In the early 1990s, after the collapse of the Berlin Wall and the break-up of the Soviet Union, many economists argued that the post-communist countries could rapidly develop a capitalist and market system. (1) Some held the market order would rapidly germinate and grow once the old state bureaucracies were swept away. As the influential Western advisor Jeffrey Sachs (1993, xxi) contended: "markets spring up as soon as central planning bureaucrats vacate the field."
In fact, markets did not spring up spontaneously; capital markets, in particular, were slow to develop. As Ronald Coase (1992, 718) observed: "The ex-communist countries are advised to move to a market economy ... but without the appropriate institutions, no market of any significance is possible." From an institutionalist perspective, markets are not the universal physical ether of economic existence, but social institutions that depend upon the evolution of detailed rules and norms (Hodgson 1988; 2001; Vanberg 2001). In the former communist countries, the requisite commercial rules, norms and institutions were scarce (Clague 1997; Kozul-Wright and Rayment 1997; Grabher and Stark 1997).
More than a decade and a half after the fall of the Berlin Wall, there is now plentiful evidence to consider in explaining the differences in performance in the transitional or post-communist countries of the former Soviet Bloc. All these countries experienced severe or extreme recessions in the 1990s, followed by a subsequent recovery. Jan Svejnar's (2002) survey showed that only Poland and Slovenia had significantly exceeded their 1989 GDP by the year 2000. Hungary, Slovakia and the Czech Republic just about matched their 1989 GDP after 11 years. Other former Eastern Bloc countries were still well below their 1989 GDP levels as late as 2001, with Russia about 40 percent lower and the Ukraine about 60 percent lower. Nauro Campos and Fabrizio Coricelli (2002) reported similar results. (2)
This article updates these earlier studies and attempts to identify the principal underlying factors that account for overall post-1989 economic growth in the transitional economies. Priority is given to institutional factors rather than standard macroeconomic variables. The sections below respectively present the relevant growth data, consider a number of hypotheses that might account for these growth outcomes, report the econometric results, discuss their significance, and offer some deeper explanations. The concluding section highlights the importance of effective national institutions, enforcing non-discriminatory rules and overcoming the negative economic legacy of ethnic and other divisions.
The statistical analysis here focuses on the growth in GDP and on absolute levels of GDP per capita. Of course, it should not be assumed that GDP is an accurate measure of welfare or well-being. Official GDP figures may also be inaccurate, particularly in assessing the size of undeclared revenues and the illegal economy.
The data cover 27 former communist countries, including nineteen in Europe and eight in Central Asia. (3) Seven of these countries--Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia, and Slovenia--joined the European Union (EU) in 2004. Some key indicators are shown in Table 1, which shows a very mixed picture. Fourteen of these former communist countries have experienced negative overall growth in the sixteen-year period. On the other hand, nine have achieved an average 1989-2005 GDP growth rate of over one percent, and in Poland and Albania growth has exceeded two percent. Taking the 1989-2005 period as a whole, only nine countries can claim a modestly successful overall growth rate. The remaining countries have experienced stagnation or decline.
Moreover, two of the more rapidly growing countries--Albania and Turkmenistan--are extremely poor compared with the more developed European countries. Their levels of GDP per capita are instead comparable to developing economies such as Algeria, Guatemala, Namibia and Thailand. The five remaining countries--Poland, Slovenia, Slovakia, Hungary and the Czech Republic--that have exceeded one percent overall growth in the 1989-2005 period have GDP per capita levels well below the 2001 figures of $34,946 per capita in the United States, and $20,860 per capita in the European Union.
The 1990s recessions in transitional economies varied considerably in severity, as shown in Table 1. The least severe was in the Czech Republic, where the downturn of 1992 brought the economy to 87 percent of its 1989 level. In Russia the 1996 downturn brought the economy to 59 percent of its 1989 level. The extremely severe recessions in Armenia, Azerbaijan, Bosnia and Herzegovina, Georgia, Kyrgyzstan, Moldova, Serbia and Montenegro, Tajikistan, and the Ukraine meant the loss of at least half of their GDP levels in 1989. In some of these countries, the recession was at least partly the result of civil war and the breakdown of internal order.
Several additional datasets exist to test a number of hypothetical explanations of the varied, post-1989 performances of the transitional economies. (4) For convenience we group these different hypotheses together according to the different types of independent variable they identify.
(a) Property rights and corruption. The first group of hypotheses addresses the conditions under which legitimate contracts and exchanges may be secured in a market economy. These explanations emphasize the importance of well-defined and enforceable property rights and of the minimization of corruption. These factors can be tested by using the indices of corruption for different countries provided by Transparency International, the indices of property rights from Freedom House, and the index of economic freedom from the Heritage Foundation. (5) A widespread opinion is that economic growth in these transitional economies would be positively correlated with well-defined property rights and negatively correlated with the degree of corruption.
(b) Governance, democracy and civil liberties. The second group of hypotheses concerns the characteristics of government and civil liberties. Some variables from the widely used Polity IV dataset were selected, namely the indices of democracy (DEMOC), autocracy (AUTOC), and regime durability. (6) Another chosen variable in this group was the Freedom House index of civil liberties. Finally, the European Bank for Reconstruction and Development data on the percentage of GDP taken in taxes were included to assess the significance of the capacity of government to raise revenue on economic performance. The aim was to test whether the post-1989 performance in the transitional economies was in any way correlated with these variables.
(c) Social fractionalization and conflict. A third group of hypotheses address measures of social fractionalization and conflict. James Fearon's (2003) indices of ethnic and cultural fractionalization were used here, as they refine and supersede other similar datasets. There is significant evidence that ethnic fractionalization has a negative effect on economic activity in other developed and underdeveloped countries, including the United States (Alesina, Baquir and Easterly 1999) and Africa (Easterly and Levine 1997). The hypotheses here are that ethnic or cultural fractionalization might have had similarly negative effects in the transitional economies since 1989. The third variable in this group is an index of militant conflict for the 1989-2004 period. The hypothesis is that armed conflict has inhibited post-1989 economic recovery. (7)
(d) The legacy of past institutions. An additional hypothesis, elaborated by Jan Winiecki (2004), Steven Fish (1998) and others, is that the degree of post-1989 recovery is significantly affected by ingrained institutional and cultural factors, dating as far back as the pre-modern era. Winiecki identifies the zone of "Western Christendom" during the sixteenth century. This includes all the countries that were then predominantly Catholic or Protestant, but not those under the sway of Islam or Russian Orthodoxy. (8) The hypothesis is that the culture and institutions of Western Christendom were more amenable to trade and entrepreneurship. Winiecki draws attention to notions of liberty, entrepreneurship and trust, inherited from the sixteenth century or thereabouts. Similarly, Katchanovski (2000) and others have argued that Protestant and modern Catholic cultures are less amenable to corruption, in part because of the diminished role of family ties in civil society. Katchanovski provides data in the percentage of the population in each country that were Catholic or Protestant in the early 1990s and thus provides a second cultural variable (CATHPROT) used here. A third variable concerning the institutional legacy of the past is simply the number of years under communist rule prior to 1991 (YRSCOMM). Arguably, the greater the longevity of the communist period the greater the inhibition of pro-market institutions and cultural norms, and the more limited the capacity to develop such institutions after 1989.9A fourth variable in this group is the 1989 level of GDP per capita (GDP1989), which acts as a baseline for subsequent growth and thus, is expected to be statistically significant in explaining the 2005 level of GDP. Another variable OTTOMAN was used to indicate whether a country was a longstanding part of the Ottoman Empire. Yet another variable (MUSLIM) indicates whether it was predominantly Muslim. The latter two variables are used to test if the institutional or cultural legacy from Ottoman rule or Muslim practices affects recent performance.
(e) The depth and timing of the 1990s recession. Proponents of "shock therapy" argue that in difficult economic circumstances the more resilient, productive and energetic firms and organizations would be more likely to survive than the others, thus leading...