Abstract
In discussions about the efficacy of various monetary policy instruments, attention is often focused on analyzing the money supply process. Although the basic framework of analysis is similar in all economies—developed, less developed, or undeveloped—the money supply process itself is highly differentiated, depending on a variety of factors, such as the openness of the economy, the level of development of the financial markets, their degree of integration, and so on. If the economy is insulated, monetary policy tends to have a far more pervasive impact on the money supply variations than it does in an open economy where short-term capital movements are unhampered (Khatkhate and Villanueva, 1972 and 1973; Kouri and Porter, 1974). Similarly, the nature of the institutional environment also influences the behavior of the financial system, as has been amply borne out by the experiences of the United States and Germany. While variables such as movements in excess reserves, borrowings from the central bank, free reserves, and so on, constitute a special feature of the U.S. institutional milieu, there is no such phenomenon as excess reserves in Germany. Hence, the specification of the models of the money supply process in different countries varies qualitatively.
I. Introduction
In discussions about the efficacy of various monetary policy instruments, attention is often focused on analyzing the money supply process. Although the basic framework of analysis is similar in all economies—developed, less developed, or undeveloped—the money supply process itself is highly differentiated, depending on a variety of factors, such as the openness of the economy, the level of development of the financial markets, their degree of integration, and so on. If the economy is insulated, monetary policy tends to have a far more pervasive impact on the money supply variations than it does in an open economy where short-term capital movements are unhampered (Khatkhate and Villanueva, 1972 and 1973; Kouri and Porter, 1974). Similarly, the nature of the institutional environment also influences the behavior of the financial system, as has been amply borne out by the experiences of the United States and Germany. While variables such as movements in excess reserves, borrowings from the central bank, free reserves, and so on, constitute a special feature of the U.S. institutional milieu, there is no such phenomenon as excess reserves in Germany. Hence, the specification of the models of the money supply process in different countries varies qualitatively.
This paper attempts to analyze the determinants of the money multiplier process in an open developing economy with a fixed exchange rate, using that of Venezuela as an illustration. In Venezuela there is minimal control on short-term capital movements; investment activity can be financed by either domestic or foreign borrowing without much impediment, so that Venezuelan investors can easily substitute domestic assets for foreign assets (principally those of the United States). Though a breakdown of financial asset holdings by sectors is not available, there is a presumption that they have been held largely by the corporate sector. In the context of an open economy, the large amount of foreign assets implies that the foreign interest rate—in the case of Venezuela, the prime lending rate in the New York market—is a relevant opportunity cost of holding domestic financial assets.
The Venezuelan economy differs from a developed open economy in that the Venezuelan banking and financial structure is relatively unsophisticated and to some extent internally fragmented. The financial asset holdings by the four sectors of the economy—household, corporate, government, and financial—are relatively less diversified than one would expect in a developed economy. For example, during 1962–69 savings by the household sector consisted mainly of either currency and bank deposits or shares and other titles of ownership, with no government bonds and only a small proportion of private bonds. The asset pattern of the corporate sector deviates only slightly from that of the household sector (Khatkhate, van der Hoeven, and Villanueva, 1974). As a result of this lack of development on a sufficient scale of domestic money market instruments, the domestic interest rate, of which no comprehensive series is yet available, would not appear to be the main relevant opportunity cost of holding financial assets, though the commercial loan rate series in Caracas has been used in this study.
This paper is organized as follows. In Section II, an annual model of the money multiplier is presented, with a detailed explanation of its assumptions, especially that underlying the excess reserve function. This section also presents and discusses the results for the behavioral equations of two estimations for the sample periods 1950–70 and 1950–72. Section III discusses in general terms the applicability of the model for forecasting and policy simulation, and some preliminary results are presented on the basis of the relevant coefficients. Appendix I contains a mathematical integration of the money multiplier model in this paper with Khan’s (1974) model for the Venezuelan economy. Finally, Appendix II reports the results obtained with a more disaggregated quarterly model which, though still experimental, might be a better basis for useful simulation and forecasting when more refined data become available.
II. The Model
Since the model presented here is that of the money multiplier alone, changes in the sources of the monetary base could be taken as determined in Khan’s (1974) model. In Khan’s framework, the monetary base is largely endogenous, because net foreign assets are determined by a number of variables within the model, including the net domestic credit of the central bank, which is treated as a policy variable. The chain of causality from the net domestic credit of the central bank to the real economy may be described as follows: the variations in net domestic credit affect the money stock, given the money multiplier; the changes in money stock bring about changes in private expenditure, trade balance, and capital flows, and these, together with government expenditure and exogenously given exports, determine the level of gross domestic product and changes in net foreign assets. However, Khan’s model specifies a linear relationship between reserve money and the money supply. This assumes no change in the behavior of the public or the banks; the money multiplier is constant. Therefore, the effects on the money supply of monetary policy instruments such as legal reserve requirements and/or interest rate regulations cannot be analyzed. Moreover, it is not possible to examine the interdependence between the monetary base and the money multiplier through the interest rate mechanism. The model of the money multiplier developed here attempts to complement Khan’s specification of the money supply process. The two models can be formally integrated, as shown in Appendix I.
The annual money multiplier model consists of three definitions and two behavioral equations; the latter are expressed in a log linear form:
where the variables are defined as follows:
endogenous variables 1
C = currency outside the banks
MO = money supply, defined within the model
Re = level of excess reserves
Rq = level of required reserves
TD = total private deposits at the banks
exogenous variables
GDP = gross domestic product in current prices 2
RM = reserve or high-powered money, defined within the model
RVZ = domestic interest rate
Rus = foreign interest rate
The domestic interest rate is represented by the 90-day commercial loan rate in Caracas. This rate is chosen from the Banco Central de Venezuela (1971)3 as it is the only reliable series available. The foreign interest rate chosen is the prime rate in the New York market. This was preferred to the Euro-dollar rate because Venezuela’s money market is integrated more fully with that of the United States than with the rest of the world money markets. This series has been retrieved from the U. S. data bank of Data Resources Inc.
Equation (1) is an identity which defines high-powered money in terms of its uses as the sum of currency outside banks and total commercial bank reserves. It should be noted that this definition does not include reserves of the mortgage and official banks and the savings and loan associations. Unavailability of aggregate data comprising those institutions, in addition to the commercial banking system, has precluded the development of a more general model. Even if data were available, the exclusion of the mortgage and official banks, although not the savings and loan associations, would have been justified since they may not respond so clearly to market forces. Nevertheless, it must be understood that the model in this paper attempts to explain only a part, though a substantial one, of the aggregate money multiplier in Venezuela. It is assumed, for the purpose of presenting the model of the money multiplier, that the monetary base (RM) is exogenously given. However, when the present model is integrated with any model determining the sources of the monetary base, as for instance Khan’s model, RM becomes endogenous.
Equation (2) defines money as the sum of currency plus total private deposits with the commercial banks, including demand, savings, and time deposits. Equation (3) incorporates a simplified assumption concerning the level of required reserves. The proportion of total required reserves to total private deposits (k) is under the influence of the central bank, because it can be altered by manipulating the various legal reserve ratios. However, k is also determined by the public’s distribution of its deposits among various types and to that extent it is not truly an exogenous variable. Lack of available data, as explained below, precludes the use of a more detailed and exact function here.
An exact definition of required reserves in Venezuela is provided by the following alternative equation:
where
Rq = level of required reserves
TD = total private deposits at the banks
D = private demand deposits
S = private savings deposits
T = private time deposits
DG = government demand deposits
SG = government savings deposits
TG = government time deposits
TDG = total government deposits
CPR = paid-up capital and reserves
and
Equation (3.1) defines the level of required reserves in terms of the legal reserve requirements which have been in effect in Venezuela since before 1950, the starting year of the period under consideration. Separate legal reserve ratios are imposed on demand, savings, and time deposits, including government deposits. These ratios have not yet been used as policy instruments, since their level has been held constant. In addition, there is a special reserve requirement which is defined as 40 per cent of the difference between all deposits, both private and government, and a specified multiple of paid-up capital and reserves. In 1971 this multiple was changed from 6 to 8.
In the absence of data relating to deposit liabilities subject to reserve requirements as consistent with the deposit figures used in this annual model, it has not been possible to apply the exact equation (3.1). Nevertheless, a preliminary attempt has been made to extend the behavioral aspects of the model by using quarterly data for recent years, so as to substitute the separate legal reserve ratio for k (see Appendix II).
Equations (4) and (5) are the two behavioral equations of the model relating to the nonbanking and the banking sectors, respectively. The first of these explains the movements in the currency-deposit ratio. This ratio is negatively related to the opportunity cost of holding currency as measured by the domestic interest rate. It is also negatively related to income, since individuals and corporations tend to become more efficient in their cash management as their income rises.
Equation (5) explains the ratio of excess reserves of the banks to their total private deposit liabilities. It is assumed that this ratio is negatively related to the foreign interest rate, which is taken as a relevant proxy for the substitution between cash and other assets in the portfolios of the commercial banks. The fact that commercial banks in Venezuela are prevented by law from holding more than a nominal amount of foreign liquid assets is not in conflict with this hypothesis; for this to be valid it is sufficient that a strong substitution effect appears in the portfolios of the nonbank public between foreign and domestic borrowing—a case widely recognized in Venezuela. Thus, a rise in foreign interest rates induces the public to turn to domestic borrowing, including borrowing from the commercial banks, thereby reducing the banks’ excess reserves and vice versa. This argument implies a passive reserve behavior on the part of the banks. Indirect evidence supporting this hypothesis is also provided by the severe disturbances in banks’ reserves which have been observed during periods of extreme uncertainty in the foreign exchange markets, as explained below in connection with the estimation of the model.
The domestic interest rate has also been included as an argument in equation (5), despite some apparent difficulties. The relatively undeveloped state of the domestic market for short-term financial instruments has made short-term domestic interest rates only a secondary factor in the banks’ portfolio choice; indeed, the banks’ choice in this respect has been rather limited. In addition, the high degree of collinearity observed in Venezuela between the domestic and foreign interest rate series makes it difficult to estimate the independent effect of each of these two series.
Finally, although efficiency considerations similar to those working in the case of the currency equation would make it desirable to include a variable to capture such an effect in equation (5), it is difficult to find a satisfactory proxy to measure the increasing efficiency of the banks. While use of the income variable might be considered, as in the case of the currency equation, this variable in Venezuela has been spuriously collinear with both the domestic and foreign interest rates, so that its inclusion is detrimental to the estimation and the forecasting performance of the equation.
Model equations (1) through (5) explain the money multiplier and, given the level of high-powered money, the money supply. The domestic interest rate and GDP are both taken as exogenous. From equations (1), (2), and (3) the following expressions for the money multiplier and the money supply are derived:
It can be seen from equation (6) that the money multiplier is a function of the currency-deposit ratio, the excess reserve ratio, and the required reserve ratio. The required reserve ratio is the main policy instrument affecting the value of the multiplier, while the other two ratios are explained by equations (4) and (5). Equation (7) is an identity which redefines the money supply as the product of high- powered money and the money multiplier.
Estimates of equations (4) and (5) were obtained by ordinary least- squares for the sample period 1950–70 and 1950–72. The model with estimated equations for the shorter sample period is denoted Model A, while that for the longer period is denoted Model B. The estimates for the two models are presented below.
model a: estimation period, 1950–706
model b: estimation period, 1950–72 6
Dummy variables were included in the estimation to reflect the period of uncertainty in the years 1961–63 (Woodley, 1964). In 1961 the Venezuelan authorities allowed the exchange rate to fluctuate in the unofficial market while the official transactions continued at the official rate. The rate in the unofficial market declined quickly by almost 50 per cent and remained at this lower level during the years 1961-63. In 1964 both the official and unofficial rates were pegged at about the level of the unofficial rate. The dummy variables capture the effect of the uncertainty of this period.7 These are defined as follows:
DUM1 = 1 for 1961, zero otherwise;
DUM2 = 1 for 1962, zero otherwise;
DUM3 = 1 for 1963, zero otherwise
The effect of the uncertainty on the currency-deposit ratio seems to have been mild and short-lived. There appears to have been a moderate increase in the public’s demand for currency in 1961, so that the currency-deposit ratio was above the level that it would have been otherwise. This is captured by the positive coefficient of DUM1 in equations (4A) and (4B). (Since the coefficients of DUM2 and DUM3 were found not to be statistically significant in each of the two currency- deposit equations, they were dropped from these equations.) The effect of the uncertainty on the excess reserves ratio seems to have been rather more pervasive, since the banks, apparently overtaken by events, held throughout the entire period substantially higher levels of reserves than would be expected from the equation. This is captured by the positive coefficients of the three dummy variables, all of them statistically significant, in equations (5A) and (5B).
All coefficients in the above equations are of the correct sign, and those in Model A are roughly of the same magnitude as their respective counterparts in Model B, as expected. Since the model is expressed in logarithmic form, the coefficients represent the elasticities of the currency-deposit and excess reserves ratios to each of the variables appearing on the right-hand side. For instance, the coefficient of In RVZ in equation (4A) indicates that a 1 per cent increase in the domestic interest rate lowers the currency-deposit ratio by approximately 1 per cent. The coefficients are all statistically significant except for those of the domestic interest rate (RVZ) in the excess reserve equations (5A) and (5B).
The overall performance of the equations, as indicated by the adjusted
Equations (4A) and (4B) show that the rise in the domestic interest rate and in GDP significantly explains the observed decline in the currency-deposit ratio.
The behavior of the excess reserves of the banks requires further explanation, because it is considered to be the most conspicuous institutional feature of the Venezuelan banking system. Over the period 1950-72 the banks held a substantial amount of excess cash reserves as a proportion of their total private deposit liabilities, the average ratio being about 10.3 per cent (see Chart 2, center panel). However, this ratio tended to decline secularly from 17.7 per cent in 1950 to 6.0 per cent in 1972 with various interruptions, the most notable being that of the 1961-63 period of uncertainty (a disturbance captured by the dummy variables). The secular decline in this ratio is largely captured by the secular rise in the domestic and foreign interest rates, while the reduction in these rates since 1970 explains the upturn in this ratio in 1971–72.
Venezuela: Plots from Money Multiplier Model A
Citation: IMF Staff Papers 1974, 003; 10.5089/9781451969344.024.A007
Venezuela: Plots from Money Multiplier Model A
Citation: IMF Staff Papers 1974, 003; 10.5089/9781451969344.024.A007
Venezuela: Plots from Money Multiplier Model A
Citation: IMF Staff Papers 1974, 003; 10.5089/9781451969344.024.A007
Venezuela: Plots from Money Multiplier Model B
Citation: IMF Staff Papers 1974, 003; 10.5089/9781451969344.024.A007
Venezuela: Plots from Money Multiplier Model B
Citation: IMF Staff Papers 1974, 003; 10.5089/9781451969344.024.A007
Venezuela: Plots from Money Multiplier Model B
Citation: IMF Staff Papers 1974, 003; 10.5089/9781451969344.024.A007
III. Forecasting, Simulation, and Policy Implications
The within-sample results from the estimated Models A and B for their respective periods of application are shown in Charts 1 and 2, respectively. Chart 1 also shows the forecast for 1971–72, using Model A; this is used to test the predictive performance of the model outside the sample period. The estimates and forecast for the years 1971 and 1972 are summarized in Tables 1-3 below:
Venezuela: Currency-Deposit Ratio, 1971–72
(In per cent)
Venezuela: Currency-Deposit Ratio, 1971–72
(In per cent)
| Forecast Value | Estimated Value | ||
|---|---|---|---|
| Year | Actual Value | (Model A) | (Model B) |
| 1971 | 22.89 | 19.90 | 21.18 |
| 1972 | 20.60 | 16.60 | 17.86 |
Venezuela: Currency-Deposit Ratio, 1971–72
(In per cent)
| Forecast Value | Estimated Value | ||
|---|---|---|---|
| Year | Actual Value | (Model A) | (Model B) |
| 1971 | 22.89 | 19.90 | 21.18 |
| 1972 | 20.60 | 16.60 | 17.86 |
Venezuela: Excess Reserves Ratio, 1971–72
(In per cent)
Venezuela: Excess Reserves Ratio, 1971–72
(In per cent)
| Forecast Value | Estimated Value | ||
|---|---|---|---|
| Year | Actual Value | (Model A) | (Model B) |
| 1971 | 5.83 | 5.83 | 5.84 |
| 1972 | 5.99 | 5.92 | 5.95 |
Venezuela: Excess Reserves Ratio, 1971–72
(In per cent)
| Forecast Value | Estimated Value | ||
|---|---|---|---|
| Year | Actual Value | (Model A) | (Model B) |
| 1971 | 5.83 | 5.83 | 5.84 |
| 1972 | 5.99 | 5.92 | 5.95 |
Venezuela: Money Multiplier (MO/RM), 1971–72
Venezuela: Money Multiplier (MO/RM), 1971–72
| Forecast Value | Estimated Value | ||
|---|---|---|---|
| Year | Actual Value | (Model A) | (Model B) |
| 1971 | 2.72 | 2.84 | 2.79 |
| 1972 | 2.85 | 3.04 | 2.98 |
Venezuela: Money Multiplier (MO/RM), 1971–72
| Forecast Value | Estimated Value | ||
|---|---|---|---|
| Year | Actual Value | (Model A) | (Model B) |
| 1971 | 2.72 | 2.84 | 2.79 |
| 1972 | 2.85 | 3.04 | 2.98 |
The forecast by Model A for the excess reserves ratio is well on track, with only a slight underestimation in 1972; however, it underestimates the currency-deposit ratio more seriously in both years, with the consequence that the money multiplier is overestimated by about 4½ per cent in 1971 and by 6½ per cent in 1972.8 The estimates derived from Model B show a similar pattern, although the error involved is somewhat reduced by the fact that these estimates are obtained from data covering those two years.
Model A was also used to produce a simulation for the years 1971–72 concerning the effect on the money multiplier of a change in the required reserve ratio. For this purpose, it was assumed that the monetary authorities had decided to maintain this ratio at the level of 1970 instead of reducing it as they did in 1971. The following values of the required reserve ratio were involved (Table 4):
Venezuela: Required Reserve Ratio, 1971–72
(In per cent)
1970 value.
Venezuela: Required Reserve Ratio, 1971–72
(In per cent)
| Assumed Value in | ||
|---|---|---|
| Year | Actual Value | Simulation1 |
| 1971 | 16.46 | 17.73 |
| 1972 | 15.73 | 17.73 |
1970 value.
Venezuela: Required Reserve Ratio, 1971–72
(In per cent)
| Assumed Value in | ||
|---|---|---|
| Year | Actual Value | Simulation1 |
| 1971 | 16.46 | 17.73 |
| 1972 | 15.73 | 17.73 |
1970 value.
The effect of this increase in the required reserve ratio on the money multiplier is obtained by comparing the forecast results from Model A before and after the change in this ratio (Table 5):
Venezuela: Effect of Increase in Required Reserve Ratio on the Money Multiplier (MO/RM) 1971–72
Venezuela: Effect of Increase in Required Reserve Ratio on the Money Multiplier (MO/RM) 1971–72
| Year | Forecast Value | Simulated Value |
|---|---|---|
| 1971 | 2.84 | 2.79 |
| 1972 | 3.04 | 2.98 |
Venezuela: Effect of Increase in Required Reserve Ratio on the Money Multiplier (MO/RM) 1971–72
| Year | Forecast Value | Simulated Value |
|---|---|---|
| 1971 | 2.84 | 2.79 |
| 1972 | 3.04 | 2.98 |
The simulated values for both years are lower than the forecast values. This result is indeed fully ensured by the following expression:
which is derived by differentiating MULT with respect to k in equation (6). It should be noted that the forecasts for the currency-deposit ratio and the excess reserves ratio are not affected by the change in k, since this policy variable does not appear in the equations determining C/TD and Re/TD.
Other simulations of the model would be possible by integrating it with another model explaining the monetary base endogenously, as in Khan’s model. For example, the effects of exogenous changes in central bank credit to the banks and/or to the Venezuelan Government on the money supply could be simulated. Such policy changes would affect the money supply through two channels. Firstly, the monetary base would be affected both as a direct result of these policies as well as the induced movements in the net foreign asset component. Secondly, to the extent that the domestic interest rate and GDP would be altered by such policy actions, changes in the money multiplier would follow.9
APPENDICES
I. Mathematical Integration of the Money Multiplier Model with Khan’s Monetary Model for Venezuela
The notation below is the same as in the article by Khan (1974). All his behavioral equations are rewritten in summary form (that is, in general functional notation). The integrated model is as follows:
endogenous variables
M = nominal value of imports
AE= private aggregate expenditure (in current prices)
GE = government expenditure (in current prices)
ΔK = net private short-term capital flows including net errors and omissions (excluding the oil and iron sectors)
RVZ = interest rate in Venezuela (weighted average of interest rates charged on loans by commercial banks)
MO = currency plus demand deposits plus time and savings deposits
GDP = gross domestic product (in current prices) at realized prices
PY = permanent GDP (in current prices) at realized prices
ΔNFA = change in net foreign assets of the Banco Central de Venezuela
RM = reserve money
MULT = money multiplier
C/TD = currency-deposit ratio
Re/TD = excess reserves-deposit ratio
exogenous variables
PM* = price of imports (1963 = 100)
X* value of exports
= required reserve ratio ΔRus* = change in U.S. prime rate
ΔYus* = change in nominal U.S. GDP
D1 * = dummy variable for speculative capital inflows during 1971
COB * = net items in the balance of payments accounts other than the trade balance and short-term private capital inflows and net errors and omissions
= imports lagged one year = private expenditure lagged one year = net foreign assets of Banco Central de Venezuela lagged one year = government expenditure lagged one year = Venezuelan interest rate lagged one year = GDP lagged one year = money supply lagged one year
reduced model
Substitute equations (1), (2) and (3) in equation (8):
Substitute equations (1), (2), (4), and (9) in equation (10):
Substitute equation (6) in equation (5):
Substitute equations (11), (12), and (13) in equation (7):
The above four equations constitute the reduced model of the system; there are four endogenous variables: namely, GDPt, MOt, RMt, and RVZt• Equations (8.1), (10.1), and (5.1) represent the reduced form of Khan’s model, while equation (7.1) is the reduced form equation of the money multiplier model.
II. Quarterly Model of the Money Multiplier in Venezuela, 1968–72
The model consists of four identities and four behavioral equations as follows: 10
The new variables, not defined in the text, are:
endogenous variables 11
DP = demand deposits, including government
SP = savings deposits, including government
TDP = total deposits, defined within the model
TP = time deposits, including government
exogenous variable 12
rT = interest rate on time deposits
The interest rate on time deposits is chosen because it is the only relevant rate in the individual’s portfolio choice among various forms of deposits and currency in Venezuela. By law no interest rate is paid on demand deposits, and the regulated savings rate was altered only once in the period covered by this study (from 3 per cent through the second quarter of 1969 to 4 per cent thereafter). While no published index of time deposit rates was readily available, one was constructed for use in this study by using half-yearly information on the financial statements of the commercial banks, after transforming this into a quarterly series and combining it with the other information.13
The three equations representing the portfolio choice of the nonbank sector in the model, equations (5) through (7), show a pattern which is consistent with a priori expectations. The rise in income, which is a proxy for the increased efficiency in the portfolio management of individuals and corporations, appears to have a dampening effect on holdings of currency and deposits other than time deposits. The direction of the substitution effects from changes in the time deposit rate also is correctly captured in the estimated equations, although the coefficient of this variable is not statistically significant in the equation for savings deposits.
The results obtained in the estimation of the excess reserves function were somewhat surprising. When an equation of the same form as that presented in the text for the annual model was estimated, the coefficient of the domestic interest rate had the wrong sign and for this reason was dropped from the regression. Regression (8) shows, in accordance with the results of the annual model, that the foreign interest rate has a role in the excess reserve behavior of the banks, but the adjusted R2 is too low to warrant any interesting forecast from this equation.
REFERENCES
Banco Central de Venezuela, Anexo Estadístico (Caracas, 1970, 1971, and 1972).
Banco Central de Venezuela, La Economía Venezolana en los Ultimos Treinta Años (Caracas, 1971).
Banco Central de Venezuela, Informe Económico (Caracas, 1969).
International Monetary Fund, International Financial Statistics, various issues.
Khan, Mohsin S., “Experiments with a Monetary Model for the Venezuelan Economy,” Staff Papers, Vol. 21 (July 1974), pp. 389-413.
Khatkhate, Deena, R., and Delano P. Villanueva, “A Behavioral Theory of the Money Multiplier in the United States: An Empirical Test,” Staff Papers, Vol. 19 (March 1972), pp. 125-44.
Khatkhate, Deena, R., and Delano P. Villanueva, “Some Aspects of the Money Supply Process in an Open Economy: The German Case” (unpublished, International Monetary Fund, April 23, 1973).
Khatkhate, Deena R., Willem van der Hoeven, and Delano P. Villanueva, “Venezuelan Financial System and the Monetary Policy Instruments” (unpublished, International Monetary Fund, 1974).
Kouri, Pentti J. K., and Michael G. Porter, “International Capital Flows and Portfolio Equilibrium,” Journal of Political Economy, Vol. 82 (May/June 1974), pp. 443-67.
Porter, Michael G., “Capital Flows as an Offset to Monetary Policy: The German Experience,” Staff Papers, Vol. 19 (July 1972), pp. 395-424.
Woodley, W. John R., “Exchange Measures in Venezuela,” Staff Papers, Vol. 11 (November 1964), pp. 337-66.
Mr. Khatkhate, Advisor in the Central Banking Service, is a graduate of the Universities of Bombay and Manchester. He was formerly Director of Research in the Reserve Bank of India. He has contributed numerous articles to academic journals.
Mr. Galbis, an economist in the Central Banking Service, is a graduate of the Universities of Barcelona (Spain) and Wisconsin. He was an economist in the Western Hemisphere Department before joining the Central Banking Service.
Mr. Villanueva, an economist in the Central Banking Service, is a graduate of the Universities of the Philippines and Wisconsin. He is the author of several articles on mathematical growth models and econometric models of the monetary sector.
In addition to colleagues in the Fund, the authors are indebted to Mrs. Ruth de Krivoy of the Banco Central de Venezuela for helpful comments.
Data for all the variables in this group up to 1970 were taken from the Banco Central de Venezuela (1971). Updated figures for 1971–72 were provided by the Banco Central de Venezuela at the authors' request.
These series are at realized prices and are taken from International Financial Statistics of the International Monetary Fund and the Anexo Estadístico of the Banco Central de Venezuela.
Updated figures for 1972 were taken from the Anexo Estadístico of the Banco Central de Venezuela.
It is not clear from Banco Central de Venezuela (1971) whether the required reserves are inclusive of those of the foreign banks.
For foreign banks k4 is now 1.00 and α = 6.
The t statistics are shown in parentheses;
As is well known, the estimates of the structural coefficients obtained from a model with “one-time” dummy variables, as in this study, are the same as those that would be obtained by effectively removing the “dummy years” from the sample period.
It has been suggested that the world uncertainty regarding international monetary relations during this period might be responsible for an increased demand for domestic currency.
Khan (1974) has found that, in the short run, a decrease in the net domestic assets variable by 10 million bolívares would increase the interest rate by 25.5 per cent and decrease GDP by 5.4 million bolívares.
The t statistics in equations (5) through (8) are shown in parentheses;
Data for all variables in this group were provided by the Banco Central de Venezuela at the authors' request.
Quarterly data for GDP at realized prices were provided by Mohsin S. Khan.
where rD and rS are interest rates on demand and savings deposits and FY is the financial income paid by banks. Series were assumed to be constant between two quarters in any half-year period.
