MONETIZATION AND STABILITY OF MONEY DEMAND IN DEVELOPING COUNTRIES: THE LATIN AMERICA CASE / LE PROCESSUS DE MONÉTISATION DE L'ÉCONOMIE ET LA STABILITÉ DE LA DEMANDE DE MONNAIE

Giordano Dell-Amore Foundation

MONETIZATION AND STABILITY OF MONEY DEMAND IN DEVELOPING COUNTRIES: THE LATINAMERICA CASE / LE PROCESSUS DE MONÉTISATION DE L'ÉCONOMIE ET LA STABILITÉ DE LADEMANDE DE MONNAIE DANS LES P.V.D.: L'EXPÉRIENCE DES PAYS D'AMÉRIQUE LATINEAuthor(s): Ali F. DarratSource: Savings and Development, Vol. 10, No. 1 (1986), pp. 59-71Published by: Giordano Dell-Amore FoundationStable URL: http://www.jstor.org/stable/25829947 .

Accessed: 06/10/2013 12:14

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Giordano Dell-Amore Foundation is collaborating with JSTOR to digitize, preserve and extend access toSavings and Development.

http://www.jstor.org

This content downloaded from 157.182.150.22 on Sun, 6 Oct 2013 12:14:57 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=fgda

http://www.jstor.org/stable/25829947?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


MONETIZATION AND STABILITY OF MONEY DEMAND IN DEVELOPING COUNTRIES: THE LATIN AMERICA CASE

Ali F. Darrat

University of New Orleans

1. Introduction

The degree of monetization in any economy is often defined as the fraction of national income received in the form of money. While most developed economies are almost totally monetized, the extent of monetization is characteristically much lower in the case of developing economies where a relatively large proportion of national income is still received in kind as in the form of barter exchange or self-consumption. Clearly, a

study of monetization and an examination of its underlying factors is vital for under standing the process of economic development *.

It is the purpose of this paper to examine empirically the effect of monetization on the money demand function in the case of some Latin American developing countries using the annual data over 1950 through 19812. Specifically, the focus of this paper is on the implication of the monetization process for the structural stability of money demand functions of these developing economies. Several rationales may be advan ced for undertaking this study. First, to my knowledge, there has not been any pre vious empirical study on monetization in the case of the Latin American countries3. Yet, the pace of economic growth and development in these countries is closely rela ted to the stability of the monetary environment which in turn may be potentially in fluenced by the process of monetization. Second, data limitation is usually cited for the lack of previous empirical studies on monetization. However, as discussed below, the re are some arguments advanced in the literature which can be utilized to suggest a simple proxy of monetization for the type of countries selected. Finally, and perhaps more importantly, several recent empirical studies have found the money demand functions for some Latin American economies to suffer from structural instability4. The

1 For a discussion of the importance of monetization in the economic development process, see Mukherjee (1967), and Chandavarkar (1977). 2 Since monetization is rather a gradual structural phenomenon, annual, as opposed to quarterly, data may seem a more appropriate time framework to study the impact of monetization on the monetary environment. Lack of quarterly data for some variables has also necessitated the use of annual series. 3 For other economies, notable studies include Laumas (1978) for India, Aghevli (1980) for the U.S., and Driscoll and Lahiri (1983) for a number of developing countries. However, among these studies, only Laumas

investigated the implication of monetization for the stability of money demand function. 4 See, for example, Darrat (1984) and the reference cited therein. Surprisingly, despite the voluminous empi rical literature on money demand relationships in Latin American countries, the structural stability of the estima ted functions is often neglected. See, for instance, Meiselman (1970), Khan (1977), Balino (1980) and Cardoso

(1983).

59



SAVINGS AND DEVELOPMENT - No. 1 - 1986 - X

present empirical inquiry is a modest contribution towards examing whether or not the

quick pace of monetization in the countries can be blamed for such instability result.

The remainder of the paper is organized as follows. Section II discusses briefly the measure of monetization used and outlines the money demand model. The regression estimates and the stability results are reported in section III. Section IV provides the conclusions and some implications.

2. Monetization and the Money Demand Model

Four Latin American developing countries, namely Brazil, Chile, Colombia, and Peru, are the focus of the present study5. These economies are generally characterized by a dominant agricultural sector that contributes a sizable proportion of national income. In the 1950's, about a third or more of these countries' gross domestic product originated in the agricultural sector6. It is well-known that in such economies a significant share of the agriculturists' output is retained for self-consumption and for future use in the production process rather than exchanged for money in the market place. Furthermo re, an important component of transactions within the agricultural sector often takes the form of payment in kind (e.g., wages in kind, commodity loans, and share crop ping). As Driscoll and Lahiri (1983) pointed out, most Latin American economies are dualistic with notable differences between the agricultural and non-agricultural sectors in terms of advancement of production, transportation facilities, marketing and invest

ment attitude. Such differences have also contributed to the non-monetizing nature of the agricultural sector. Given these arguments, and following Melitz and Correa (1970), Laumas (1978), and Driscoll and Lahiri (1983), in this paper we approximate the non-monetized income by the agricultural output in each of the four Latin American economies7. Note that, as Chandavarkar (1977) indicated, the main distinguishing characteristic of a non-monetized sector is that a sizable proportion (though not neces

sarily all) of its output is not exchanged for money.

5 Unavailability of consistent data has precluded the inclusion of other Latin American countries'. 6 Except for Chile where the contribution of the agricultural sector amounted to about 15% of GNP in the 1950's. By 1981, the agricultural share in GNP shrank to approximately 10% in Brazil, Chile, and Peru. For

Columbia, this share has been rather steady and still relatively high in 1981 at 23% of GNP. 7 That the agricultural income could be used as a proxy for the non-monetized income is an approach that was earlier suggested by researchers in the United Nations. See, in particular, United Nations (1969).

60



A.F. DARRAT - MONEY DEMAND IN LATIN AMERICA

As to the money demand model, the general theoretical foundation postulates that real money demand is a positive function of real income as a measure of the bubget con straint, and a negative function of the yields on alternative assets available to wealth holders as a measure of the opportunity cost. Two comments regarding the specifica tion of these variables are now in order. Firstly, for developing economies, the relevant measure of real income in monetary analysis is only the monetized component of

gross domestic product and not the total domestic product. Thus, the money demand functions estimated and tested here for structural instability employ the non

agricultural real gross domestic product as their proxy measure of the budget con straint. Secondly, the expected inflation rate, following the general practice8, is used as the prime measure of the opportunity cost of holding money in our group of the Latin American countries. Interest rates were not included here because ? to the extent that financial assets are available ? authorities in these countries often impose controls on interest rates, and furthermore consistent data on interest rates are gene rally unavailable.

Assuming a partial adjustment mechanism between the actual and desired stock of

money, the estimated money demand function in lag-linear form can be written as:

log mt = a0 + log y^ + a2 Jtt + a3 log m^ + et

Where m is real money stock, y is real monetized (non-agricultural) gross domestic product, jt is expected inflation 9, and e is the stochastic term assumed white-noise with zero mean and constant variance. A priori, the underlying theory suggests that a1f a3 > 0 and a2 < 0.

For all countries studied, the money stock is defined narrowly, that is, currency at the hands of the non-bank public plus demand deposits at commercial banks. It may be

argued, however, that different definitions of the money stock would be more appro priate for different countries because of different institutional arrangements. Yet the narrow definition was chosen in order to use a comparable and consistent definition of the money stock across all countries studied. To generate real money stock series, the

money stock was deflated by the consumer price index (1975 = 1). Finally, for the

8 See, among others, Meiselman (1970), Aghevli and Khan (1978), Galbis (1979), Crockett and Evans

(1980), and Driscoll and Lahiri (1983). 9 Note that the inflationary variable must enter linearly since it assumes negative or zero values during some

early years of the estimation period in which case the logarithms are undefined.

61




expected (unobservable) inflation rates, the variable can be measured as a weighted average of current and past actual inflation rates. Assuming static inflationary expecta tions and following Crockett and Evans (1980) and Driscoll and Lahiri (1983), the ac tual inflation rate in any given year (the growth rate of the consumer price index) is used here as a proxy for the inflation rate expected in the next period. Such an as

sumption appears somewhat legitimate in view of the fact that annual data are being employed in the empirical estimation. All the data used in this paper were taken from the International Financial Statistics, and the United Nations Monthly Bulletin of Stati stics (various issues)10.

3. Empirical Estimates and Stability Results

The above equation was estimated for the four Latin American agricultural countries over the annual period 1950 through 1981. Table 1 contains the parameter estimates. Across countries, these empirical estimates appear quite consistent with the prior theo retical predictions regarding the signs of the coefficients and their statistical significan ce. After correcting for first-order serial correlation (except for Peru where no autocor relation was evident), the estimated equations seem free from remaining significant serial correlation n. The overall goodness of the money demand model across coun tries is excellent as indicated by the high values of R2 (adjusted for degrees of free

dom) and the low values of the standard-error of the regression. Plots of actual and predicted real money demand (not shown here) also confirm the closeness of the fit as the estimated equations trace quite well both the level of real money holdings and their

turning points in each country.

10 Where necessary, data were also drawn from Economic Survey of Latin America, and other official docu ments of the Latin American governments. The most serious data problems were encountered in isolating agricultural income from total gross domestic product for the countries studied. 11 Although reported in Table 1, the traditional Durbin-Watson statistic is known to be biased in the presence of a lagged dependent variable among the regressors. Thus, we calculated the unbiased Durbin-h statistic, except for Chile where the statistic could not be computed due to a large coefficient variance.

62




Table 1

ESTIMATES OF THE MONEY DEMAND FUNCTIONS FOR FOUR LATIN AMERICAN AGRICULTURAL ECONOMIES (ANNUAL DATA, 1950 THROUGH 1981)

log mt = oq + ai log /"t + a2 nt + a3 log mt-i + et

Country ao ai a2 a3 R S.E. RHO D.W. h

Brazil 0.729 0.205 - 0.36.10 0.663 0.96 0.0788 0.26 1.86 0.71 (1.72) (2.58) (2.97) (4.46) (1.41)

Chile -0.528 0.477 - 0.39.10 0.479 0.92 0.1789 0.76 1.52 ?

(0.38) (3.28) (2.77) (2.75) (6.18) Columbia -0.283 0.208 - 0.74.10 0.810 0.97 0.0451 -0.28 2.16 - 0.54

(2.19) (2.52) (3.54) (7.80) (1.46) Peru -0.590 0.223 - 0.56.10 0.830 0.98 0.0838 ? 1.83 0.55

(2.07) (2.36) (4.11) (9.01)

Notes The numbers in parentheses below the coefficient estimates are the absolute values of the t-statistics. FT2 is the coefficient of multiple determination adjusted for degrees of freedom. S.E. is the standard-error of the regression. Except for the equation in Peru, the estimated equations are corrected for first-order serial correlation using the Beach-Mackinon maximum likelihood proce dure. RHO is the estimate of the first-order serial correlation coefficient. D.W. is the Durbin-Watson statistic to test for remaining serial correlation, h is the Durbin-statistic to test for first-order serial correlation in the presence of a lagged dependent variable among the regressors.

The empirical estimates also show that the adjustment process between the actual and the desired money balances is somewhat slow across countries and requires mo re than one year to be completed. As to the estimates of real income elasticities, the short-run elasticities are directly given by the estimated coefficients on the income variables since both the dependent and the real income variables enter the regres sions logarithmically. These short-run elasticities are considerably small, about 0.2, except for Chile which is characterized by a higher short-run elasticity of about 0.50. The long-run real income elasticities are calculated as the ratio of the real income coefficient over the speed of adjustment (one minus the coefficient on the lagged de

pendent variable). According to these estimates, the long-run real income elasticities are 0.61 for Brazil, 0.92 for Chile, 1.09 for Colombia, and 1.31 for Peru. Thus, there is an evidence for economies of scale in cash management for Brazil, and that money

63




appears to be a luxury good in Peru 12. As to Chile and Colombia, their long-run real income elasticities are near unity. It is important to note that these real income elastici ties do not suffer from upward bias common in other empirical studies since real inco me as measured here is generated only within the monetized sector in each of the countries considered. The major concern in this paper, however, is not so much with the sizes of the various regression coefficients but rather with their structural stability over time and the effect of monetization on the overall stability of the money demand

relationships.

Thus, the objective is to test whether the quick pace of monetization over the estima tion period in our group of Latin American countries has any significant implication for the stability of money demand. The structural stability of the above money demand equations is explored using the traditional Chow (1960) test. However, in recent litera ture [e.g., Schmidt and Sickle (1977)], the Chow test has come under severe attack

basically due to its apparent sensitivity to the presence of even moderate heterosceda

sticity (unconstant disturbance variance across periods). Since the stability issue is our

primary concern, it is advisable to supplement the Chow test with an alternative test that does not require homoscedastic disturbance terms. In this study, we utilize the test recently proposed by Gupta (1978) that seems to be adequately powerful and less sensitive than the Chow test to the presence of heteroscedasticity13. In the absence of explicit prior knowledge on how to classify the data series into hypot hetically different regimes, we follow Farley, Hinich, and McGuire (1975) and split the

sample period at the mid-point in order to maximize the power of the tests. The calcu lated F-statistics of the two stability tests are reported in Table 2. In the case of Colom bia, both tests suggest that the money demand function is structurally stable over time. This empirical finding may be expected on a priori grounds since the monetization process in Columbia was rather slow and smooth during the estimation period. Equally interesting is the empirical finding that the money demand relationships in Brazil, Chile, and Peru are all structurally unstable according to both the Chow and the Gupta

12 Similar greater-than-unity real income elasticities are reported in several studies for other developing coun tries. For example, see Aghevli, Khan, Narvekar, and Short (1979), and Darrat (1983). 13 See Gupta (1978) and Ali and Silver (1983) for extensive Monte-Carlo examinations of the properties and

empirical power of the Gupta test. This test has been previously used in the literature to test for money demand

stability by Hein (1979) and Darrat (1983) among others.

64




tests14. With unstable money demand functions, monetary policy aimihg at stabilizing the domestic economy becomes rather difficult to formulate and implement. Therefore, it may seem desirable to investigate the underlying source of instability in these money demand equations.

Table 2

STABILITY TESTS RESULTS

Country The Chow Test The Gupta Test

Brazil 13.70* 15.81* Chile 4.10* 4.38*

Colombia 0.36 0.21 Peru 6.18* 5.72*

Note * Indicates rejection of the stability hypothesis at the 5 percent level of significance. (The critical F-statistic is 2.80 with vi = 4, v2 = 23).

It is sometimes hypothesized [e.g., Chandavarkar (1977)] that instability in the money demand function in developing economies is primarily caused by fluctuations in the agricultural (non-monetized) income. Following Laumas (1978), this hypothesis is te sted by estimating alternative money demand equations for Brazil, Chile and Peru that are identical to the previous money demand equations except for the inclusion of the

agricultural real income as an additional explanatory variable. The resulting equations were then re-tested for structural stability. Table 3 reports the calculated F-statistics of the two stability tests for the three money demand equations15. The tests results indi cate that the money demand functions for Brazil, Chile, and Peru continue to exhibit structural instability even after the inclusion of the non-monetized real income. These

empirical results accord with the hypothesis that the instability in the money demand functions in these three countries does not stem from the process of monetization.

14 This result is consistent with the evidence presented in Darrat (1984) on the basis of quarterlydata. 15 To economize on space, the regression estimates from these alternative equations are not reported here but are available from the author upon request. As mentioned before, our main interest in this paper resides with the stability property of the estimated equations rather than with the coefficient estimates per se. It is worth

noting, however, that the inclusion of the agricultural real income into the money demand equations did not

significantly alter the equations' overall estimates.

65




Table 3

STABILITY TESTS RESULTS (MONEY DEMAND EQUATIONS WITH THE INCLUSION OF AGRICULTURAL NON-MONETIZED REAL INCOME)

Country The Chow Test The Gupta Test

Brazil Chile Peru

9.30* 3.30* 6.66*

12.17* 4.01* 6.13*

Note * Indicates rejection of the stability hypothesis at the 5 percent level of significance. (The critical F-statistic is 2.68 with vi = 5, v2 = 21).

As a further check on the above finding, and in order to determine the relative contribu tion of the individual explanatory variables in causing structural instability, a final exer cise was performed using the Gujarati (1970) stability test16. The main advantage of this test over the Chow and the Gupta tests is its ability to examine the stability proper ty of individual coefficients rather than testing for the stability of the entire relationship. Thus, (0,1) dummy variables were employed to generate multiplicative slope-dummy variables for all explanatory variables (including the non-monetized real income). Sta bility of individual variables was then explored by testing the hypothesis that the coeffi cient on each slope-dummy variable is significantly non-zero. If the t-statistic obtained for any given variable is smaller than a predetermined critical value, then the stability hypothesis for that particular variable cannot be rejected. The calculated t-statistics for each variable's stability test are given in Table 4. For Brazil, the test results suggest that the most likely candidate causing instability in her money demand relationship is a

significant drift in the coefficient on the expected inflation variable rather than the exi stence of the non-monetized sector. Therefore, it may be argued that the money de mand instability in Brazil is the result of its unstable inflationary environment. For Chile too, the results suggest that the monetization process has not caused instability in her money demand relationship. Rather, the empirical results show that the adjustment process is the likely candidate causing instability in the Chilean money demand func

16 This test was recently implemented by Batten and Hafer (1983).

66




tion. This may be an evidence that either the Chilean money demand relationship is inherently unstable or that the type of adjustment used (the standard Koyck-lag) is inappropriate for Chile. Hence, an alternative adjustment mechanism may lead to a

more stable money demand function. As for Peru, the test results indicate that the existence of the non-monetized sector may be partly to blame for money demand instability. However, as in Brazil, the Gujarati test shows that the volatility of the infla tionary experience of Peru is also a significant factor causing instability in her money demand relationship17.

Table 4

STABILITY TEST RESULTS OF INDIVIDUAL EXPLANATORY VARIABLES (THE GUJARATI TEST)

Absolute Values of t-Statistics for

Country Monetized Non-Monetized Expected Lagged Dependent Real Income Real Income Inflation Variable

Brazil 0.56 1.96 4.12* 0.85 Chile 0.58 0.86 0.22 2.56* Peru 1.92 3.37* 2.35* 0.93

Note * Indicates rejection of the stability hypothesis at the 5 percent level of significance (The critical t-statistic is 2.06 with 26 degrees of freedom).

4. Conclusions and Implications

The main purpose of this paper has been to investigate empirically the effect of the monetization process on the money demand function and its structural stability in the

17 To check whether money demand instability stems from the common exclusion of interest rates, a proxy was experimented with for all equations, namely a weighted average of short-term interest rates in major OECD countries. This foreign interest rates variable was found in Darrat (1983) to exert a significant influence on

money demand in some OPEC developing countries to the extent that its exclusion caused instability in their

money demand functions. However, the inclusion of such variable into the money demand functions of the Latin

American countries did not prove fruitful as the variable was generally insignificant and sometimes appeared with the wrong signs.

67




case of four agricultural Latin American countries. One may expect that the quick pace of monetization in these economies would cause structural instability in their money demand functions. The likelihood of unstable money demand functions in such econo mies is even enhanced given the often erratic seasonal fluctuations in their dominant

agricultural sectors.

The principal conclusion from this paper is that, except for Colombia, the money de mand functions for the remaining agricultural Latin American countries (Brazil, Chile, and Peru) appear structurally unstable. Except for Peru, the empirical results also indi cate that such instability result was not caused by the monetization process per se. In Brazil (and also in Peru), money demand instability appears to stem largely from a volatile inflationary experience and increased uncertainty regarding the real value of

money holdings. As to Chile, its money demand instability may be the result of an erratic behavior of the underlying adjustment process. All in all, the empirical findings of this paper generally reject the hypothesis that the monetization process has significantly caused structural instability in the money de mand functions in these Latin American developing economies. Therefore, the resul

ting unfavorable environment facing stabilization policies in these economies should not be blamed on their monetization process. Rather, irresponsible monetary policy and the consequent erratic price behavior [as detailed in Harberger (1978)] seems to be among the major causes of money demand instability particularly in Brazil and Peru. Indeed, if anything, the monetization process is conducive to economic growth and stability as it would set free idle productive resources from the non-monetized sector that, if properly channeled, would help promote economic growth and develop ment. Therefore, these countries should induce the private sector to engage in more

monetary transactions and support any systematic effort to enlarge the sphere of their monetized sector.

REFERENCES

Aghevli, B.B. (1980), ? Effects of Banking Development on the Demand for Money ?, in Money and Monetary Policy in Less Developed Countries, edited by W. L. Coats, Jr., and D. R. Khatkhate. Oxford, England: Pergamon Press Ltd.

?, and M.S. Khan (1978), ? Government Deficits and the Inflationary Process in Developing Countries*. International Monetary Fund, Staff Papers, 25 (September), pp. 383-416.

68




?, P. P. Narvekar, and B. K. Short (1979), ? Monetary Policy in Selected Asian Countries ?. International

Monetary Fund, Staff Papers, 26 (December), pp. 775-824.

Ali, M.M., and J. L. Silver (1983), ? Test for Equality Between Sets of Coefficients in two Linear Regressions Under Heteroscedasticity ?. Unpublished Manuscript, Department of Economics, University of Kentucky.

Balino, T. J. T. (1980), ? The Demand for Money and Its Components in Argentina: Annual Estimations 1935 69 ?, in Money and Monetary Policy in Less Developed Countries, edited by W. L. L. Coats, Jr., and D. R. Khatkhate. Oxford, England: Pergamon Press Ltd.

Batten, D. S., and R. W. Hater (1983), ? The Relative Impact of Monetary and Fiscal Actions on Economic

Activity: A Cross-Country Comparison ?. Federal Reserve Bank of St. Louis, Review, 65 (January), pp. 5-12.

Cardoso, E. A. (1983), ? A Money Demand Equation for Brazil?. Journal of Development Economics, 12

(February/April), pp. 183-193.

Chandavarkar, A. Q. (1977), ? Monetization of Developing Countries ?. International Monetary Fund, Staff Pa

pers, 24 (November), pp. 665-721.

Chow, G. C. (1960), ? Tests of Equality Between Sets of Coefficients in Two Linear Regressions ?. Econometri

ca, 25 (July), pp. 591-605.

Crockett, A. D., and O. J. Evans (1980), ? Demand for Money in Middle Eastern Countries ?. International

Monetary Fund, Staff Papers, 27 (September), pp. 543-577.

Darrat, A. F. (1983), ? The Demand for Money in Some Major OPEC Members: Regression Estimates and

Stability Results ?. Working Papers in Economics, No. E-58-83, Department of Economics, University of

Kentucky.

?, (1984), ? The Demand For Money in Brazil: Some Further Results ?. Journal of Development Economics,

forthcoming.

Driscoll, M. J., and A. K. Lahiri (1983), ? Income Velocity of Money in Agricultural Developing Economies ?.

Review of Economics and Statistics, 65 (August), pp. 393-401.

Farley, J. U., M. J. Hinich and T. W. McGuire (1975), ? Some Comparisons of Tests for a Shift in the Slope of a

Multivariate Linear Time Series Model ?. Journal of Econometrics, 3 (August), pp. 297-318.

Galbis, V. (1979), ? Inflation ad Interest Rate Policies in Latin America, 1967-76 ?. International Monetary Fund, Staff Papers, 26 (June), pp. 334-366.

Gujarati, D. (1970), ? Use of Dummy Variables in Testing for Equality Between Sets of Coefficients in Linear

Regressions: A Generalization ?. The American Statistician, 24 (December), pp. 18-22.

Gupta, S. A. (1978), ? Testing the Equality Between Sets of Coefficients in Two Linear Regressions When

Disturbance Variances are Unequal ?. Unpublished Ph.D. Dissertation, Purdue University.

Harberger, A. C. (1978), ? A Primer on Inflation ?. Journal of Money, Credit and Banking, 10 (November), pp. 505-52.

Hein, S. E. (1979), ? Empirical Evidence on the Macroeconomic Demand for Money Relationship in the United

States ?. Unpublished Ph.D. Dissertation, Purdue University.

Khan, M. S. (1977), ? Variable Expectations and the Demand for Money in High-Inflation Countries ?. The

Manchester School, 45 (September), pp. 270-293.

69



SAVINGS AND DEVELOPMENT.- No. 1 - 1986 - X

Laumas P. S. (1978), ? Monetization, Economic Development and the Demand for Money ?. Review of Econo mics and Statistics, 60 (November), pp. 614-618.

Meiselman, D. I., editor (1970), Varieties of Monetary Experience. Chicago: University of Chicago Press.

Melitz, J., and H. Correa (1970), ? International Differences in Income Velocity ?. Review of Economics and

Statistics, 52 (February), pp. 12-17.

Mukherjee, M. (1967), ? The Role of Transaction in Kind in Developing Economies ?. Review of Income and

Wealth, Series 13 (December), pp. 335-351.

Schmidt, P., and R. Sickles (1977), ? Some Further Evidence on the Use of the Chow Test Under Heteroskeda

sticity ?. Econometrica, 45 (July), pp. 1293-1298.

United Nations (1969), ? Monetized and Subsistence Agriculture and Agricultural Activities in GDP ?. Economic Bulletin for Africa 9: 35-39.

LE PROCESSUS DE MONETISATION DE L'ECONOMIE ET LA STAB I LITE DE LA DEMANDE DE MONNAIE DANS LES P.V.D.: L'EXPERIENCE DES PAYS D'AMERIQUE LATINE

RESUME

L 'dlargissement du domaine de l'6conomie mon&aire, appe!6 processus de mon&isa tion est une caract4ristique importante de la croissance 6conomique dans les P.V.D. Le but principal de cet article est d'analyser d'un point de vue empirique I'impact possible du processus de mon&isation sur la demande de monnaie dans un groupe de pays d'Amfrique Latine.

Certains 6conomistes ont affirm^ it priori qu'un processus rapide de mon&isation peut determiner une instability structurale des fonctions de demande de monnaie dans les P.V.D. Se basant sur 3 tests de stability les r6sultats empiriques de cet article mon trent qu'en g?n?ral le processus de mon&isation dans un groupe de pays d'Amfrique Latine n'est pas responsable de I'instabilite apparente de leurs fonctions de demande de monnaie. En effet, pour ne pas introduire d'autres arguments, le processus de mon&isation conduit vers un processus de croissance 6conomique et de stability fi nancidre. Ce processus de mon&isation peut HbGrer des ressources du secteur non

70




mon&aire qui stimulent la croissance et le d?veloppement si elles sont convenable ment utilises. Les rGsultats indiquent aussi que I'instability de la demande de mon naie dans les pays d'Amfrique Latine semble etre due en grande partie k des episo des inflationnistes irr?gulieres et & une forte incertitude concernant la valeur r?elle du stock detenu de monnaie.

71



Documents

MONETIZATION AND STABILITY OF MONEY DEMAND IN DEVELOPING COUNTRIES: THE LATIN AMERICA CASE / LE PROCESSUS DE MONÉTISATION DE L'ÉCONOMIE ET LA STABILITÉ DE LA DEMANDE DE MONNAIE