THE IMF AND THE RUBLE ZONE. Richard Pomfret. Professor of Economics, Adelaide University

Transcription

1 September 2001 THE IMF AND THE RUBLE ZONE by Richard Pomfret Professor of Economics, Adelaide University Mailing Address: Richard Pomfret, School of Economics, Adelaide University, SA 5005, Australia.

2 2 ABSTRACT When the fifteen Soviet successor states joined the IMF in 1992, the most pressing monetary question was whether to retain the ruble or to issue national currencies. IMF officials have argued that the IMF s role was to present the arguments for and against alternative arrangements. This is not how policymakers in the new independent countries and other observers perceived the IMF s position. In 1992, the IMF was seen as advocating retention of the ruble zone, although its position changed and in 1993 it supported introduction of new national currencies. This paper analyses why the IMF adopted a position, and discusses some of the consequences.

3 3 THE IMF AND THE RUBLE ZONE The International Monetary Fund (IMF) has been given a hard time over its policies towards the former Soviet Union. After a brief experience with the region, starting essentially from the report on the dying Soviet economy (IMF, 1991), the Fund was suddenly expected to provide financial and technical support for the fifteen successor states which all became IMF members in The IMF was then criticized for both doing too little and doing too much; Sachs (1994), for example, accuses the IMF of failing to provide enough financial support to Russia, while Wedel (1998) criticizes the IMF for the abuse of its power by distorting policymaking. This paper focuses more narrowly on the role of the IMF in providing policy advice, dealing specifically with the choice between maintaining the common currency inherited from the USSR or issuing national currencies. This was an important episode, because the choice arose immediately after the dissolution of the USSR in December 1991 and was seen by most of the successor states as falling squarely within the IMF s technical competence. In a recent paper co-authored by the Director of the IMF Department dealing with the former USSR, Odling-Smee and Pastor (2001) attempt to set the record straight and to correct some misconceptions that have arisen about the IMF s policy advice to the fifteen countries and its role in the break-up of the ruble zone. Of their paper s two substantive sections, the first is a clear and objective account of the dissolution of the ruble zone, but the second is a more tendentious review of the IMF s advice, exonerating the IMF from charges that it advocated maintaining the ruble as a common currency. In offering an alternative view of the IMF s advisory role, my aim is not to use hindsight to criticize the IMF for lacking perfect foresight.

4 4 It is, however, important not to let the protagonist s view stand as a definitive account of what happened, and it is also important to understand why so many well-informed observers believe that the IMF advocated retention of the ruble zone in The Dissolution of the Ruble Zone The facts, and even the fundamental cause, of the dissolution of the ruble zone are not in dispute. In January 1992 all fifteen Soviet successor states used the ruble, and each of the new nations central banks was a credit-creating centre. With hindsight, the institutional situation was unstable because of the free-rider problem. 2 Each government gained all the seigniorage from its own credit creation, but only bore a fraction of the inflationary costs, which were spread over the whole ruble zone. In early 1992, however, this flaw was far from universally recognized as decisive. Moreover, breaking the institutional constraint was difficult, because the dissolution of the USSR was unexpected and many participants attempted to maintain the common currency in order to reduce economic disruption. 3 The collapse of the ruble zone took two years, during which serious monetary stabilization was not possible for Russia or other ruble-users. 4 The three Baltic countries and Ukraine issued national currencies in The role of IMF advice on whether or not to abandon the ruble was less significant for these four countries, because they had decided their answer before joining the IMF. Russia appeared torn between trying to break up the ruble zone in order to achieve control over monetary policy and using levers available within the ruble zone (such as its monopoly over banknote printing) for political ends. The role of the IMF in this choice was significant but not critical in Moscow, because Russia had other sources of

5 5 economic advice internally and externally. 5 The role of the IMF as a technically competent analyser of monetary matters was particularly important for the other ten countries. Exit from the ruble zone during 1993 was staggered (Table 1) because countries placed differing weights on the benefits of monetary independence and the costs of losing perceived Russian largesse and of disrupting trade with other ruble users. The Kyrgyz Republic introduced a national currency in May Over the next four months parallel currencies which had been circulating in Azerbaijan, Belarus and Georgia became national currencies The largest exodus from the ruble zone occurred In November 1993, Turkmenistan, Kazakhstan, Uzbekistan, Armenia and Moldova introduced their national currencies, after which only war-torn Tajikistan continued to use the Soviet ruble. The ruble zone had collapsed, and the Soviet successor states had national currencies over which they had full control The IMF s View The actual situation and understanding of the currency zone issue evolved rapidly during 1992 and 1993, and it is important to locate evidence in time. During the early months of 1992, there was an active debate over the merits of a common currency. After the July 1993 introduction of Russian banknotes, and declaration that the old Soviet rubles were no longer legal tender in Russia, the options were reduced to having a national currency or using a Russian-managed common currency, and in September and October 1993 it became apparent that the Russian terms for membership in the ruble zone were unacceptable to the other CIS countries.

6 6 Odling-Smee and Pastor (2001, 12) portray the IMF as playing the role of providing the policy options in a neutral manner: In discussions with the ten countries (other than Russia) that did not declare early on for national currencies, the IMF staff explained the economic advantages and disadvantages of the national currency and cooperative ruble area options In annexes they provide documentation from October 1991, February 1992, and May 1992 showing the IMF rehearsing both sides of the arguments. The May 1992 document, presented at a meeting of Heads of Central Banks of the ruble zone in Tashkent, is about Coordinatin of Monetary Policy in the Ruble Area and the only real reference to the national currency option is a warning about the dangers of failure to cooperate in managing the common currency: Eventually, states could be forced to introduce their own currencies before they were adequately prepared to conduct an independent monetary policy. (Odling-Smee and Pastor, 2001, 30). In 1992 a noncooperative ruble area (managed by the Central Bank of Russia) was considered to be a political non-starter, because the other successor states would not accept Russian domination. When after July 1993 Russia presented this as the only alternative to national currencies, the balance of argument shifted and the IMF prepared to help more actively in the introduction of national currencies. This is a selective provision of documentation. An unmentioned influential paper was Max Corden's report for the UNDP/World Bank Trade Expansion Program, which with some caveats, recommended creation of a ruble zone and specifically pointed to the Central Asian countries as the ones for which the argument was likely to be strongest: "If a republic is small, if the argument for fixing the exchange rate to the ruble is strong because trade with Russia is expected to dominate the country's trade for a long time, and if Russia is expected to succeed in stabilizing its economy, then the case for going all the way into a monetary union with Russia becomes strong. Perhaps these conditions apply for the Central Asian republics. If the

7 7 intention is to maintain a fixed exchange rate indefinitely, it is better to lock it in through an institutional arrangement and thus avoid any foreign exchange speculation." Corden (1992, 14-15). 7 Corden s influence was large because of his academic stature as one of the leaders in international economics over the last four decades and because in he was working for Johns Hopkins University in Washington DC and had connections with both the IMF, for whom he had been a full-time adviser in , and World Bank. Although not officially an IMF document, this argument was cited by IMF officials. Several IMF documents in the first half of 1992 explicitly advocated maintaining the common currency. The argument was set out in the International Monetary Fund s spring 1992 World Economic Outlook (p. 41-2), where introduction of new currencies was discouraged on the grounds that (a) macroeconomic stability should be achieved first and (b) intra-cis trade would be disrupted. 8 The first Economic Reviews of CIS countries published by the IMF in spring 1992, repeated the policy recommendation that the Central Asian countries should remain within the ruble zone. The Central Asian and Caucasus countries and Moldova followed this advice in The IMF had reversed its position by the end of When the Kyrgyz Republic introduced its national currency in May 1993, it received immediate support from the international financial institutions. After the confiscatory issue of Russian rubles in July 1993 indicated that Russia had abandoned the option of a cooperative common currency area, the IMF clearly stood ready to support the issue of national currencies by the remaining ruble zone members, and was presumably pointing out that the balance of advantages had swung overwhelmingly in that direction. In sum, the situation is fairly clear in 1993 when the weaknesses of the ruble zone had become more apparent, and the IMF was giving appropriate advice. The

8 8 situation in 1992 was, however, less clear. There is considerable evidence, stronger in the spring and weaker as time passed, that the IMF was advocating maintenance of the ruble zone as the best practical option. Does this history of IMF thinking matter? The endorsement of the ruble zone on optimum currency area grounds and failure during the first two-thirds of 1992 to provide strong warnings of the inflationary consequences of retaining the ruble became prima facie evidence in several Soviet successor states of IMF fallibility on technical matters. This was important because on other matters, where the economic case for recommended policies was much stronger, the IMF found that it had squandered credibility. 10 The credibility issue would be less severe if the IMF had truly presented pros and cons of alternative monetary arrangements as Odling-Smee and Pastor suggest. This was not how the IMF position was perceived in the new capital cities. 11 The first generation of IMF resident representatives were invariably associated with a company line rather than being viewed as two-handed economists. The impatience with assessing a menu of options partly lay on the side of the new governments who sought answers rather than options, but it also reflected an IMF ethos. Both the IMF and the World Bank have difficulty reconciling the pressures for a single company line on key policy issues with the inherent imprecision of answers to major economic questions. It would have been better to air pros and cons of a common currency rather than trying to present a unified front in the first half of Similarly, the Washington consensus of the early and mid-1990s was not shared equally by all IMF and World Bank staff, but it remained the company line until a new position was adopted in the later 1990s. The challenge to the existing consensus was articulated forcefully by the World Bank s new chief economist (Stiglitz, 1998),

9 9 but diversity of opinion was not welcomed and Stiglitz had to go before the two institutions reached a more consensual position. Why did key policymakers believe that the IMF advised retention of the ruble zone in 1992? The simple answer is because it is true. For most of 1992 the IMF warned of the disruption to trade that would result from separate currencies and advocated stabilizing the ruble before considering issuance of national currencies. Even in 1993, when publications in Washington had shifted their emphasis towards support for national currencies, IMF resident representatives were still advocating caution. A fundamental point is that the IMF position as perceived by policymakers in the new independent states was forged much more by verbal communications, which tended to come quickly to the point, than by publications (typically in English), which careful set out assumptions and caveats. Odling-Smee and Pastor (2001, 19) argue that In contrast to the neutral position that the IMF held with respect to the two options, there has been a widespread view that the IMF opposed national currencies and sought to maintain the ruble area. Their accompanying footnote lists exponents of the widespread view and includes Yegor Gaidar, Jeffrey Sachs and Anders Aslund, all central figures in Russian policymaking and in contact with the IMF in Brigitte Granville has written: For more than a year after the collapse of the Soviet Union, the EC, IMF and other international financial institutions urged the former Soviet republics to preserve a single currency. This poor advice was based largely on the notion that these states should seek to minimise the dislocation of central planning s organic enterprise links (Granville, 1995, 65-6). 12 It is beyond credibility that all four, and the many others accepting this view, were all mistaken. 13

10 10 Lack of wider education on the prospects of national currencies was reflected in the generally negative headlines greeting new currencies. Here again Odling-Smee and Pastor are selective in their citations, quoting 1994 Economist headlines about Kirgizstan: The Reliable Referendum or The Invalid Responds to Treatment, rather than the May 1993 headlines of The Battle of the Som (The Economist) or Out of Steppe (The Far Eastern Economic Review). The Wall Street Journal ran a negative story about the lack of internal acceptance of the new Kyrgyz currency in its early days, which was practically obsolete before it was printed. Even in November 1993, The Economist headlined its page and a half story on the final collapse of the ruble zone: The Cost of Monetary Chaos. The IMF cannot be blamed for the journalistic bias against change, but it had certainly not contributed to a public airing of the advantages and disadvantages of alternative scenarios Conclusions The IMF faced unprecedented policy and technical issues with the sudden and unexpected collapse of the USSR and accession of fifteen new member states. This paper does not attempt to assess the IMF s overall performance in response to these challenges. What it does challenge is attempts to rewrite history to absolve the IMF of error. On the ruble zone, the IMF presumably had internal discussions and a variety of views undoubtedly flourished among its staff, but the company line for most of 1992 favoured maintenance of the common currency. This was not a stupid position, and indeed was shared by one of the most pre-eminent academic international economists (Corden, 1992), but it turned out to be mistaken in the post-soviet context.

11 11 The mistake was important because the IMF had an initial stock of credibility on technical matters which was squandered on this issue. The fact that politics mattered, and that with greater willingness to cooperate or to accept an independent central bank (neither politically feasible in 1992), the common currency could have been desirable, is not relevant. The technical advice was, with hindsight, wrong. All advisers are in a difficult position because policymakers want advice on what to do, without a range of caveats. The IMF erred in failing to identify from the start that the common currency was an issue on which it should not have succumbed to the desire for clear advice in That failure had a bad legacy even though the error was corrected by There is no lesson in this, other than that ex post claiming of infallibility serves nobody well. The advisers should be aware that they might be wrong and the policymakers too should be aware that the ultimate decision is theirs and any advice may be wrong.

12 12 Table 1: Date of Introduction of National Currency National Currency Previous Arrangements Estonia 20 June 1992 Latvia 20 July 1992 Coupon May 1992 Lithuania 1 October 1992 Coupon May 1992 Ukraine 12 November 1992 Coupon January 1992 Kyrgyz Republic 15 May 1993 Azerbaijan June 1993 Parallel currency August 1992 Georgia 11 June 1993 Coupon April 1993 Belarus September 1993 Parallel currency May 1992 Turkmenistan 1 November 1993 Kazakhstan & Uzbekistan 15 November 1993 Armenia 22 November 1993 Parallel currency Moldova 29 November 1993 Coupon July 1992 Tajikistan July 1995 Note: In several countries coupons or parallels currencies circulated while the ruble was still legal tender, and in some cases they were the sole means of payment for some transactions. The date given in column 2 is when the national currency became sole legal tender, apart from Azerbaijan where the manat, introduced as a parallel currency in August 1992, had become the dominant currency by June 1993 when it was announced that it would be sole legal tender (Pomfret, 1996, 110-1), but a coup d état interrupted these plans and the manat became officially sole legal tender only on 1 January In Latvia, Lithuania, Ukraine, Georgia, Uzbekistan and Tajikistan the initial national currency (Latvian rouble, talonas, karbovanets, coupon, sum coupon and Tajik ruble) were subsequently replaced by new currencies (lats, litas, hryvnia, lari, sum and somoni).

13 13 References Casella, Alessandra, and Jonathan Feinstein (1989): Management of a Common Currency, in Marcello de Cecco and Alberto Giovannini (eds.) A European Central Bank? (Cambridge University Press: Cambridge UK), Corden, W. Max (1992): Integration and Trade Policy in the Former Soviet Union, paper prepared for the UNDP/World Bank Trade Expansion Program, Washington DC, January. Flandreau, Mark (1993): On the Inflationary Bias of Common Currencies: The Latin Union Puzzle. European Economic Review, 37, Gomulka, Stanislaw (1995): The IMF-Supported Programs of Poland and Russia : Principles, Errors and Results, Journal of Comparative Economics, 20(3), Granville, Brigitte (1995): Farewell, Ruble Zone, in Anders Aslund (ed.) Russian Economic Reform at Risk (Pinter, London UK), Granville, Brigitte (2001): The Problem of Monetary Stabilization, in Brigitte Granville and Peter Oppenheimer (eds.) Russia s Post-Communist Economy (Oxford University Press, Oxford UK), Hofmann, Volker, and Friedrich Sell (1993): Credibility, Currency Convertibility and the Stabilisation of the Rouble, Intereconomics, 28(1),

14 14 IMF, World Bank, EBRD and OECD (1991): A Study of the Soviet Economy, 2 volumes. International Monetary Fund, Washington DC. Nikolić, Milan (2000): Money Growth Inflation Relationship in Postcommunist Russia. Journal of Comparative Economics, 28, Pomfret, Richard (1996): Asian Economies in Transition: Reforming Centrally Planned Economies. Edward Elgar: Cheltenham UK. Sachs, Jeffrey (1994): Russia s Struggle with Stabilization: Conceptual Issues and Evidence, paper presented at the 1994 World Bank Annual Conference on Development Economics and published in the conference proceedings, as a supplement to the World Bank Economic Review, 1995, Stiglitz, Joseph (1998): More Instruments and Broader Goals: Moving Towards the Post-Washington Consensus, WIDER Annual Lectures 2, United Nations University World Institute for Development Economics Research (UNU/WIDER), Helsinki, January. Wedel, Janine (1998): Collision and Collusion: The Strange Case of Western Aid to Eastern Europe St. Martin s Press, New York NY.

15 15 Endnotes 1 Fourteen of the fifteen former Soviet republics joined the IMF between April and September Tajikistan joined in April The free-rider problem has been analyzed in the context of other currency unions by Casella and Feinstein (1989) and Flandreau (1993). 3 Other political considerations included the use by Russia of its monopoly over banknote production to favour some CIS countries with cheap credit and fear that exiting from the ruble zone would expose a country to higher prices for energy and other imports. 4 The attempt by the first reformist government to establish price stability in Russia was fatally undermined by the increases in net credit to other ruble-using countries (Granville, 2001, 102-5). Although financing of payments imbalances became less automatic in July 1992, this change was shortly followed by the fall of the reformist government and appointment of a chairman of the Russian Central Bank who was not committed to price stability. Russia only achieved price stability after July I emphasize the importance of gaining national control over monetary policy, but there is also evidence of a direct relationship between the end of the ruble zone and Russian inflation, which is consistent with simply stopping the central banks free-riding behaviour; Nikolić (2000, 114) reports finding a structural break in Russia s inflation rate time series at the start of Some of these advisers view less benignly the IMF s role in Moscow (see next section), but the role was more influential and decisive in other capitals where few people had experience of national policymaking before In his analysis of the role of the IMF in Russia in Gomulka (1995), although critical of IMF advice on the currency issue, concludes that other (domestic) factors were more influential than IMF advice 6 The situation was messier than this account implies; see, Pomfret (1996,118-29). Ruble credits issued by different governments were valued differently and many parallel or substitute currencies circulated in , but the essential problem remained until November Corden s May 1992 World Bank working paper, Trade Policy and Exchange Rate Issues in the Former Soviet Union, Policy Research Working Paper WPS 915 (Washington DC, World Bank, May 1992), p. 27, contains essentially the same quotation. Russia s trade dominance turned out to be shortlived; by 1996 over half of the Central Asian countries foreign trade was outside the CIS -- Economic

16 16 Survey of Europe in (Geneva, United Nations Economic Commission for Europe, 1997), p This was generally viewed as the IMF position in In a January 1993 article, Hofmann and Sell (1993) question the credibility of what they refer to as the IMF concept of a three-stage program of first achieving macroeconomic stabilization through budgetary and central bank discipline, second implementing structural reforms, and finally addressing exchange rate and currency reform. 9 See, for example, the article New Currencies need the Support in IMF Survey, 26 October 1992, p There were many examples of this through the 1990s. Perhaps the most pervasive consequence was to undermine the IMF s emphasis on the need for monetary stringency in order to end high inflation. Some countries, eg. Uzbekistan, remain sceptical of IMF technical advice; in meetings in Tashkent in July 2001 senior policymakers still cited the currency issue in support of this scepticism. 11 Odling-Smee and Pastor (2001, 20) acknowledge the gap between the published record and the IMF position: Although the IMF retained a public posture of neutrality, by September 1992, it was privately advising delegates from countries still in the ruble zone that the balance of the economic arguments now favored national currencies. Note the tone of this statement the IMF had a position. 12 Brigitte Granville, now Director of the International Economics Program at the Royal Institute of International Affairs in London, was attached to the Russian Ministry of Finance between November 1992 and January 1994 and she was the main economic adviser involved in ruble zone negotiations during this period. 13 My own position as a UN adviser in six of the successor states during 1993 brought me into contact with policymakers and central bank officials who all believed that the IMF supported maintenance of the ruble zone even in early To some extent this reflected an era, when the IMF was instinctively secretive. During the 1990s the IMF made huge strides towards greater transparency and media-friendliness.