Constructing market-based economies in central Asia: a natural experiment?

• Author: Pomfret, Richard
• Position: Report

1. Introduction

   Following the collapse of central planning in Eastern Europe and the Soviet Union, an active and urgent debate ensued on the merits of alternative strategies for the transition to a market-based economy. In econometric studies of economic performance (typically measured by change in GDP) the strongest results were for variables that picked up the difference between Eastern Europe and the former Soviet Union, whether measured as distance from Dusseldorf or years under Communism or levels of human capital, and for some obvious negative correlations (e.g. between growth and war or hyperinflation). The extensive literature produced little or no consensus on the relative merits of gradual versus rapid transition or on the optimum sequencing of reform, because beyond the similarities of a Communist regime and central planning the thirty-something transition economies were a diverse group.

   A better basis for comparative analysis may be to take a smaller less heterogeneous group of economies. In this paper I ask whether the five Soviet Central Asian republics provide such a natural experiment. Apart from their similarity of geography and culture, the five Central Asian successor states were, together with Azerbaijan, the poorest Soviet republics in 1991. Their role in the Soviet division of labour was supplying a small range of primary products (cotton, oil and gas, and minerals). The Central Asian republics had no prior history as independent nation states. Their territory had been incorporated into the Russian Empire in the nineteenth century, so that they had roughly similar lengths of experience in the Tsarist Empire as well as the same Soviet history. The republics' borders were more or less arbitrary creations of Stalin; they could conceivably have been amalgamated as the Turkestan Soviet Republic, but instead they gained independence in December 1991 as five separate states. In four of the five republics, First Secretaries appointed by Mikhail Gorbachev became national presidents and quickly established super-presidential political regimes, with power heavily concentrated in the executive branch and with weak parliaments and legal institutions. The exception was Tajikistan which experienced a bitter civil war until 1997, but since then the president has established a super-presidential system. (4)

   When the five Central Asian countries became independent in December 1991, the political priority was nation-building. At the same time they faced three major economic shocks: the end of central planning, hyperinflation, and the collapse of demand and supply chains. Although the central planning system had been abandoned in the final years of the Soviet Union, reform had hardly begun in Central Asia, but when Russia introduced a big bang liberalization of prices in January 1992 the Central Asian countries, which still shared a common currency with Russia, were forced to follow. In the transition to a market economy, government spending exceeded revenues and inflation accelerated, turning to hyperinflation in 1992-3, which was exacerbated by the common currency zone. Finally, when transport networks were nationalized and borders were erected, the Central Asian economies were vulnerable to the rapid collapse of Soviet-era demand and supply links.

   All five countries suffered serious disruption from the dissolution of the USSR. Falling output and rising prices became much worse after the formal dissolution of the USSR removed residual central control over the Soviet economic space. Attempts to maintain economic links by retaining the ruble as a common currency in 1992-3 were abandoned by the end of 1993. (5) After 1993 each national government had considerable freedom of action in fiscal and monetary policy and in structural reforms.

   National strategies for transition to a market-based economy varied among the five countries, ranging from the most rapid reform of any former Soviet republic in Kyrgyzstan to minimal change in Turkmenistan. Despite similar backgrounds and economic challenges and commonalities such as super-presidential political systems under autocratic rulers, rampant corruption, geographical obstacles to trade, and reluctance to engage in serious regional cooperation, during the 1990s the five countries' economies became more differentiated. The divergence of transition strategies among countries with similar initial conditions and unchanging (or hard to change) geographical and cultural constraints encouraged researchers to view the five countries as a natural experiment of the efficacy of alternative approaches to creating a market-based economy.

   The end of the second decade since the dissolution of the Soviet Union is a good time for reflection because many developments have taken time to work themselves out (Pomfret, 2010a). Table 1 presents summary statistics of initial conditions and outcome indicators, Table 2 provides indicators of reform and Tables 3 and 4 report annual data on growth and inflation.

2. Dissolution of the USSR and the Transition from Central Planning

   During the 1990s, the national leaders cemented their personal power by creating super-presidential regimes, in which the balance of power between executive and legislature was overwhelmingly weighted towards the former. In Tajikistan, the only one of the five countries not to evolve peacefully from Soviet republic to independent state under unchanged leadership, the bloody civil war dominated political developments until 1997 and delayed implementation of a serious and consistent economic strategy, but by the end of the decade President Rakhmon had constructed a political system similar to that of his neighbours. In the priority task of nation-building the five countries have been successful; simply surviving for twenty years was an achievement that many did not anticipate in 1991. (6)

   What was the economic goal? By 1992 the centrally planned economy had collapsed and there was no alternative to creating a market-based economy (although there was much discussion of the varieties of capitalism). The experiment thus became how to create a market-based economy that could deliver superior economic performance, proxied by increased real GDP.

   Which country was most likely to succeed? Viewed from the starting gate, there were several potential favourites. The biggest windfall gain from the shift from Soviet administrative prices to world prices was expected to go to Turkmenistan: Tarr (1994) estimated a terms of trade gain of 50%, second only to Russia among former Soviet republics, and well-ahead of third-place Kazakhstan (19%). Kazakhstan, however, had the best endowment of human capital, reflected in a per capita income roughly twice as high as the average for the four southern Central Asian countries (World Bank, 1992) and the country's substantial energy resources were about to be tapped by the Chevron-Tengiz oil project--the largest foreign investment agreement in Soviet history. Uzbekistan benefitted from the position of Tashkent as the administrative capital of Soviet Central Asia, insofar as the new country inherited greater bureaucratic capacity and material (e.g. the large inherited aeroplane stock gave the new national carrier an initial edge and inherited military equipment reduced pressure on the defence budget). The poorest prospects, apart from Tajikistan which was embroiled in civil war, appeared to be those of Kyrgyzstan, which was poorly endowed with human and resource capital and geographically isolated. However, poor prospects could be an advantage if they pushed the country into more radical (and more appropriate) reform.

   Which country won the race in the 1990s? Based on real GDP in 1999 as a percentage of 1989 GDP, the answer is clear. All former Soviet republics experienced a transitional recession, but the decline was least in Uzbekistan. The worst-performing Central Asian economy, unsurprisingly, was Tajikistan. The other three countries had more or less identical performances, with GDP about a third lower in 1999 than in 1989 (Table 3), despite the diversity of transition strategies. (7)

   Why was Uzbekistan so successful? The main explanations have been in terms of a gradual transition strategy, favourable world market conditions for its major export, cotton, and good administration (and favourable inheritance from the Soviet era). (8) These are not necessarily competing explanations, and there is some interaction. Gradualism reduced the magnitude of the short-term transitional recession compared to countries adopting rapid reform. Good administration helped to reduce the output fall, e.g. maintenance of irrigation channels and ginning facilities kept cotton output up in contrast to Turkmenistan and Tajikistan where cotton output declined. Compared to elsewhere in Central Asia, some state enterprises were relatively successful in the early and mid-1990s, e.g. Uzbekistan Airways provided the best international air services and turned Tashkent airport into the regional hub. However, what transformed these factors into the best performance in the former Soviet Union was the 1992-5 upturn in world cotton prices (Figure 1) combined with the relative ease of transporting cotton to world markets.

   Uzbekistan's success was seized upon by some participants in the virulent debate about the appropriate speed of reform as evidence of the benefits of gradualism. (9) On the other hand, supporters of "shock therapy" (or Big Bang reforms) argued that Uzbekistan had failed to lay the foundations for sustained economic growth. Their scepticism was borne out by Uzbekistan's mediocre performance in the next decade, but that had less to do with gradualism than with poor policies. When cotton prices fell in the second half of 1996, Uzbekistan resorted to draconian foreign exchange controls to maintain state control over the rents from cotton. The controls bred a series of policies injurious to a...