Chapter

IV Transition to Indirect Instruments: Selected Experiences

Author(s):
Tomás Baliño, Charles Enoch, and William Alexander
Published Date:
July 1995
Share
  • ShareShare
Show Summary Details

This section analyzes the experience of selected countries in making the transition from direct to indirect instruments. The sample consists of industrial and nonindustrial countries chosen to be broadly representative of the membership’s experience in introducing indirect instruments in recent years.

Their experience suggests that the introduction of indirect monetary policy instruments is most effective and most smoothly accomplished under conditions of a stable macroeconomic environment and sound fiscal policies. In addition, it is crucial to develop a sound and competitive financial system and an adequate supervisory framework. The experience of the industrial countries shows that deep financial markets may take a long time to develop—a consideration that nonindustrial countries should keep well in mind. The transition in most industrial countries has been gradual and fairly smooth. In contrast, some other countries started with large disequilibria and distortions, making rapid and substantial changes necessary. During the transition process in the latter group, it was essential to establish proper incentives and conditions for financial market development as well as guidelines to assess and ensure the soundness, competitiveness, and efficiency of the banking sector.

Experience of Industrial Countries

This review of industrial country experience is brief because most of these countries have been analyzed in detail elsewhere.25 These experiences are best reviewed separately from those of developing countries and economies in transition, owing to the wide difference in the initial conditions of countries in each group.

Context of the Transition

Most industrial countries abandoned direct controls and began to rely exclusively on indirect instruments of monetary policy between the early 1970s and late 1980s. This was part of a broader process of financial deregulation, liberalization, and innovation. Interest rate and credit controls were eliminated, and new financial instruments and institutions were allowed to develop. In addition, international financial transactions were progressively liberalized. Full reliance on indirect instruments was not only a natural response of the central banks to these developments but was also a necessity in light of the growing ineffectiveness of direct instruments as financial markets became increasingly integrated and economic agents had more opportunities to circumvent controls.

Brief Overview

Canada abandoned direct controls in 1967. The United States, too, has a long tradition of relying mainly on indirect instruments, although some direct credit controls were used as late as 1980.

In most Western European countries, the movement toward full reliance on indirect instruments for the conduct of monetary policy has been gradual. France and the United Kingdom made significant moves toward their use in the late 1960s and early 1970s. However, both countries reverted for a period to some form of direct controls. Greater integration of the economies of the countries that now form the European Union and, in particular, the pressure exerted by financial integration in the second half of the 1980s increased the pace of the transition. It also induced a convergence in the techniques used to implement monetary policy.26 In Germany, open market operations became the Bundesbank’s principal vehicle for short-term reserve management in the early 1980s, thereby reducing earlier emphasis on rediscount operations and minimum reserve requirements. Belgium, Denmark, Italy, the Netherlands, and Spain adopted indirect methods of monetary control in the late 1980s and early 1990s, as did Finland, Norway, and Sweden. A similar process is now under way in Greece, Ireland, and Portugal, with a view toward monetary integration in Europe.

Although Japan has never used credit ceilings, it has used other direct measures, the most important being interest rate controls and moral suasion (window guidance). Its transition to indirect instruments was part of a broader liberalization started in the second half of the 1970s, which is almost complete. In particular, dependence on moral suasion has been significantly reduced since the early 1980s, and window guidance was formally abolished in 1991. While interest rates on bank loans have been free since the 1940s, the liberalization of interest rates on deposits started in the second half of the 1970s and has proceeded steadily. New Zealand, on the contrary, went through a successful “big bang” experience in 1985-86.27

Problems with Transition

The comparatively smooth transition in most industrial countries can be explained largely by the fact that (1) financial markets were already well developed when the central banks increased their reliance on indirect instruments; (2) the macroeconomic situation was generally not far out of balance; and (3) interest rates, although often controlled, were usually close to market-clearing levels and did not cause great distortions.

Despite these relatively favorable conditions, two countries, France and the United Kingdom, suffered setbacks in their first attempts to liberalize the financial system and adopt indirect instruments of monetary policy. In the United Kingdom, the Competition and Credit Control Act of 1971 deregulated interest rates, abolished credit controls, and permitted commercial banks and building societies to compete in each other’s markets. Henceforth, monetary policy was to rely entirely on indirect mechanisms. However, the introduction coincided with a significant deterioration in the fiscal stance, which compromised the ability of the Bank of England to contain the subsequent credit growth. In November 1973, the Supplementary Special Deposit Scheme (the “corset”) was introduced, imposing high marginal reserve requirements on financial institutions. Under the scheme, banks were required to place additional deposits at the Bank of England, in increasingly penal proportions, as their eligible liabilities rose above specified baseline levels. The scheme was abolished in February 1975 but reintroduced in 1976-77 and 1978-80. However, the abolition of exchange controls in 1979 provided additional scope for disintermediation from the domestic banking sector. As a result, with deposits increasingly being made at institutions overseas, the scheme was seen to be ineffective and shortly thereafter it was finally abolished.

In France, the 1967–69 liberalization attempt was designed to ensure that interest rate management would become the central bank’s main policy instrument. However, there were inconsistencies in the use of monetary instruments, and monetary policy suffered from a lack of well-defined policy objectives. To facilitate domestic monetary management at times when interest rate policy was focused on the external sector, the Bank of France intended to use reserve requirements flexibly. Unlimited access to rediscount facilities made increases in those requirements substantially ineffective, however, as they were automatically offset by additional refinancing with the central bank. Reserve requirements on liabilities affected only a small part of the financial sector because several large banks were funded through the government’s budget.

The segmentation and heterogeneity of the French financial system also significantly constrained the central bank’s ability to conduct an interest rate policy and, hence, hampered the transmission of monetary policy impulses. The financial sector’s permanent indebtedness to the central bank and the uneven distribution of that debt made the authorities unwilling to set the discount rate at a penalty level.

The large banks never used their refinancing ceilings in full and aligned their lending rates to the rediscount rate. By contrast, the other banks, with permanent liquidity shortages, were mainly price takers. They were usually at their refinancing limits with the central bank and were borrowers in the interbank market. Hence, an increase in the money market rate made many of these banks unprofitable. This vulnerability was a major concern for the Bank of France and impeded effective interest rate management. Hence, in times of conflict between internal and external goals, the authorities reverted to direct monetary controls and tightened exchange controls. Thus, France’s first liberalization lasted less than two years (see Quintyn, 1993). The Bank of France reimposed bank-by-bank credit ceilings in 1969, which were abolished only in 1985.

These two episodes highlight the importance of avoiding conflicts between fiscal and monetary policy objectives and, particularly in the case of France, the importance of a well-functioning money or interbank market that can rapidly transmit the effects of changes in instrument settings throughout the banking sector. The two countries’ experience also shows the need for healthy and competitive financial institutions and underscores the importance of using the appropriate instrument mix and approaching the introduction of indirect instruments and financial sector liberalization in a comprehensive way. The U.K. example also points to the catalytic effect of opening up the capital account, since such opening undermined the effectiveness of direct instruments, leaving no alternative to indirect methods of monetary control.

Experience of Selected Developing and Transition Economies

This section concentrates on the experience of nonindustrial countries. To keep the scope of the exercise tractable, a maximum of five countries from each region were selected for study, and only countries that were well advanced in the transition to indirect instruments were included. However, the final sample contains countries that made the transition relatively easily as well as countries that encountered significant difficulties. The sample comprises five countries from Latin America and the Caribbean (Argentina, Chile, Jamaica, Mexico, and Venezuela),28 two from Europe (Hungary and Poland), five from Asia (Indonesia, Malaysia, Philippines, Sri Lanka, and Thailand), four from Africa (Burundi, The Gambia, Ghana, and Kenya), and three from the Middle East (Egypt, Israel, and Tunisia).

Analytical Focus

Nonindustrial countries have faced a wider variety of problems in their transition to indirect instruments than industrial countries, mainly because certain institutional and economic conditions were often lacking. To better understand the complexities in implementing indirect instruments, the subsequent analysis focuses on (1) the initial institutional and macroeconomic conditions in nonindustrial countries. (2) the implementation experience, (3) the characteristics of the monetary policy instruments, and (4) financial sector efficiency and monetary control.

It is difficult to date unambiguously the start of the transition to indirect instruments in particular countries. For analytical purposes, it has been assumed in this paper that the transition started when the central bank began to auction treasury bills or central bank bills. Similarly, it is difficult to be precise about the duration of the transition, which is needed to compare a country’s performance before and after the transition. Therefore, the length of the transition was determined case by case, with the endpoint marked by full reliance on indirect monetary instruments. The latter is defined as the point at which interest controls have been eliminated and the directed credits from the central bank amount to no more than 25 percent of total credit.

To varying degrees, all 19 countries in the sample have experienced higher savings mobilization, increased financial intermediation, a move toward positive real interest rates, and, thus, financial deepening following the transition process.29 However, there were substantial differences in performance (chiefly reflecting each country’s willingness to use the new instruments fully), its timing of the reforms, its initial macroeconomic conditions and policies, as well as its institutional and structural characteristics.

Institutional and Macroeconomic Factors at the Outset of Reform

Although the experience of the sample countries during the transition to indirect instruments has been diverse, certain initial conditions were common to most of them. The introduction of indirect instruments was part of a broader set of reforms, which included not only liberalization of the financial sector but also macroeconomic stabilization and liberalization of the economy in general. Most of the countries had an IMF-supported program at the time. Measures to open up the financial sector to new entrants and to allow banks more operational freedom were common to all cases. Table 4 summarizes some of the salient institutional and macroeconomic characteristics at the outset of the reforms. The data on institutional characteristics are mainly derived from information gathered by IMF staff during its regular annual consultation with each of its members. Some involve a substantial element of judgment. (Additional details for some of the countries are included in the case studies in Part II of this paper.)

Table 4.Initial Conditions in Selected Countries in Transition to the Use of Indirect Instruments
ConditionPercent of total 19 countries
Initial institutional conditions
IMF-supported program89
Public ownership of banking sector greater than 50 percent63
Effective supervision32
Independent central bank11
Attainment of monetary policy targets63
Existence of capital controls95
Initial macroeconomic conditions
Annual rate of inflation greater than 20 percent47
Negative real interest rates53
Ratio of fiscal deficit to GDP greater than 5 percent58
Excess liquidity79
Source: IMF staff estimates and International Financial Statistics. various issues.Note: The countries are Argentina, Burundi, Chile, Egypt. The Gambia, Ghana, Hungary, Indonesia, Israel, Jamaica, Kenya, Malaysia, Mexico, the Philippines, Poland, Sri Lanka, Thailand. Tunisia, and Venezuela.
Source: IMF staff estimates and International Financial Statistics. various issues.Note: The countries are Argentina, Burundi, Chile, Egypt. The Gambia, Ghana, Hungary, Indonesia, Israel, Jamaica, Kenya, Malaysia, Mexico, the Philippines, Poland, Sri Lanka, Thailand. Tunisia, and Venezuela.

About 60 percent of the sample was characterized by a banking system owned mainly by the public sector. This is important because, as discussed earlier, the monetary policy transmission mechanism depends on the soundness and competitiveness of the banking system.30 Further common features were weak and segmented money and interbank markets, lack of effective bank supervision, and a low level of central bank autonomy. The latter was important in that it made many central banks vulnerable to pressures to accommodate the government’s fiscal needs. In only two-thirds of the sample countries were the authorities able to enforce direct controls effectively and attain their monetary objectives.

About half the countries in the sample had large macroeconomic imbalances at the start of the reform process characterized by, among other factors, inflation rates well over 20 percent (one-fifth of the sample had triple-digit inflation rates), negative real interest rates, and high fiscal deficits (exceeding 5 percent of GDP). As a consequence, indirect instruments were adopted under less than ideal conditions, which required that they be part of a more comprehensive reform and stabilization program. Thus, the success of the introduction of indirect instruments and that of the stabilization program were closely linked.

As noted earlier, excess liquidity, as reflected in levels of bank reserves that greatly exceed both statutory requirements and banks’ normal precautionary requirements, is a common characteristic of monetary frameworks with direct controls. Eighty percent of the sample countries were coping with excess liquidity in the financial system at the time of the reform process. This posed a special challenge to the authorities, since market-based indirect instruments might not be developed sufficiently early in the reform to absorb large amounts of excess liquidity.

Implementation Experience

There were significant differences across countries with respect to the measures taken during the reform process as well as to the general pace of reforms. Nor was the experience in all cases smooth; about half of the countries in the sample experienced temporary reversals of reforms. Table 5 summarizes the key features of the transition.

Table 5.Features of the Transition to the Use of Indirect Instruments in Selected Developing Countries
FeaturePercent of total 19 countries
Gradual58
Reversal of monetary reforms47
Fiscal consolidation68
Fixed exchange rate regime53
Abolition of capital controls158
Source: IMF staff estimates.Note: Sample as noted in Table 4.

Includes partial liberalization efforts.

Source: IMF staff estimates.Note: Sample as noted in Table 4.

Includes partial liberalization efforts.

The pace of transition to a market-oriented system of indirect instruments has varied across the nonindustrial countries. Argentina, Chile, and Israel adopted indirect instruments within a year, while the rest took a more gradual approach, and many took three or more years to complete the transition. The latter continued to employ direct instruments, including credit or interest rate controls, directed credits, and subsidized central bank credit, after the introduction of indirect instruments. This was similar to the approach followed by many industrial countries, as discussed earlier. In addition, some central banks sought to complement indirect instruments with informal means, such as moral suasion (for example, Indonesia, Malaysia, Thailand, and Venezuela). Mexico reversed the direction of its policies for a substantial period during the transition—at the time of the debt crisis—and only started to rely fully on indirect monetary instruments six years after their introduction.

In all sample countries, monetary policy reforms were part of a broader financial reform package, including improvements in bank supervision, revisions of the legal framework, and reorganization of the banking system. In addition, in about 70 percent of the countries, efforts were undertaken to contain excessive fiscal imbalances.31

Central banks in all the sample countries had to improve control over credit expansion and in many cases absorb excess reserves. For instance, Bank Indonesia drastically curtailed the scope of its subsidized rediscount (“liquidity credits”) as a first step of its financial reforms in 1983. Most of the countries also had to eliminate or tighten access to facilities that allowed banks to borrow automatically from the central banks. In addition, excepting Indonesia, all countries limited central bank financing of the government to some extent.32

Most or all interest rates were liberalized early in the transition to provide adequate scope for market-based instruments. However, only 35 percent of sample countries abolished bank-by-bank credit ceilings at the beginning of the liberalization process, while the rest abandoned them later in the process, after they had gained experience with the use of indirect instruments.

Introducing indirect instruments often required supporting measures that varied depending on each country’s situation. For example, Poland’s shift to indirect instruments was impeded by a slow and unreliable paper-based payments system that was geared to a centrally planned economy, which the authorities had to modernize. Thus, in designing their monetary instruments, the Polish authorities had to take into account modernization of the payments system that was simultaneously taking place (see Baliño, Dhawan, and Sundararajan, 1994). Conversely, the development of monetary instruments required appropriate changes in the payments system.

Other reforms also had to be undertaken. Several countries, including Egypt, Ghana, Israel, Mexico, and Thailand, began to restructure their commercial banks, many of which had accumulated large non-performing loan portfolios. Banking supervision and regulation had to be strengthened in most of the sample countries to avoid excessive risk taking on the part of banks and to prevent systemic problems. Moreover, increasing competition was an object of reform in all the sample countries. Therefore, reducing barriers to entry and eliminating regulations that restricted competition were key elements of the reforms. Thus, enhanced banking competition and the introduction of indirect instruments reinforced each other. Also, some countries—such as those in Central and Eastern Europe—had to make special efforts to gather and analyze economic information that would shed light on key monetary relationships, knowledge of which was needed to operate indirect monetary instruments successfully.

The need for a comprehensive approach to the introduction of indirect instruments is particularly strong in the case of economies in transition (Sundararajan, 1991). Such introduction can only take place once a two-tier banking system has replaced the monobank system in a way that clearly separates money from reserve money and vests in the monetary authority sole responsibility, as well as sufficient technical and institutional capabilities, for managing the latter. This transformation has required a comprehensive approach to strengthening a wide range of central banking functions and financial structures simultaneously, so that lack of progress in one area would not impede progress in another. Also in those countries, difficulties in establishing and identifying stable relationships between reserve money and broader monetary aggregates have often required cautious strategies for the shift to indirect instruments. These strategies often provide for the temporary retention of credit ceilings while indirect instruments are established and other key functions, such as the payments system, are improved.33 Thus, the reforms that those countries have been undertaking typically need to cover a broader range of issues than those in market economies.

The process of reform has often included temporary reversals. The cases of Argentina. Burundi, Chile, Jamaica, the Philippines, and Venezuela illustrate that point. All of these countries except Burundi and Jamaica faced a serious financial crisis, which prompted the temporary reintroduction of controls on interest rates to alleviate the burden of high real interest rates on borrowers and banks. Reversals in Burundi and Jamaica were a direct response to excessively high fiscal imbalances, which the authorities were reluctant to finance at market rates of interest. Thus, inadequate banking supervision and failure to sustain adequate macroeconomic policy were key factors in the reversal of financial reform.

The overall experience of the sample countries suggests that a successful shift to indirect instruments does not depend on the choice of exchange rate regime. However, that choice could influence the desirable speed of transition to indirect instruments, since fixed exchange rate regimes typically require greater interest rate flexibility. The shift to indirect instruments does not require the removal of capital controls. In most cases, there was a revealed preference to phase in capital account liberalization only gradually as the authorities became confident that they had the necessary instruments to manage these flows. Capital account liberalization often had two main effects on the money market. On the one hand, it stimulated large capital inflows, which the monetary authorities had to manage. On the other, those inflows contributed to the development of the money and capital markets, for example, by fostering competition from foreign banks and other financial institutions. They also made direct monetary controls ineffective and added impetus to the shift to indirect instruments. Moreover, the deepening of financial markets allowed central banks to use more sophisticated instruments (such as in Indonesia and Mexico). This is consistent with the experience of industrial countries discussed earlier.

Use of Monetary Policy Instruments

Table 6 summarizes the situation with regard to monetary instruments at the end of the transition period. All sample countries started with open market type operations in the primary market owing to the absence of well-developed secondary markets for government securities or central bank paper. During the transition, countries experimented with different types of financial instruments, varying maturities, and different auction frequencies.

Table 6.Use of Monetary Instruments in Selected Countries at the End of the Transition Period
Use of instrumentPercent of total 19 countries
Open market operations
Market
Primary markets50
Predominantly primary markets130
Predominantly secondary markets20
Instrument
Treasury bills only53
Central bank bills only16
Both31
Maturity2
1-3 months63
3-6 months58
More than 6 months21
Weekly auctions95
Type of auction
Uniform price74
Multiple price26
Retains discretion to modify auction outcome63
Reserve requirements
Less than 15 percent58
Remunerated32
Rediscount facilities90
Access limited79
Source: IMF staff estimates.Note: Sample as noted in Table 4.

The central bank basically intervenes through the issue market but may sporadically also do repos or reverse repos with the banks or primary dealers.

As some countries issue different maturities simultaneously, the percentages do not add up to 100.

Source: IMF staff estimates.Note: Sample as noted in Table 4.

The central bank basically intervenes through the issue market but may sporadically also do repos or reverse repos with the banks or primary dealers.

As some countries issue different maturities simultaneously, the percentages do not add up to 100.

The survey indicates a preference for the use of treasury securities as the underlying instrument, versus central bank securities, although about a third of the countries used both instruments simultaneously. Most countries chose maturities of not more than three months. Almost all countries eventually adopted a system of weekly auctions, often after experimenting with longer intervals. A majority opted (also after experimentation) for a uniform price auction. In about 65 percent of the cases, the authorities retained some discretion to modify the auction outcome, typically through the right to reject bids even if the full amount of securities at auction would not be fully allocated at the cutoff rates. Generally, this right is exercised to reject outlying bids or discourage collusion. All countries except Mexico maintained reserve requirements, albeit at different levels. Half of them kept their reserve requirements higher than 15 percent; only in about one-third of the sample did the central bank remunerate required reserves. Reserve requirements therefore imposed a considerable tax on financial intermediation and at the same time provided the authorities with a cheap source of finance.34

In most sample countries, many of the building blocks necessary for an active and efficient money market were in place by the end of the transition period. The necessary instruments, including prime-quality assets like treasury bills and bonds, central bank bills and bonds, banker’s acceptances, and certificates of deposit, had been created. Most countries had also developed legal and technical procedures for clearing and settlement. But despite all these developments, secondary markets remained weak. Thus, as shown in Table 6, the central bank’s money market interventions were still mainly conducted through the primary or issue markets in about three-fourths of the sample. In about one-half of the countries, these open market type operations were still the dominant technique, but some central banks had also started to supplement them with interventions in the secondary markets (mainly repos or reverse repos). Only 20 percent of the central banks relied entirely on open market operations through secondary markets.

In addition to the introduction of open market type operations, most countries in the sample kept a rediscount facility in place. However, in order to improve monetary control, many countries restricted access by setting their rediscount rate at a margin above the market rate; others put limits on access to this facility.

Financial Sector Efficiency and Monetary Control

This section examines the impact of the use of indirect monetary instruments on efficiency in the financial sector and on the degree of monetary control by considering the level and the volatility of two sets of key financial variables before and after the transition to indirect monetary instruments.35 The first set includes proxies for the performance of the banking system; the second set includes variables that are used as proxies for the degree of monetary control.36

Financial Sector Efficiency

To assess the implications of introducing indirect monetary instruments, the levels of the following monthly data in the pre- and posttransition periods were examined: interest rate spreads (difference between lending and deposit rates), the share of banks in the overall deposit and credit market for all financial institutions (bank and nonbank), and the share of credit to the private sector in total domestic credit. Standard F-tests were utilized to test for significant shifts in the means of the above variables in the periods preceding and following the transition to indirect monetary instruments.37Table 7 provides the information on the sample countries and the results of the analysis.

Table 7.Trends in Financial Variables Between Pre- and Posttransition Periods1
Egypt2The GambiaGhanaIndonesiaIsraelJamaicaKenya2MalaysiaMexicoPhilippinesSri LankaThailandTunisiaVenezuela
Interest spreads
Mean06.0011.838.106.00140.203.396.532.7112.218.265.825.722.155.44
Mean15.0013.535.2011.88325.45.085.213.065.656.363.936.522.714.19
Mean14.0013.263.403.0727.1010.304.921.404.104.502.742.262.582.51
F-test4.01*11.23*8.47*19.20*5.51*4.75*8.01*7.63*5.65*10.28*8.40*
Deposit market share of banks
Mean00.970.990.820.970.900.4O0.680.890.780.970.920.68
Mean20.991.000.981.000.800.440.820.950.920.921.000.74
F-test0.960.822.146.29*2.353.90*1.081.230.95
Loon market share of banks
Mean00.830.850.850.620.800.760.54
Mean20.870.920.700.680.910.810.68
F-test2.94*7.62*1.043.87*2.005.30*
Share of credit to private sector in domestic credit
Mean00.260.380.210.940.420.340.520.920.460.760.490.670.830.91
Mean20.281.450.251.190.451.300.740.880.760.850.580.960.890.80
F-test9.03*2.115.34*2.728.04*3.14*3.78*7.24*4.51*6.83*1.205.87*
Sources: IMF staff estimates and International Financial Statistics, various issues.

Mean0, mean1, and mean2 are the sample means for the pretransition, transition, and posttransition periods, respectively. The F-test (Chow test) is the test statistic for the stability of the regression coefficients between the pretransition (mean0) and posttransition periods (mean2). The autoregressive regressions are of the form xt = βxt-1, where x is the financial aggregate variable. The F-tests that are significant at the 1 percent confidence interval are indicated by an asterisk.

Limited data are available for the posttransition period.

Sources: IMF staff estimates and International Financial Statistics, various issues.

Mean0, mean1, and mean2 are the sample means for the pretransition, transition, and posttransition periods, respectively. The F-test (Chow test) is the test statistic for the stability of the regression coefficients between the pretransition (mean0) and posttransition periods (mean2). The autoregressive regressions are of the form xt = βxt-1, where x is the financial aggregate variable. The F-tests that are significant at the 1 percent confidence interval are indicated by an asterisk.

Limited data are available for the posttransition period.

Interest rate spreads can be used as a proxy for the efficiency of financial intermediation and the level of competition in the financial sector.38 The experience of most of the sample countries indicates that the interest spreads of banks initially widened during the transition and then narrowed significantly in the posttransition period, suggesting increased efficiency of services in the financial sector. In all but three countries (The Gambia, Jamaica, and Tunisia), spreads fell between the pre- and posttransition periods.39 The results of the F-tests indicate that those changes were statistically significant.

An analysis of the deposit market shares indicates a significant increase in deposit mobilization by the banking sector relative to nonbank financial intermediaries for all countries. Loan market shares show similar trends.40 The ratio of loans to deposits has also increased in most of the sample countries, suggesting increased intermediation as a result of the elimination of credit ceilings. Available data for some sample countries indicate that this trend was accompanied by a decline in excess reserves, which is also consistent with improved financial intermediation and a better functioning payments system.

The share of total credit extended to the private sector has increased in most countries. In some cases (The Gambia. Indonesia, and Jamaica), this ratio is greater than unity, indicating the repayment of outstanding credit by the public sector.

Finally, after the transition, all countries in the sample had moved toward or maintained positive real lending rates.41 This suggests a more efficient allocation of credit following the removal of controls.

Monetary Control

To assess the impact on monetary control of the introduction of indirect monetary instruments, the level and volatility of the following monthly data series were examined: money multiplier (M2/M0), ratio of narrow money to broad money (M1/M2), and short-term interest rates (money market or treasury bill rates). Volatility is measured as the coefficient of variation (the mean-weighted sample standard deviation) in the periods before, during, and after the transition to indirect monetary instruments. Standard F-tests were used to test for significant shifts in the standard deviations before, during, and after the transition. Table 8 provides the results of the analysis.

Table 8.Trends in Monetary Indicators1
Egypt2The GambiaGhanaIndonesiaIsraelJamaicaKenya2MalaysiaMexicoPhilippinesSri LankaThailandTunisiaVenezuela
Multiplier (M2/M0)
Mean02.902.461.672.632.265.010.463.681.783.423.296.504.494.30
Mear03.142.572.203.562.242.940.474.653.263.372.857.784.874.26
Mean23.252.642.525.442.902.620.484.746.123.492.758.404.463.21
F-test023.05*2.835.22*3.42*4.71*6.63*8.24*3.50*4.25*5.02*2.276.34*
Sdev00.030.110.020.030.020.130.010.020.030.020.020.050.050.05
Sdev10.100.250.140.170.090.050.040.050.260.290.050.080.080.11
Sdev20.020.060.050.050.040.130.020.020.060.120.030.040.030.06
F-test015.31*2.186.17*4.08*3.45*5.18*4.19*3.12*2.173.71*4.66*2.94*
F-test127.56*1.477.81*3.17*5.31*3.04*2.91*2.94*3.67*4.47*3.52*2.50
M1/M2
Mean00.360.580.770.630.050.290.690.410.330.400.370.180.560.51
Mean10.290.620.740.420.030.310.660.350.280.230.420.140.490.38
Mean20.260.590.650.330.080.380.720.310.400.270.380.110.450.33
F-test022.004.52*7.20*5.74*3.90*2.97*3.27*4.83*1.853.40*2.613.12*
Sdev00.020.060.020.010.080.040.010.060.040.030.030.010.010.04
Sdev10.030.080.040.090.110.090.020.020.160.250.020.060.030.06
Sdev20.010.020.010.040.070.030.010.060.050.040.020.030.010.06
F-test014.72*1.622.96*8.45*7.83*5.34*18.55*5.28*1.225.59*1.715.94*
F-test123.99*2.337.84*6.52*7.12*3.99*14.96*6.11*1.915.02*1.472.08
Short-term interest rates3
Mean020.413.16114.9811 4213.863.5932.9912.2512.3713.069.82
Mean118.525.59.47210.1119.0316.596.7166.0218.4712.429.2911.31
Mean218.029.313.9018.250.3849.794.7421.3818.9916.397.1011.72
Sdevo00.030.260.420.080.010.320.120.020.050.130.01
Sdev10.820.020.730.610.120.040.210.240.210.090.170.06
Sdev20.650.090.140.200.180.380.100.180.120.050.230.01
F-test012.815.40*10.97*39.61*10.01*5.80*19.31*3.12*9.01*5.82*
F-teat023.28*6.10*5.22*16.30*52.97*14.30*4.71*27.24*4.27*4.21*3.68*
Sources: IMF staff estimates and International Financial Statistics, various issues.

Mean0, mean1, and mean2 are the sample means for the pretransition, transition, and posttransition periods. Sdev0, sdev1, and sdev2 are the mean-weighted sample standard deviations for the pretransition, transition, and posttransition periods, respectively. The F-test02 (Chow test) is the test statistic for the stability of the regression coefficients between pre- and postransition periods. The F-test01 and F-test12 (Goldfeld-Quandt test) are the test statistics for the stability of the standard deviations between the pretransition and transition periods and between the transition and posttransition periods, respectively. The autoregressive regressions are of the form xt-1 = βxt,1, where x is the financial variable. The F-tests that are significant at the 1 percent confidence interval are indicated by an asterisk.

Limited data are available for the posttransition period.

Money market rates for Indonesia, Kenya, Malaysia, the Philippines, Sri Lanka, Thailand, and Tunisia and treasury bill rates for The Gambia, Ghana, Israel, Jamaica, and Mexico.

Sources: IMF staff estimates and International Financial Statistics, various issues.

Mean0, mean1, and mean2 are the sample means for the pretransition, transition, and posttransition periods. Sdev0, sdev1, and sdev2 are the mean-weighted sample standard deviations for the pretransition, transition, and posttransition periods, respectively. The F-test02 (Chow test) is the test statistic for the stability of the regression coefficients between pre- and postransition periods. The F-test01 and F-test12 (Goldfeld-Quandt test) are the test statistics for the stability of the standard deviations between the pretransition and transition periods and between the transition and posttransition periods, respectively. The autoregressive regressions are of the form xt-1 = βxt,1, where x is the financial variable. The F-tests that are significant at the 1 percent confidence interval are indicated by an asterisk.

Limited data are available for the posttransition period.

Money market rates for Indonesia, Kenya, Malaysia, the Philippines, Sri Lanka, Thailand, and Tunisia and treasury bill rates for The Gambia, Ghana, Israel, Jamaica, and Mexico.

In all countries except Jamaica, the volatility of the multiplier increased substantially between the pre-transition period and the transition itself, underlining the potential for a loss of monetary control. However, in most of the sample countries, the variability of the money multiplier fell substantially between the transition and posttransition periods, implying a potential improvement in monetary control, A comparison between the pre- and posttransition periods yields a mixed picture: eight countries showed a higher volatility, six a lower one, and two showed no change. The results of the F-tests indicate that most changes in volatility were statistically significant.

In all the sample countries, the transition process has been characterized by unstable money or credit aggregates, making their interpretation and control difficult. In most of them, the volatility of the M1/M2 ratio rose during the transition period but fell after the transition. The F-tests indicate a significant change in this ratio before and after the transition period, implying a structural break after the monetary reform process. This decline in the level and volatility of the M1/M2 ratio may indicate a shift toward long-term deposits and increased confidence in the banking system or may be a result of positive and higher real interest rates.

The volatility of short-term interest rates mostly increased during the transition period. This may be a sign of a more active interest rate management or simply reflect the fact that interest rates became more market determined. The posttransition experience is less uniform, as about 70 percent of the countries experienced lower interest rate volatility. This is consistent with the expectation that after an initial jump caused by liberalization, interest rate volatility goes down as the central bank and the market gain experience operating in a liberalized environment. However, in most countries, volatility still remained above the level of the pretransition period.

    Other Resources Citing This Publication