Partly true. Indeed, it took Western European countries about a decade after the war to re-establish full current account convertibility, first among themselves and then with the rest of the world. At no time, however, did these countries have foreign exchange regimes as distorted as in Uzbekistan today. This is evidenced by the fact that curb market premiums remained very small and multiple currency practices were rare. In Asia, Japan, Hong Kong and Singapore dismantled restrictions quickly, while countries like Korea and Taiwan took somewhat longer. By the mid-1960s most newly industrialized Pacific rim countries were free of current account restrictions. Most of the developing world followed in the 1980s and transition countries in the early 90s. Today, only a handful countries (e.g., Turkmenistan, Iran, Iraq, Sudan, Zimbabwe, Myanmar, Cuba and North Korea) have restriction on current account convertibility comparable to those in Uzbekistan. Current account convertibility should not be confused with capital account convertibility which took much longer to introduce in industrial countries and has not yet been fully established in many emerging economies.

Wrong. Unlike industrial countries after the war, Uzbekistan does not have the luxury to postpone current account convertibility. The economic environment in the 1950s and 60s differs profoundly from today. At that time international trade was still small, goods and services were mainly exchanged within bilateral trade agreements and most countries maintained fixed exchange rates and the gold standard. Since then, world trade and investment have grown exponentially, increasingly driven by private banks and corporations rather than governments. If a country wants to reap the benefits of globalization, it has to play by the minimum rules of the international market; current account convertibility is clearly one of them. This is already evidenced by Uzbekistan’s poor record in attracting foreign equity investment (in 2000 the lowest per capita level of all transition countries) and the decline of its foreign trade (by about 40% in the last four years).

Wrong. The main beneficiary of current account convertibility would be Uzbekistan itself. Removal of foreign exchange restrictions would increase consumer choice and satisfaction, reduce opportunities for corruption, improve competition and, by providing proper price incentives, enable the country to identify the industries where it has a comparative advantage. In the macroeconomic area, it would help Uzbekistan reverse the worrying trend of declining net foreign exchange inflows, low per-capita growth and high inflation. An increase of foreign trade would benefit both Uzbekistan and its trading partners. These were the points made by IMF Executive Directors, from Western countries and elsewhere, at their Board meeting on February 7, 2001, when they unanimously urged Uzbekistan to move to current account convertibility as soon as possible.

The IMF has neither the intention nor the means to "push" a country to pursue certain policies. Within its mandate, it shares with member countries its analysis of macroeconomic and exchange rate policies, taking into account 57 years of institutional experience in now 183 member countries. In the past, the IMF has helped many of its members with technical and financial assistance when they introduced current account convertibility. This is, after all, one of the founding principles of the IMF (Article VIII of the Articles of Agreement , Sections 2,3 and 4—henceforth referred to as Article VIII). By becoming a member of the IMF in 1992, Uzbekistan has subjected itself to the Fund’s jurisdiction in this area and pledged that it will not engage in discriminatory currency arrangements or in multiple currency practices that are not approved by the Fund. Today 149 (or about 80%) of IMF member countries have accepted the obligations of Article VIII (see map) and most others apply its principles in practice.

Wrong. A currency can be regarded as fully convertible when any holder is free to convert it at market rates—fixed of flexible—into one of the major international reserve currencies (like US Dollar, Euro or Yen). Russia and Kazakhstan, like most transition countries, have accepted the obligations under Article VIII and therefore maintain foreign exchange regimes that are free from current account restrictions. This means that there are no legal restrictions on obtaining foreign exchange to import goods and services, repatriate profits and service foreign debt. This is best evidenced by the fact that in none of these countries the curb market premium (if it exists at all) exceeds 10 percent. The only CIS country that has foreign exchange restrictions similar to Uzbekistan is Turkmenistan.

It is not current account convertibility, but the lack thereof that may lead to foreign debt default. For example, some Uzbek enterprises have been unable to service their non-government guaranteed loans because they could not legally convert their sum earnings into foreign exchange. Debt defaults also occur of course in countries with convertible currencies because enterprises, banks or government for whatever reason do not have sufficient domestic currency funds to buy the foreign exchange needed. But this has nothing to do with legal restrictions on convertibility.

Wrong. Real incomes of the population are determined by many factors and the degree of currency convertibility is only one of them. Experience in many countries shows, however, that economic liberalization, including in the foreign exchange area, leads to improvements in efficiency and investment, an increase of productivity and ultimately higher real incomes. Incidentally, this is corroborated by comparing average monthly wages, both in US Dollar terms and in real terms (i.e., corrected for consumer inflation) in Kazakhstan and Uzbekistan. In 1994 average wages stood at about USD 40 in both countries, but by the year 2000 dollar wages in Kazakhstan had more than doubled, while they were flat at best in Uzbekistan (see chart 1). If one uses a more appropriate measure of inflation or the curb market exchange rate (which is the only one relevant for the population), real wages have actually fallen in Uzbekistan over the past three years.

The exact opposite is true. By restricting convertibility and maintaining a highly overvalued official exchange rate Uzbekistan has discouraged the inflow of foreign

Maybe. Current account convertibility implies that there are no restrictions on making payments for imports of goods and services. It does not imply that residents are also free to purchase foreign exchange for making capital transactions, e.g., investments abroad (like depositing money at a non-resident bank). In practice it is of course difficult to distinguish the two, especially for cash operations. This is why reasonable documentation requirements (above generous thresholds) are in conformity with current account convertibility as defined in Article VIII. Experience in many countries shows, however, that capital flight can not be prevented by imposing currency restrictions, neither for current nor capital transactions. Residents will always find a way to illegally transfer capital abroad if they have no incentive to keep their funds in the country. This underlines the importance of sound macro economic policies, positive real interest rates and a business environment conducive to investment

Not necessarily. "Currency speculation," in the terminology of some Uzbek officials, refers to the use of the illegal curb market by individuals who cannot legally obtain foreign exchange for importing or saving. But such a market and all its associated opportunities for corruption only exists because there are trade and foreign exchange restrictions. Also, is it really appropriate to call a family father who buys cash forex to protect his savings against inflation a speculator? "Speculation" in wider sense occurs of course because those who obtain foreign exchange at the legal exchange rates can derive substantial profits from the black market premium.

Wrong. It is not clear what is meant by "internal convertibility," a term sometimes encountered in discussions with Uzbek officials. The standard definition is the right of residents to hold foreign-exchange denominated assets (e.g. US dollar deposit accounts with domestic banks), but this was already granted in Uzbekistan in 1994. If "internal convertibility" means some form of current account convertibility, the above statement is wrong. As pointed out below, the measures taken in 2000 have moved Uzbekistan’s foreign exchange regime away rather than closer to current account convertibility.

Wrong. The administrative adjustments of the exchange rates combined with loose monetary policies have actually led to an acceleration of inflation, which is already high by international standards. According to official statistics, consumer prices rose by 28.2 percent in 2000, compared to 26 percent in 1999; according to alternative estimates based on data collected by the IMF and other international organizations actual inflation was about twice as high. By any account, consumer price inflation in Uzbekistan was much higher than in Kazakhstan (9.8 percent) and Kyrgyzstan (9.6 percent).

Wrong. Few countries have seen a slump of their official exchange rate in 2000 that is comparable to the 132 percent devaluation of the official exchange rate and 18 percent of the curb market exchange rate in Uzbekistan (see chart 3). During the same time the Kazakh Tenge depreciated by 5 percent and the Kyrgyz sum by 6 percent.

Wrong. As explained above, current account convertibility is defined as the unhindered access to foreign exchange for current transactions and has therefore nothing to do with the level of the exchange rate. This is particularly the case if, like in Uzbekistan, the legal exchange rates continue to be manipulated by administrative means rather than market mechanisms. None of the measures taken in 2000 imply greater access to foreign exchange. On the contrary, import licenses have been revoked, some consumer good imports were centralized in one import company and cash conversion limits were tightened in August. Moreover, the foreign exchange market has become more segmented and several new exchange restrictions violating Article VIII have been introduced.

Not necessarily. There is a distinction between convertibility restrictions and multiple currency practices, both of which fall under the IMF’s jurisdiction. By administratively unifying the official exchange rate and the commercial exchange rate the government could eliminate a multiple currency practice and thus remove one of the legal obstacles to accepting the obligations under Article VIII. While a positive step in itself, this has nothing to do with introducing current account convertibility per se. The latter requires the removal of access restrictions to foreign exchange for current account transactions. Note that the existence of a black market exchange rate does not legally constitute a multiple currency practice. In reality, however, the curb market premium will become much smaller or even disappear once all restrictions on current transactions (such as maximum cash conversion amounts in exchange booths) are eliminated.

The Uzbek authorities sometimes argue that the administrative adjustments of the exchange rates in 2000 have reduced both their spread and their overvaluation and that this is a preparatory step for introducing convertibility. Moving the rates closer to their presumed equilibrium level may indeed be helpful. But how can the authorities know the "right" exchange rate level? A much better approach would be to step-by-step remove access restrictions and let the exchange rate(s) gradually adapt to a new equilibrium. The success of such a strategy depends crucially on the accompanying policies, i.e., reasonably tight monetary and fiscal policies as well as structural reforms that enable the economy to adjust to the new exchange rate. For example, the state order system will need to be reformed to make sure that a devaluation is also reflected in higher producer prices for cotton and wheat.

True. Demand for foreign currency is indeed strong, as evidenced by the increase in the velocity of money and the dollarization (many transactions, such as personal saving as well as purchases of second-hand cars and real estate are almost exclusively in dollars). This is evidence of the population losing faith in the domestic currency due to high inflation, little confidence in banks and negative real deposit interest rates. Whoever has sum holdings that he does not need for daily sum-denominated transactions (such as food purchases on the bazaars), has already converted them into hard currency on the black market.

Wrong. The high degree of dollarization mitigates the risk that the exchange rate will substantially depreciate beyond the level of the curb market, aside from a possible short-term overshooting for psychological reasons. People simply do not hold excess sum that could be used to buy dollars at exchange booths and thus drive up the rate. This depends crucially on a tight stance of monetary policy and a phased increase of wages. After the removal of access restrictions the CBU may in the short term wish to sell foreign exchange in the cash market to prevent a speculative attack (there may be expectations that the liberalization will be short-lived.) But the need to draw on official reserves for such interventions is likely to be very limited. At the prevailing curb market rate, the CBU would need less than 15% of official reserves to buy up the entire cash in circulation in the economy.

Unlikely. Most consumer prices in Uzbekistan already reflect the market-determined curb market exchange rate. The impact on overall inflation of introducing current account convertibility and unifying exchange rates (at a level somewhere between the official exchange rate and the curb market rate) will therefore be limited, provided that financial policies remain sufficiently tight. However, price increases can be expected for those goods and services that are directly determined by the official exchange rate (flour and some bread products, transportation and utilities). Such price hikes will primarily affect living standards of urban citizens with low fixed incomes (budget employees, pensioners, recipients of social transfers). Provisions will need to be made in the budget to compensate these people through direct income transfers. But the majority of the population living in rural areas will automatically benefit from foreign exchange liberalization, provided that state order prices for cotton and wheat are increased along with the devaluation of the official exchange rate.

Unlikely, but would this be so bad anyway? At its first attempt at introducing convertibility in 1995-96 Uzbekistan saw indeed a surge of low-quality consumer good imports. This reflected the preferences and low purchasing power of the population, but more importantly the maintenance of an increasingly overvalued exchange rate. The restrictions imposed in late 1996 may have put an end to such imports through legal channels, but this does not mean that they no longer exist. The black market is supplying the population with all the goods that it demands—illegally and at the equivalent of the market-determined curb market rate (imported chewing gum can still be bought at every street corner). If after the introduction of convertibility the exchange rate settles at a level close to the prevailing curb rate, there is no reason to assume that such imports will increase.

More profoundly, the question arises whether the government should at all try to restrict imports of consumer goods that are clearly demanded by the population (otherwise people would not buy them at the present curb rate). In a market economy, not the government but the consumer decides what he or she wants to buy. If the authorities are concerned about public safety of low-quality items (such as toys or food items), they can introduce a technical licensing requirement. If they are concerned about the protection of domestic competitors, they can introduce temporary customs duties (after further five years of institution building, the customs authorities should be in a better position to enforce such duties than in 1996). Regulating the inflow of certain consumer goods through the foreign exchange regime is ineffective, highly distortive and promotes the growth of the black market.

True. The introduction of convertibility would indeed have severe negative effects on those banks and enterprises that are now heavily subsidized through the privilege to buy foreign exchange at the overvalued official rate. Increased foreign competition would threaten those industries that are not fit for international competition. More immediately, a strong devaluation would increase debt servicing and import cost, especially of those specializing in import substitution (exports have been discouraged through the overvalued exchange rate).

Wrong. Ultimately, the cost of the exchange rate adjustment will automatically fall on the budget, since most foreign loans carry government guarantees and many enterprises are directly or indirectly owned by the state. If it chooses to continue to subsidize certain enterprises, the government can do so by making explicit transfers from the budget. The resources could come from taxing those who gain from convertibility, such as cotton and gold exporters (through resource taxes). In fiscal terms, the introduction of convertibility means nothing more than making the implicit transfers through the foreign exchange regime explicit in the budget. In principle, the current system that redistributes income from exporters to selected importers could therefore be continued even after foreign exchange liberalization.

The question rises, however, if the government should really continue to pursue its forced industrialization and import-substitution program (sometimes referred to as "localization"). Similar policies failed in many developing countries in the 1960s and 70s, for several reasons. First, they involve costly inter-sectoral transfers; in 1999, Uzbekistan’s exporters subject to the surrender requirement transferred around 16% of GDP to selected importers. Secondly, they are not transparent (implicit taxes and subsidies are not shown in the budget), causing problems of accountability and governance. Thirdly, they are not sustainable, as evidenced by the drying up of foreign exchange inflows to Uzbekistan. Finally, governments have traditionally proven to be bad judges of what "strategic" industries they should support. A market-determined unified exchange rate will be the best guide to Uzbekistan’s relative strengths. In the long run, the country’s comparative advantage will lie in those sectors that are profitable without relying on government subsidies.

Wrong. The causality is exactly the opposite: exports from Uzbekistan are declining because the lack of current account convertibility. Both official and commercial exchange rates are highly overvalued relative to the market-determined rate. This system imposes an implicit tax on exporters who have to surrender all or (in the case of non-traditional exporters) half of their foreign-exchange denominated earnings at the administratively controlled official exchange rate. In 1999, the implicit tax rate on strategic exports (such as cotton or gold) was 56% and for other exports 22%. If the foreign exchange regime is liberalized and the official exchange rate devalued, this will immediately improve the international competitiveness of Uzbekistan’s exports. On a small scale this already happened in 2000, when the government temporarily removed exports duties and surrender requirement for selected fruits and vegetables.

In any event, an increase of exports is not a necessary precondition for foreign exchange liberalization as Uzbekistan’s current level of foreign exchange reserves is sufficient to finance a move towards current account convertibility (this is elaborated in the next paragraph). Moreover, it may prove futile to wait for a cyclical recovery of world gold and cotton prices. There is little evidence that supports a substantial increase of these prices in the near future.

Wrong. In principle no additional international liquidity is needed when introducing convertibility, provided the authorities are prepared to let the market forces work fully and quickly, i.e., allow the establishment of a unified exchange rate truly determined by demand and supply. In practice, however, a country will want to ensure adequate international liquidity when making a systemic change like liberalizing the foreign exchange regime, for three reasons. First, to insure against exogenous shocks during the adjustment process (e.g., a drop in cotton prices in the case of Uzbekistan or a surge in oil prices in Poland). Secondly, to be able to stabilize the exchange rate by selling foreign exchange in the case of a speculative attack. Finally, to temporarily subsidize households, enterprises and banks without having to raise taxes.

Uzbekistan international reserves (about USD 1.2 billion at the end of 2000) are sufficient to address all these concerns, certainly for the time before and immediately following the introduction of convertibility. Relative to the size of imports and money supply, Uzbekistan is in a much better position than Poland in 1990, which had at its disposal USD 2.5 billion including credit lines from IMF and other donors. Today, Uzbekistan’s gross international reserves cover about 5.5 months of imports (4.5 months in Poland in 1990), which is a sufficient cushion in the case of unexpected export shortfalls. Moreover, Uzbekistan’s gross international reserves are enough to buy the entire money supply at the official exchange rate and exceed it almost threefold at the curb market rate (in Poland in 1990 reserves were only 23% of the money supply); such an amount is clearly sufficient to stabilize the unified market exchange rate.

The government may require external funds to support those negatively affected by a devaluation because it finds it difficult to mobilize these resources domestically (by imposing new taxes or allowing an inflationary budget deficit to emerge). Note, however, that these needs will not arise immediately with the removal of convertibility restrictions. Many significant steps towards convertibility can be taken without affecting favored banks and enterprises (like legalizing the black market, eliminating foreign exchange licenses, allowing non-strategic enterprises to operate at a truly-market determined commercial exchange rate). The cost will also depend crucially on which enterprises, banks and households the government chooses to subsidize. The IMF, the World Bank, the Asian Development Bank and bilateral donors stand ready to finance some of the adjustment cost once credible and irreversible steps towards current account convertibility have been taken. In the medium run, the introduction of convertibility will increase exports and FDI and thus create the basis for new investments, both financed by foreign credits and equity.