Country experiences with monetary operations and government securities markets have been highly diverse, but a general pattern seems to have emerged since early 1992: most countries have moved toward more central bank operational autonomy and increased reliance on indirect instruments and on market-based interest rates. In addition, with few exceptions, reforms have led to a slow but sustained growth of the interbank and government securities markets. The sequencing of reforms has varied widely across countries, because of different political and macroeconomic conditions, and faster developments in one central banking area have not always been accompanied by similar progress in others (Table 2.1).

Country experiences with monetary operations and government securities markets have been highly diverse, but a general pattern seems to have emerged since early 1992: most countries have moved toward more central bank operational autonomy and increased reliance on indirect instruments and on market-based interest rates. In addition, with few exceptions, reforms have led to a slow but sustained growth of the interbank and government securities markets. The sequencing of reforms has varied widely across countries, because of different political and macroeconomic conditions, and faster developments in one central banking area have not always been accompanied by similar progress in others (Table 2.1).

Table 2.1.

Country Rankings in Monetary Operations and Government Securities Markets

article image
Note: Estonia and Lithuania have currency board arrangements and thus do not actively manage banking system liquidity. Hence, they are excluded from this classification.

Recent Developments

The following sections review specific steps taken by central banks during 1997 to enhance the monetary transmission mechanism through the use of indirect instruments, and also survey the important changes since the onset of reforms.

Monetary Policy Framework

Since 1992, the enactment of central bank laws in the Baltics, Russia, and other countries of the former Soviet Union has provided most of the countries’ central banks with a considerable degree of formal autonomy to carry out an independent monetary policy.1 Central bank legislation has also provided a clear objective for monetary policy—price stability. This target is set out in the recently enacted central bank laws of Armenia, Azerbaijan, Georgia, the Kyrgyz Republic, Russia, Tajikistan, and Uzbekistan.2 Over the same period, central bank credit to the government has declined steadily, as governments have moved toward market financing through the issue of government debt, and some countries have explicitly prohibited central bank lending to the government (Estonia and Lithuania, which operate currency boards, and Kazakhstan and Kyrgyz Republic).

At present, most countries surveyed in this report are using monetary programming techniques and adopting monetary operations to foster market development and monetary control. Most countries have established medium-term operating frameworks and targets for monetary policy that follow performance criteria for central bank balance sheets developed under IMF-sponsored macroeconomic adjustment programs.3 Success on this front has been mixed, as changing macroeconomic conditions often conspired against the formulation of monetary policy based on stable long-run relationships between monetary aggregates and economic activity.

Less progress has generally been achieved in short-term forecasting and liquidity management, which has hampered the coordination of a mix of indirect monetary policy instruments. Day-to-day liquidity management has also been hindered by limited coordination between operations in the domestic money market and foreign exchange operations (Belarus, Moldova, Russia, and Turkmenistan), and difficulties in forecasting certain balance sheet items such as government cash positions (Georgia, Kazakhstan, Latvia, Tajikistan, and Turkmenistan) and net other items. More important, the lack of fully developed interbank and secondary government securities markets affects the functioning of the monetary transmission mechanism, thereby limiting the ability of central banks to link short-term liquidity management with the medium-term objectives established in the monetary program. These deficiencies have tended to limit the scope for more active liquidity management through the use of indirect instruments. Nevertheless, substantial progress has been achieved in the Kyrgyz Republic, Russia, and, to a lesser extent, Kazakhstan.

Markets and Interest Rate Management

While interbank money markets are in place in most of the 15 countries and have begun to play a role in redistributing liquidity among financial institutions, activity in many cases remains limited. After suffering setbacks in 1995 and 1996 because of banking crises or concerns regarding counterparty risk, Kazakhstan, Latvia, Lithuania, and Russia resumed the development of interbank money markets in 1997. This was due in part to a decline in credit risks as unsound banks were liquidated or reorganized and supervision was strengthened (Kyrgyz Republic), and to a gradual buildup of the stock of usable collateral (Kyrgyz Republic and Moldova). The interbank market is also increasingly used for liquidity management in Armenia, although market growth has been hindered by bank solvency problems. In most cases, inadequate legislation on collateral remains an obstacle, and the lack of accurate information to banks on settlement balances in Georgia, Kazakhstan, Russia, and Tajikistan has prevented these countries from using the market for end-of-day funding operations. In Russia, the lack of settlement balance information was a serious hindrance for end-of-day interbank activity, although the situation improved in February 1997 when the central bank introduced multiple clearings. Russia has also experienced market segmentation (both geographically and across market participants), which has been a further constraint on the efficient redistribution of liquidity. In Turkmenistan, interbank market activity remains stagnant, despite improvements in the timeliness of settlement balance account information.

Regarding the primary market for government securities, the share of government deficits financed through the issue of treasury bills has increased steadily. Primary issuance of bills, however, has seldom been used by the central banks for short-term liquidity management. In some cases—Armenia, Kazakhstan, and the Kyrgyz Republic—the stock of government securities in the central bank’s balance sheet has been built up by securitizing existing government debt and government credit lines. To compensate for the lack of monetary instruments used to absorb liquidity, central banks in Belarus, Estonia, Kazakhstan, and Uzbekistan have issued central bank securities and in Armenia, Latvia, and Russia have offered central bank term deposits on a temporary basis. Interest rates on these instruments are determined at auctions, except in Latvia and Russia, where the rates on central bank deposits are administered. The primary market is active in all countries except Estonia and Tajikistan, and volumes outstanding are growing. Since the central banks of Estonia and Lithuania operate under a currency board agreement, they do not engage in active liquidity management. Tajikistan has plans to introduce treasury bills in the near future. Two countries, Azerbaijan and Uzbekistan, began to issue treasury bills in 1996, while Georgia began to auction treasury bills in August 1997. Turkmenistan issues a limited amount of treasury bills, but not in an auction format.

In the primary issues of securities, the central bank usually acts as the government agent and the ministry of finance decides on the timing, volume, and cutoff prices. There has been a lengthening in the maturity of the securities issued in the more established markets as inflation has declined. In a number of countries—Armenia, Kazakhstan, Latvia, Lithuania, Moldova, Russia, and Ukraine—nonresident investors strongly influence demand.

The central banks in Kazakhstan, Latvia, Ukraine, Russia, and, more recently, Armenia, Lithuania, and Moldova have begun to conduct transactions in the secondary market for government securities. Over-the-counter secondary market trading was recently introduced in Armenia and Lithuania. During 1997, the Bank of Lithuania reintroduced repurchase operations in treasury bills, while the Central Bank of Armenia initiated such operations in treasury bonds after an earlier, experimental use of similar operations in bills. Ukraine also conducts limited repurchase operations for liquidity management purposes. Open market operations in the secondary market still play a limited role in liquidity management owing to a lack of market depth and, to a certain extent, the absence of an institutional framework for conducting repurchase agreements. Many of the countries have favored the implementation of electronic trading systems to reduce counterparty risk and to increase the transparency of operations in markets where participants lack a meaningful track record and information is scarce. The Kyrgyz Republic installed an electronic trading system in April 1998, and Moldova had plans to implement one in 1998. In other countries, secondary markets are small or nonexistent, owing in part to the recent move to market-based monetary policy (Armenia4 and Azerbaijan), or because of the absence of a meaningful primary treasury bill market (Turkmenistan). The absence of adequate systems for payments, book entry, or information linking market participants, and uneven tax treatment of earnings from securities portfolio, as in Lithuania, Moldova, and, until 1997, Armenia, also limit market development.

Controls on deposit and lending rates have been eliminated in most countries. Kazakhstan abolished its indirect control on interest rates in 1996; Ukraine eliminated the mandatory link between lending rates and the refinance rate in 1995; and Georgia liberalized prudential limits on the mobilization of household deposits in January 1997. Belarus, however, restricts lending rates on loans using central bank funds from exceeding the refinance rate plus a pre-set margin; the country eliminated minimum rates on deposits in 1995, but reintroduced them in April 1997. Uzbekistan maintains a limit on lending rates at 150 percent of the refinance rate. In 1996 and 1997, Turkmenistan loaned a substantial amount to selected industries and agriculture by presidential decree at rates below its central bank refinance rate. Some countries have linked the refinance rate to market-based interest rates. In Azerbaijan and Moldova the refinance rate is linked to the credit auction rate, and in the Kyrgyz Republic to the treasury bill rate. In other cases, the refinance rate is determined on the basis of the expected inflation rate, with the aim of keeping it positive in real terms, as in Kazakhstan, Tajikistan, Turkmenistan, and Ukraine. This approach is also followed in Belarus, but it has relatively little impact since most credit (almost 90 percent) is extended at a subsidized rate below the refinance rate.

Instruments and Monetary Policy Implementation

Direct instruments are now seldom used in the 15 countries. Bank-by-bank credit ceilings have been phased out in all cases, including Tajikistan in 1997. Turkmenistan and Uzbekistan are the last countries to impose limitations on cash withdrawals. While most countries have lifted controls on interest rates, Belarus and Turkmenistan currently use directed credits with subsidized interest rates set below their refinance rates on a regular basis, and in Uzbekistan directed credits are granted for crop financing in specific situations.

Reserve requirements have been an important monetary policy instrument in the countries under review, and compliance has improved over time, including in Belarus, Georgia, Kazakhstan, and Russia. In Russia, compliance had been particularly weak in 1995, but it improved in 1996 as a result of regulations allowing the central bank unilaterally to transfer balances of delinquent banks from their correspondent accounts to reserve accounts. Most countries have chosen or recently moved to uniform ratios for different maturities and currencies.5 Countries with nonuniform requirements—Belarus, Uzbekistan, and, until February 1998, Russia—typically set lower ratios for foreign currency deposits. Russia has made several changes in its reserve requirements since 1997 and in early 1998 introduced a uniform 11 percent reserve requirement for all commercial banks. Uzbekistan also maintains different ratios for different categories of banks. Meanwhile, Belarus has established a reserve ratio that is differentiated by maturity, ranges from 5 percent to 17 percent, and applies to both domestic and foreign currency deposits of up to one year. Lithuania is another country that differentiates by deposit maturity. Most countries stipulate that reserve requirements on both domestic and foreign currency deposits be met in the domestic currency.6 The exceptions are Armenia, Azerbaijan, Belarus, and Kazakhstan, which allow the commercial banks to choose the currency in which to meet reserve requirements on foreign currency deposits.7 While Azerbaijan, Belarus, Georgia, Russia, Tajikistan, Turkmenistan, and Uzbekistan keep reserve balances blocked throughout the compliance period, the rest of the countries have adopted reserve averaging.8 Only Belarus, Estonia, the Kyrgyz Republic, and Moldova remunerate required reserves.9

Overdraft facilities that provide uncollateralized short-term credits at penalty rates are in place in Azerbaijan, Belarus, Russia, and Tajikistan.10 In Turkmenistan, the central bank has introduced an auxiliary credit facility that provides banks with short-term credits. The rate is determined at a margin over the refinance rate to limit recourse to this facility.

Many countries—Armenia, Georgia, Kazakhstan, Kyrgyz Republic, Latvia, Moldova, and Ukraine—have moved toward the adoption of collateralized standing facilities (e.g., a Lombard window) to reduce the central bank’s credit risk, as the stock of usable collateral (usually treasury bills) has increased. Lending through this facility is short term at a penalty rate. Russia has introduced a repo facility to allow primary dealers on the securities market to settle transactions. Azerbaijan established a rediscount mechanism in 1997, which allows banks to sell government securities to the central bank at a discount in an emergency. In Armenia, the central bank provides extended, uncollateralized “systemic” credit for up to one year. Countries with a currency board—Estonia and Lithuania—provide extensions of central bank credit to commercial banks within the limits imposed by the amount of foreign exchange reserves in excess of those needed to provide full backing of base money.

Credit auctions were introduced in most of the countries as a first step toward market-based monetary instruments and have been used at some time in all countries except those with a currency board—Estonia and, since 1995, Lithuania. Tajikistan introduced auctions in December 1997, and has carried out several since then. As the volume of treasury bills and government securities in the market has risen, the rate of collateralization required for credit received in the credit auction has increased, gradually transforming the uncollateralized credit auctions into the equivalent of repurchase agreements. This phenomenon has occurred in Armenia, Latvia, Lithuania, Russia, and Ukraine. In addition, most countries have broadened the range of admissible collateral beyond treasury bills to include foreign exchange holdings, promissory notes, and some less liquid assets. Because of these developments, credit auctions are being phased out and are currently used actively only in Moldova, Tajikistan, and Uzbekistan. Credit auctions have been replaced by repo auctions in Latvia, and were recently suspended in the Kyrgyz Republic. In Moldova, the credit auction was expected to be fully replaced by open market operations by the end of 1998. In Turkmenistan, the auction was recently discontinued to avoid excessive liquidity expansion. In Russia, until mid-1996, low interbank market rates reduced interest in the credit auctions. The lower interbank rates were largely a reflection of stronger demand from regional markets outside Moscow. Auctions have not been frequent in Belarus, owing to the authorities’ emphasis on directed credits. In Azerbaijan, recent improvements in regulation and information dissemination may foster the development of credit auctions.

Inflation, Financial Deepening, and Dollarization

Monetary programming has been hampered by unstable relationships between monetary aggregates and economic activity (this stems from sharp changes in money velocity) and inaccuracies in measurements of domestic product on which quantitative projections of money demand are based. Medium-term targets for inflation, for example, may be difficult to achieve simply because they are based on projections of the growth in monetary aggregates, which may be subject to wide fluctuations in velocity. Only recent data, usually comprising no more than three years, provide meaningful information, since the earliest observations displayed fluctuations that were largely caused by the one-time effect of the price liberalization shocks beginning in 1992. In addition, in most of the economies high inflation and a lack of confidence in the banking sector have induced a flight from the domestic currency (reflected in a large share of bank deposits held in foreign currencies) and a marked preference for liquidity (reflected in the relative importance of cash holdings in the broad money aggregate). Despite the difficulties in implementing a monetary program, some reference monetary program is needed in each of the countries under review. With the passage of time, the economies will achieve greater stability in monetary relationships, and the monetary authorities’ ability to implement sound policies based on these data will be enhanced. Of course, those countries implementing a fixed exchange rate regime cannot follow an independent monetary policy. In fact, the more the exchange rate is fixed, the less the scope for an independent monetary policy. Monetary operations will be geared toward preserving the exchange rate parity and therefore will reflect conditions in the market of the peg currency, with a risk differential.

Developments in Monetary Aggregates

The economies of the 15 countries in this study seem to have gone through at least two clearly identifiable stages. At the beginning of the transition, a combination of limited supply and government controls kept prices arbitrarily low and resulted in a significant accumulation of currency holdings and bank deposits (mostly demand). This phenomenon, often referred to as “monetary overhang,” fueled high inflation rates at the time of the elimination of price controls. As prices skyrocketed, the real value of monetary aggregates fell significantly (Figure 2.1). With varying degrees of success, stabilization based on monetary restraint and tight fiscal policies rapidly brought inflation rates down to moderate levels. During this second phase, the real value of monetary aggregates stabilized, hovering around levels that were, on average, below those in developing countries (Figure 2.2 and 2.3). Figure 2.4 shows a high and increasing share of the currency component in broad money, which is much higher in the Baltics, Russia, and other countries of the former Soviet Union than in other developing countries. This reflects in part the reactions of economic agents in those countries to the recent inflation spike, and attendant uncertainties, particularly with respect to the credit quality of banks, in the earlier stage of reform.

Figure 2.1.
Figure 2.1.

Real Broad Money

(Deflated by CPI; 1995 = 100)

Sources: IMF staff estimates; and IMF, World Economic Outlook.Note: Simple average of 15 countries. 1997 data are preliminary.
Figure 2.2.
Figure 2.2.

Inverse Velocity

(M2/GDP; in percent)

Sources: IMF staff estimates; and IMF, World Economic Outlook.Note: Simple average of 15 countries. 1997 data are preliminary.
Figure 2.3.
Figure 2.3.

Inverse Velocity, 1997

(M2/GDP; in percent)

Sources: IMF staff estimates; and IMF, World Economic Outlook.Note: Simple average of 15 countries. 1997 data are preliminary.
Figure 2.4.
Figure 2.4.

Composition of M3

(In percent)

Sources: IMF staff estimates; and IMF, World Economic Outlook.Note: Simple average of 15 countries.

Money velocity has fluctuated significantly from 1991 through 1997, increasing substantially at the beginning of the reform, as the public reduced cash balances to minimize its exposure to price changes, a typical response to high and accelerating inflation. The public was reluctant to use the nascent banking sector, as indicated by a growing currency component of M3 (Figure 2.4), because of a lack of confidence in financial institutions, which was reinforced by several banking crises; a lack of familiarity with banking practices; and, in many cases, relatively low deposit rates, which stemmed from a paucity of profitable investment opportunities and excess liquidity. As Figure 2.5 shows, in most cases real deposit rates have turned positive only recently, albeit at low levels that have not induced a sizeable increase in bank deposits.

Figure 2.5.
Figure 2.5.
Figure 2.5.

Real Deposit and Lending Rates

Source: IMF staff estimates.Note: Real deposit rate data not available for Turkmenistan.


Dollarization presents both benefits and risks. For countries that have experienced extreme price instability and capital flight, dollarization has often played a key role in remonetization of the economy. But dollarization may also increase risks in the financial sector, including risks to foreign currency loan portfolios resulting from domestic currency devaluations, and to the ability of central banks to serve as lender-of-last-resort. Dollarized economies also have experienced losses in seignorage revenues.

Data on dollarization of deposits is at best fragmentary. There is no information on cross-border deposits, which are likely to have played an important part in the flight to quality during the first stages of the transition. Further caution is warranted when analyzing early developments, since changes in the dollar share of M3 may be in part explained by changes in the real exchange rate.11

Taking the above factors into account, the dollar plays a clear role in bank deposits in most of the 15 countries in this study, and its share of bank deposits has remained relatively stable as inflation has subsided and deposit rates have become positive in real terms (Figure 2.5).12 On average, dollar deposits represent close to 40 percent of total deposits and, given the strong preference for liquidity, foreign currency holdings may also be substantial.13

Monetary Policy Indicators

Even in those countries where inflation has been contained and some of the basic ingredients for an effective monetary policy—central bank autonomy and active government securities and interbank markets—have been put in place, much remains to be done in terms of liquidity forecasting and liquidity management. Figure 2.6 shows that reserve money displays significant volatility on a monthly basis, with average percentage changes ranging from around 4 percent in the Baltic countries to more than 10 percent in Tajikistan and Turkmenistan.

Figure 2.6.
Figure 2.6.
Figure 2.6.

Reserve Money

Source: IMF staff estimates.

Although progress has been made in most countries (as shown by the declining path of the 12-month volatility of changes in reserve money), monthly fluctuations are still significant.

More important, in most cases the link between reserve money and short-term market rates is weak or nonexistent, suggesting either that market participants do not engage in active liquidity management or that the interbank and securities markets are not yet developed to the point of actively reflecting changes in overall liquidity.14 In either case, the transmission of monetary actions to market interest rates is likely to be highly inefficient, thus limiting the scope for an effective monetary policy.

An alternative that monetary authorities may use to influence interest rates is the direct management of a reference or refinance rate. This mechanism is particularly relevant to underdeveloped markets where the mechanism of price discovery is hampered by insufficient volume and a high degree of macroeconomic uncertainty, which in turn may yield highly volatile interest rates and deter participation in the absence of such a reference rate. Although a case-by-case analysis is necessary to assess the direction of causality, a cursory look at Figure 2.7 seems to support the hypothesis that, in most of the countries reviewed in this volume, short-term interest rates (deposit rate and three-month treasury bill) tend to follow the refinance rate set by the government.

Figure 2.7.
Figure 2.7.
Figure 2.7.

Reserve Money

Source: IMF staff estimates.Note: Treasury bill rate data not available for Belarus, Georgia, and Tajikistan.

Country Rankings

Table 2.1 presents country rankings that illustrate the relative development of monetary operations and government securities markets in the economies under review. Classification levels and criteria are the same as those used in previous background studies prepared for the Basle meetings.

The first criterion covers the quality of central bank facilities, that is, the mechanisms through which the central bank is able to manage liquidity at its own discretion, and the collateralized standing facilities that are available to banks to relieve liquidity shortages. A rating of “substantial” progress indicates that the central bank has: the ability to conduct operations in the secondary market for securities, either in the form of outright purchases and sales of securities, as in Kazakhstan, Russia, and Ukraine, or through repos and reverse repos, as in the Kyrgyz Republic, Latvia, Russia, and Ukraine; or a well-designed and functioning Lombard-type facility with well-specified rules of access (Kyrgyz Republic, Moldova, Russia, and Ukraine). “Moderate” progress implies that some or all of those facilities are in place, but they are subject to shortcomings, as in Armenia, Georgia, and Uzbekistan. The remaining countries were assigned a “limited” progress rating (Azerbaijan, Belarus, Tajikistan, and Turkmenistan).15

The second criterion ranks banks on the basis of their operating framework for monetary policy implementation. Substantial progress implies that short-term monetary policy is based on the central bank’s monetary program and short-term liquidity forecasts and that the bank uses monetary instruments consistent with its program, as in Kazakhstan, Kyrgyz Republic, and Russia. Where short-term coordination between the monetary program and the use of short-term indirect instruments was in place but could be strengthened, as in Armenia, Azerbaijan, Belarus, Georgia, Latvia, Moldova, and Ukraine, a moderate progress rating was assigned. The remaining countries—Tajikistan, Turkmenistan, and Uzbekistan—showed limited progress.

The third criterion evaluates the development of two markets: the interbank money market and the secondary market for government securities. Countries with active markets for interbank money and secondary exchanges of government securities—Kazakhstan, Latvia, and Russia—earned a substantial ranking. Those with an active market in one of the two areas—Georgia, Kyrgyz Republic, Moldova, Ukraine, and Uzbekistan-were ranked moderate. Countries with activity in neither type of market—Armenia, Azerbaijan, Belarus, Tajikistan, and Turkmenistan—were ranked limited.

Overall rankings were calculated as follows: substantial progress was given to countries that received substantial ratings in at least two categories. A rating of moderate progress required at least a moderate ranking in at least two of the three criteria. The remaining countries received limited ratings.

Priorities for Reform

Significant progress has been achieved since the initial stages of reform at the start of the decade, and most of the 15 countries under review have already put in place the basic ingredients conducive to an effective market-oriented monetary policy. But active liquidity management through indirect instruments within a short-term monetary policy framework, including full coordination between monetary and foreign exchange operations, still lies ahead for most of the central banks.16 Although the underlying reasons are varied, all the countries must deal with the constraints imposed on liquidity forecasting and management by the lack of well-functioning interbank money markets and secondary markets for government securities—two elements that are crucial for the development of indirect monetary control.

Several factors impede active trading in the existing interbank markets: (1) a seeming lack of trust among market participants, reinforced by several past episodes of bank distress and concerns about the health of the financial system, as in Kazakhstan, Kyrgyz Republic, Latvia, and Lithuania; (2) inadequate regulation and supervision of bank activities; and (3) insufficient dissemination of information. An additional impediment relates to the large and irregular injections of liquidity through the extension of directed credits in some countries. Naturally, the phasing out of directed credits will be critical to development of the interbank markets in those cases. In this context, the buildup of the stock of collateral is likely to be beneficial for reducing credit risk, as is already apparent in the Kyrgyz Republic and Moldova. Moreover, efficient interbank clearing and settlement systems and timely information on correspondent account balances can stimulate daily trading of banks’ liquidity positions, as in Georgia and Russia. Finally, improving the supervisory capacity of the central bank can go a long way to contain credit risk, where market activity is hindered by concerns about the financial condition of individual banks (Armenia, Kazakhstan, Kyrgyz Republic, and Lithuania).

Another area for reform is that of the secondary markets for government securities. Except for a few countries—Belarus, Kazakhstan, and Russia—these markets are small or nonexistent.17 Standardization of trading arrangements, elimination of distorting taxes, and better information dissemination and transparency of primary auction results can help foster secondary market activity. In the medium term, a comprehensive framework for public debt management, yet to emerge in many cases, may further stimulate secondary markets. Pending tasks include extending the maturity profile of debt, capturing household savings, and creating a system of primary dealers to help develop secondary markets. Enhanced supervision and enforcement of performance criteria for primary dealers, and the establishment of rigorous settlement procedures, including a rapid move toward a delivery-versus-payment settlement, can generate a safer environment and more liquid securities markets.

As the volume of collateral increases, central banks should be expected to move toward collateralization of auctioned credit as a first step in replacing this instrument by repurchase auctions. The same is true for central bank uncollateralized credit facilities, which should eventually be transformed into Lombard windows. Such reforms, beside protecting the quality of the central bank assets, generate demand and liquidity in the underlying instrument. In the design of their facilities and open market operations, however, central banks must take care not to discourage the use of the interbank market as the first source of bank liquidity.

Yet another priority is the work that remains to be done regarding data collection and calibration of bank liquidity forecasts. Monitoring of banks’ excess reserves and interest rate movements should be conducted on a frequent basis, with special emphasis on the quality of the data. The use of desegregated data series and error control techniques, and a close monitoring and forecasting of government payments and receipts, which in turn requires a tight coordination with the treasury, may improve the accuracy of short-term projections.

A final priority concerns the need for the central bank to have the necessary volume of liquid and fully marketable securities in its own portfolio to be able to conduct substantial sterilization operations in the open market—all of which should be coordinated with the public debt program.


Exceptions are Belarus, Turkmenistan, and Uzbekistan.


Other aspects of central bank legislation are discussed in Chapter 7.


Of the 15 countries in the region, only Belarus, Lithuania, Turkmenistan, and Uzbekistan do not currently have programs supported by IMF financial arrangements.


Armenia introduced repurchase agreements on a small scale in 1996, followed by more significant measures in early 1998.


Ratios vary from 8 percent in Latvia and Moldova to 20 percent in the Kyrgyz Republic. Tajikistan introduced uniform reserve requirements in January 1997; Ukraine moved to a uniform reserve requirement of 11 percent in April 1997; and Armenia reintroduced a uniform reserve ratio of 8 percent in August 1997.


This is in line with the IMF’s Monetary and Exchange Affairs Department (MAE) recommendation for situations when the exchange rate is stable, and/or there is limited currency substitution. In the reverse case, it may be preferable to maintain required reserves on foreign currency deposits in foreign currency.


In Armenia, since August 1997 at least 50 percent of reserve requirements on foreign currency deposits has to be met in foreign currency.


In Georgia reserves are not averaged, as reserve balances (up to 90 percent) may be used as collateral for credit auctions. In Russia there is a very restrictive reserve averaging regime, and banks do not use this facility.


Only excess reserves are remunerated in Estonia.


Effective June 1998, uncollateralized short-term credits are no longer issued in Russia.


In practice, foreign currency holdings do not adjust continuously to changes in the macroeconomic environment. As a result, the currency composition of portfolios may fluctuate significantly in periods of sharp temporary movements in the real exchange rate, as the value of the stock of foreign currency assets vis-à-vis domestic currency assets changes accordingly.


In Estonia, household foreign currency deposits were allowed only in 1994.


The standard measure of dollarization (dollar deposits over M3) is inappropriate, since it ignores cash foreign currency holdings while including in the denominator domestic currency holdings, which are relatively important in these countries. In many cases the structure of reserve requirements has to some extent favored dollar intermediation, by stipulating lower ratios on foreign currency deposits.


This may also be because in thin markets reported interest rates often reflect sporadic or diverse operations and transactions, including, for example, market transactions, or open market operations of different sizes or terms.


Estonia and Lithuania have currency board arrangements and thus do not actively manage banking system liquidity. Hence, they are excluded from this classification. In the second quarter of 1997, however, Lithuania issued rules on, and started to conduct, repo transactions to inject liquidity and deposit auctions to mop up liquidity, in anticipation of its forthcoming exit from the currency board arrangement.


Although open market operations are conducted on a frequent basis in many countries (Belarus, Kazakhstan, Kyrgyz Republic, Russia, Ukraine, and Uzbekistan), there seems to be no case in which the central bank relies solely on open market operations to attain its target level of liquidity.


Government securities markets have not developed in Estonia or Latvia because the two governments have not issued domestic debt instruments owing to their strong fiscal positions.