoth-esis, at least up to a point. At the theoretical level, Mohtadi and Roe (2003) model corruption as the mo-nopolistically competitive behaviour of private ctor agents who can either invest in productive activity or in rent-eking (corruption). In their model, young democracies suffering from insufficient checks and bal-ances and lack of transparency, provide rent-ekers with greater access to public officials and hence greater opportunities for collecting public ctor rents, at least up to a point, without making the corrupt acts of rent-ekers and officials open to public scrutiny. Becau of free entry into rent-eking, competition among rent-ekers ultimately reduces returns to individual rent ekers even as it drives aggregate rents up. But as
六后5 Schneider and Schmitter (2004: 68) emphasize the behaviours of political actors; Linz and Stephan (1996: 6)
emphasize behaviours and attitudes, while Diamond (1999: 69) emphasizes the behaviours, norms and beliefs of
political actors. All three emphasize the importance of the time it takes for actors to learn democracy and become
做包子的全过程
habituated to it.
Corruption and Democracy 3 the institutions of transparency and accountability in new democracies ri as they mature, aggregate rents
and corrupt activity fall becau rents per rent eker fall and becau the cost of rent-eking (including the
probability of getting caught and punished) to rent-ekers and the government officials, who accept bribes,
ris. Taken together, this combination implies an inverted U pattern between corruption and the durability
or maturity of new democracies.
The ca evidence, at least from Indonesia and Thailand, is broadly consistent with this picture. In
both polities, corrupt networks were more or less tightly controlled by political elites in government, the bu-
reaucracy and the army (Rock 2003, Rock 2000, Rock and Bonnett, 2004). As Rock (1994, 2000) and Rock
and Bonnett (2004) argue with respect to Thailand’s bureaucratic polity, democratization led to the break-up
of a centralized corruption network between political elites, nior bureaucrats and nior army officials on
the one hand and the Sino-Thai entrepreneurs who drove the growth process following the growth coalition
asmbled by General Sarit in 1960. In this centralized network, government officials, including army of-
ficers, provided protectionist rents to a surprisingly small number of Sino-Thai entrepreneurs in exchange for
kickbacks. As in Indonesia, the government protected private property and extracted rents at a low enough
‘tax’ rate to entice entrepreneurs to invest, which they did.
A combination of rapid growth and democratization ultimately led to at least a mi-democratic
polity by the early 1980s (Chai-Anan 1990). For a while, during the time Prem was the prime minister
(1980-88), Thailand’s bureaucratic polity evolved toward both a broker polity (Ramsay 1985) and a North-拉网捕鱼
east Asian style developmental state (Anek 1988) as core economic agencies, peak business associations, and
key business leaders regularly met in a high level Joint Public Private Sector Consultative Committee to work
our problems associated with Thailand’s policy shift which favoured the export of manufactures.
But this transformation did not last as unscrupulous up-country provincial politicians subquently
captured both the legislature and the prime minister’s office (Girling 1997; Callahan and McCargo 1996;
King 1996). They ud their control of both to carry out a frontal and corrupt assault on the state to re-
ward their supporters and build their coffers for the next election (King 1996: 136-137).6 They did so, by among other things, politicizing the core institutions of macroeconomic policy—the Ministry of Finance,
the Central Bank, and the national planning agency, the National Economic and Social Development Board
(Rock 2000: 197-198; Murray 1996). This led at least one long time analyst to ask whether new democracies
could manage their macro-economies (Ammar 1997). The ri of shadowy provincial businessmen in politics
and their corrupt frontal assault on the state ultimately led bureaucratic and political elites in Bangkok to try
and slow the spread of corruption by enacting a new constitution in the late 1990s designed to reign i
risksn the
corruption associated with money politics and rural vote buying (Callahan 2005).7 Although it is difficult
to know whether the new constitution reduced corruption in Thailand, the new constitution re-centralized
1921电影politics by significantly reducing the number of political parties (Hicken 2006). One outcome of this process
was the ri of another provisional businessman Thaksin Shinawata, who became prime minister in a govern-
ment that for the first time in Thai history captured a majority in parliament for his Thak Rai Thai Party
(McCargo and Ukrist 2005).
6 McCargo and Ukrist (2005: 73-74) describe this process as related to the factional basis of Thai politics in which
political parties and faction leaders within them are allocated cabinet positions and jobs on the basis of their electoral
strength. Tho holding cabinet posts are obliged to reward the faction leaders in their parties, usually through corrupt
means.
7 They did so becau corruption had become endemic as at least one Thai cabinet came to be described as little more
than a ‘Mafia Cabinet” (Murray 1996: 372).
