This Selected Issues paper reviews developments during 2001–02 in the Belarusian pension system, and discusses potential short- and medium-term reforms needed to put pension system finances on a sustainable path. The paper presents a short overview of demographic trends and developments in Belarus and their impact on the pension system. It discusses the financial difficulties of early 2002 and the policy measures the authorities took to deal with them. A critical review of the methodology used in preparing annual Social Protection Fund budgets is also presented.


This Selected Issues paper reviews developments during 2001–02 in the Belarusian pension system, and discusses potential short- and medium-term reforms needed to put pension system finances on a sustainable path. The paper presents a short overview of demographic trends and developments in Belarus and their impact on the pension system. It discusses the financial difficulties of early 2002 and the policy measures the authorities took to deal with them. A critical review of the methodology used in preparing annual Social Protection Fund budgets is also presented.

II. Money Demand in Belarus1

Executive Summary

Demand for rubel broad money has improved significantly since mid-2001. Declines in inflation, relative exchange rate stability, high real interest rates, declines in barter transactions, and robust real GDP growth have largely contributed to this outcome.2

The authorities expect further improvements in money demand and envisage significant increases in money supply in their economic program for 2003. To achieve this objective, they have adopted specific measures in programs to increase money demand in 2002 and 2003 (Box 1 and Table 1). It is important to understand these and other changes in money demand in order to formulate an appropriate monetary policy for the period ahead.3 Although this is a preliminary effort to investigate trends in demand for rubel broad money in Belarus, this paper suggests that stable money demand relationships can be identified in Belarus, and that the usual theoretical predictors apply.

Table 1.

Selected Economic Indicators, 2000–03

article image
Sources: Belarusian authorities; and Fund staff estimates.

Baseline projections.

Calculated as GDP divided by average tubel broad money. For the authorities projections, velocity is calculated based on the annual Monetary Guidelines.

Without US$207 million for the sale of the government share in Slavneft and a US$40 million loan from the Russian government, NFA of the NBB would have been US$194 million at end-2002.

A. Background

1. There have been wide swings in rubel broad money velocity during 1995-2002. Velocity doubled from end-1998-—-in the aftermath of the Russian crisis—to mid-2001, and declined only gradually during the relative stabilization since mid-2001. However, despite a significant improvement since mid-2001, velocity at end-2002 had only returned to the level of end-1995, and it remains well above the level of neighboring countries.


Velocity of Rubel Broad Money and Real Rubel Broad Money

Citation: IMF Staff Country Reports 2003, 119; 10.5089/9781451805123.002.A002


Velocity of Domestic Currency Broad Money in Belarus and the Neighboring CIS Countries

Citation: IMF Staff Country Reports 2003, 119; 10.5089/9781451805123.002.A002

2. Financial markets in Belarus are underdeveloped and financial intermediation is weak. The economy remains dominated by the government, including in the banking system, where five of the six major banks are state-owned.4 It is estimated that SO percent of GDP is produced by the public sector, and equity markets are almost non-existent. Accordingly, the chief options for savers are cash (foreign and domestic), bank deposits (again, in foreign and domestic currency), government bonds, and—more recently—precious metals.

The Authorities’ Program to Increase Money Demand

The Belarusian authorities adopted a program to increase money demand in 2002.1 This program, which was extended in 2003, has had only limited influence thus far. The program contains both measures that are likely to prove helpful and steps that could hinder development of market relations in Belarus.

The program aims to increase rubel broad money demand by: (i) increasing real GDP growth; (ii) increasing the share of deposits in the total money stock; (iii) balancing the stock of cash money in circulation with sales of goods, including on a regional basis; (iv) reducing arrears and barter; and (v) developing new asset markets (securities, precious metals, privatization, land reform).

The program comprised specific measures and a timetable to implement them. These measures included: (1) reducing the number of activities subject to licensing requirements; (2) developing measures to reduce the tax burden in Belarus to the level in Russia; (3) saturating markets with domestically produced consumer goods; (4) gradually liberalizing controlled wholesale prices and reducing the number of socially important goods subject to price controls; (5) maintaining a positive real return on deposits; (6) making reserve requirements on new foreign currency deposits in rubels; (7) suspending licenses of those firms engaged in providing services to individuals in foreign currency; (8) developing precious metal markets; (9) increasing efforts to attract deposits; (10) expanding the usage of plastic cards; (11) equipping shops with cash registers; (12) balancing cash supply and sales of goods and services on a regional basis; (13) developing the insurance market; (14) forcing certain enterprises to abstain from barter; (15) reducing “non-monetary” (in-kind) payments for energy; (16) forcing restructuring, reorganization, and bankruptcy procedures for loss-making companies; (17) issuing government securities for individuals; and (18) expanding the privatization of state assets.

Progress has been made on a number of these measures, and many will support macroeconomic stabilization in Belarus. Others will not, however, including the attempt to saturate the market with domestically produced goods, or to match monetary expansion to supply of goods on a regional basis. In the end, there is no substitute for the difficult task of rebuilding confidence in the rubel through a sustained period of low inflation.

1/ Council of Ministers and NBB Resolutions No. 1875/30 of December 28, 2001, and No. 1825/30 of December 30, 2002.

3. Since early-1996, monetary policy has been largely accommodative of the broader policy goals of the government. These goals have included providing subsidized and directed credits to selected sectors of the economy, mainly to agriculture and housing construction. The major sources of monetary expansion have been increases in net credits from the NBB (of which a large part was direct credit to government) and directed lending to priority sectors (“state programs” or “socially important programs”) by the state-owned commercial banks. As a result, Belarus has had the highest inflation rate among the CIS countries over the past five years, a factor that has undermined demand for money and contributed to dollarization of the economy.


Net Domestic Credit of the NBB, Rubel Broad Money, and CPI

Citation: IMF Staff Country Reports 2003, 119; 10.5089/9781451805123.002.A002


CPI, Exchange Rate, and Dollarization

(In percent)

Citation: IMF Staff Country Reports 2003, 119; 10.5089/9781451805123.002.A002

1/ Curb market exchange rate until September 2000, when the exchange rate was liberalizert.2/ Deposits in forign currency as percent of total brand money.

4. The stance of monetary policy was loosened still further following the Russian financial crisis of August 1998.1 The authorities responded to these shocks by easing monetary policy, which caused even more inflation.2 Not surprisingly, the velocity of rubel broad money doubled and the real money stock halved from September 1998 to March 2000.

5. Macroeconomic policies have been tighter since early 2000. At end-1999, the authorities signed a treaty with Russia to form a union state, including establishing a monetary union,3 This factor has contributed to stability insofar as it imposed a timetable for convergence of macroeconomic outcomes in Belarus and Russia. Belarus agreed to unify exchange rates and peg the Belarusian rubel (Rbl) to the Russian ruble (Rub), while the Central Bank of Russia (CBR) agreed to provide Rub 4.5 billion (about US$150 million) in the form of a stabilization loan to Belarus.


Quarterly Changes in CPI in Belarus and in Russia

(In percent)

Citation: IMF Staff Country Reports 2003, 119; 10.5089/9781451805123.002.A002

6. Inflationary pressures have eased since mid-2000 with liberalization of the foreign exchange market. Reducing the spread between the official and curb market rates from 184 percent at end-1999 to 47 percent at mid-2000, and then abolishing the multiple exchange rate system in September 2000. eliminated an implicit tax on exporters, facilitating growth in exports. The resulting inflows of foreign currency helped ease pressure on exchange rates and increased confidence in the currency.


Exchange Rate Spread, Inflation, and Exports

(In percent unless otherwise specified)

Citation: IMF Staff Country Reports 2003, 119; 10.5089/9781451805123.002.A002

7. The authorities adopted a crawling band exchange rate regime in late 2000. The band is identified in terms of the ruble, given the agreement on monetary union with Russia.4 The authorities believe that, in Belarus, the exchange rate is more effective in anchoring inflationary expectations than monetary aggregates, mainly because it is easily understood and widely monitored. At the same time, they recognize that the transmission mechanisms of monetary impulses in Belarus are not well-understood.


Real Refinance Rate and Changes in Velocity

Citation: IMF Staff Country Reports 2003, 119; 10.5089/9781451805123.002.A002

1/ Annual refinance rate divided by four minuts quarterly inflation.

8. Exchange rate stability is also supported by high interest rates. The NBB increased its refinance rate to 175 percent in February 2000 from 90 percent at end-1999 in an effort to make rubel assets more attractive in comparison with those denominated in foreign currency. Since then, the (annualized) real refinance rate has averaged about 15 percent, a factor that has contributed to the increase in money demand.5 In particular, high interest rates have contributed to an increase in deposits and have reduced the share of cash in broad money. In addition, high interest rates may have also helped to curb demand for credits.6

9. The authorities’ efforts at macroeconomic stabilization were bolstered by a Staff-Monitored program (SMP) in 2001 and a monetary program agreed with the CBR in 2002. Both programs included quantitative targets on NDA, NIR and an indicative target on reserve money. The CBR program was supported by financial resources—three equal tranches of Rub 1.5 billion with one year maturity.7 Both the SMP and the CBR program ended on September 30, and were followed by a marked loosening of—monetary and thus higher inflation policy—in the fourth quarter.

10. Interenterprise arrears and barter remain endemic, despite the authorities’ efforts to reduce them.8 While the share of barter transactions has been reduced significantly in international trade—given pressure to limit it from Russia and other CIS countries—the share of barter in domestic transactions remains high.


Barter Transactions and Arrears

Citation: IMF Staff Country Reports 2003, 119; 10.5089/9781451805123.002.A002

B. Modeling Money Demand in Belarus

11. Mainstream theories of money demand relate the optimal stock of real money balances to returns on competing assets (negatively) and to real income (positively).9 Money demand in Belarus is considered in Krivorotov (2000). He argues that money demand in Belarus is a function of real output, expected inflation, returns on instruments in domestic currency, the relative yield on foreign currency assets and changes in quasi-money (accounts receivable, barter, etc.). However, the impact of these variables is not measured.

12. A standard model of money demand is developed in this paper, which relates money demand to real transactions and a set of variables capturing the opportunity cost of holding money:


where mp stands for real money, y is a scale variable capturing real economic activity, and R is a vector of variables capturing opportunity costs of holding money.

Data and methodology

13. We use ruble broad money (M2) for money balances. Mainly because of data constraints, we use real GDP (rGDP) and real wages (RW) to capture real economic activity. To capture opportunity costs, we use changes in CPI (∆CPI), changes in exchange rates (∆EXR), the real refinance rate (RR), and the real deposit rate (DR).10 (Due to lack of data, barter and arrears are not included as explanatory variables.) All variables except real interest rates are in logs.

14. Data are monthly from January 1996 to September 2002, a period of relative stability, although there are some breaks in the time series. Dummy variables are used to capture seasonality, as well as three significant shocks:

  • (September 1998) for the Russian crisis;

  • (December 1998) for the currency reform; and

  • (October 2000) for exchange rate unification.

15. Calculations are done using the Johansen procedure, in three steps (using three lags).11 First, we conduct unit root tests. Second, we do a vector autoregression (VAR) analysis to find long-term relations among the variables. Finally, in a single equation, changes in real money balances are regressed on contemporaneous changes in the explanatory variables, lags of changes in real money balances and the explaining variables, and the first lag of cointegrating vectors from the VAR.


16. The results of unit root tests show that all variables are integrated of order (1) (Table 2). As the variables are non-stationary, we purge the non-stationarity by differencing, and estimate only differenced variables:

Table 2.

Augmented Dickey-Fuller (ADF) Statistics for Testing a Unit Root 1/2/

(Sample period January 1996-September 2002)

article image

Using seasonality dummy variables.

* indicates rejection of null hypothesis at 5 percent level and ** at 1 percent level.

Including a constant.

indicates rejection of null hypothesis at 5 percent level

at 1 percent level.


The results suggest that only growth and changes in inflation significantly explain changes in real money balances (Table 3).12 Real money balances are inversely related to inflation and positively related to growth.13 However, valuable information concerning the long-run equilibrium properties of the underlying process is lost in the above equation. Therefore, a VAR analysis is conducted.

Table 3.

Coefficients in Growth Rate Model 1/

(Changes in variables)

article image

The equation also includes seasonality dummy variables

AR 1-5 test for equation (1): F(5.59)=1.92

AR 1-5 test for equation (2): F(5.59)=2.74*

17. The results of the VAR analysis suggest that there is a cointegration relationship in levels among real rubel broad money, the real refinance rate, and real GDP. In this relationship, real money balances are inversely related to the real refinance rate and positively related to real GDP (Table 4).

Table 4.

Long-Run Cointegrating Relation

article image

18. The single equation analysis suggests that changes in real money balances have inertia and are inversely affected by increases in inflation. While changes in real deposit rates increase money demand, changes in the refinancing rate reduce it. Feedback from the error-correction term (ECT)—which represents deviations from the long-term relationship between real money, real refinance rate, and real GDP—has a minus sign as expected (Table 5). But the relationship is stronger than would be expected, suggesting that there are omitted variables (such as arrears or barter) in the equations.14 Moreover, Chow-tests indicate that there may be a structural break in the equation in November 2000, probably related to the liberalization of the foreign exchange market in the previous month. Therefore, the findings of the single-equation analysis should be interpreted with caution.

Table 5.

Coefficients of Error-Correction Equations 1/2/

article image

The equation also includes sasonality dummy variables.

ECT stands for the error-correction term.

C. Summary

19. While preliminary, the results above suggest that it is possible to identify a stable demand for money relationship in Belarus. The standard predictions generally seem to apply, in that demand for money falls with inflation and rises with output and own interest rates. However, considerable work remains to be done in this area, given data deficiencies.


  • Banerji, Angana, 2002, “Money Demand,” in Russian Federation: Selected Issues (SM/02/63, Supp. 1, 02/22/02), pp. 1537.

  • Bogetic, Zeljko, and Zorica Mladenovic, 2002, “Inflation and the Monetary Transmission Mechanism in Belarus, 1996-2001,unpublished draft (December).

    • Search Google Scholar
    • Export Citation
  • Johansen, Soren, 1988, “Statistical Analysis of Cointegrating Vectors,Journal of Economic Dynamics and Control, Vol. 12, Issues 2-3, pp. 23154.

    • Search Google Scholar
    • Export Citation
  • Johansen, Soren, 1992, “Testing Weak Exogenity and the Order of Cointegration in UK Money Demand Data,Journal of Policy Modeling, Vol. 14, Issue 3, pp. 31334.

    • Search Google Scholar
    • Export Citation
  • Johansen, Soren, and Katarina Jusehus, 1990, “The Full Information Maximum Likelihood Procedure for Inference on Cointegration-with Applications to the Demand for Money,Oxford Bulletin of Economics and Statistics, Vol. 52, pp. 169210.

    • Search Google Scholar
    • Export Citation
  • Johansen, Soren, 1992, “Testing Structural Hypotheses in a Multivariate Cointegration Analysis of the PPP and UIP for UK”, Journal of Econometrics, Vol. 53, pp. 21144.

    • Search Google Scholar
    • Export Citation
  • Johansen, Soren, 1994, “Identification of the Long-run and the Short-run Structure—An Application to the ISLM Model,Journal of’ Econometrics, Vol. 63, pp. 736.

    • Search Google Scholar
    • Export Citation
  • Krivorotov, D. V., 2000, “Theoretical and Methodological Aspects of Evaluating Demand for Money Under Conditions of the Belarusian Economy” (mimeo).

    • Search Google Scholar
    • Export Citation
  • Sriram, Subramanian S., 1999, “Survey of Literature on Demand for Money: Theoretical and Empirical Work with Special Reference for Error-Correction Models,IMF Working Paper No. 99/64 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

Prepared by Etibar Jafarov.


However, owing to methodological problems, official data on real GDP growth are thought to be overestimated. (See Appendix III of accompanying staff report.)


It is especially important given the authorities comrnitment to fix the exchange rate vis-a-vis the Russian ruble from January 1, 2004. Higher inflation would lead to real appreciation of the currency, which in the absence of compensating improvements in productivity would hurt the balance of payments.


An exception is Priorbank, which sold fifty percent of its shares to Raiffeisen AG of Austria in January 2003.


Poor weather conditions also played a role.


The authorities reduced the required reserve ratio from 18 percent to 16 percent in August 1998 and increased directed credits to agriculture and housing.


See companion Selected Issues paper.


De facto, the authorities have been targeting the U.S. dollar.


Nevertheless, in late 2001 and late 2002, real interest rates were again allowed to move into negative territory as inflation picked up.


Though lack of hard budget constraints on state-owned enterprises make interest rates a relatively ineffective means of rationing credit in an economy where market oriented reforms are lagging.


These tranches were disbursed in July 2001, June 2002, and November 2002. The first tranche was rolled over for another year in July 2002.


Tighter macroeconomic policies and soft budget constraints have contributed to increases in money surrogates that tend to undermine demand for money.


For review of the literature on money demand, see Sriram (1999).


The curb market rate is used prior to exchange rate unification in September 2000.


This is an error-correction model. For details, see Johansen (1988 and 1992), Johansen and Juselius (1990, 1992, and 1994).


Changes in real wages, in the exchange rate, real deposit and refinance rates are excluded from the equations based on F-tests. As growth is not significant in equation (1), it is excluded from equation (2). But in mat case, we find an autoregression in the second equation,


But see also Bogetic and Mladenovic (2002), which finds different results.


Banerji (2002) shows that declines in barter contributed to increases in money demand in Russia.

Republic of Belarus: Selected Issues
Author: International Monetary Fund