A Survey of Recent Empirical Money Demand Studies
Author:
Mr. Subramanian S Sriram
Search for other papers by Mr. Subramanian S Sriram in
Current site
Google Scholar
Close

This paper surveys a selected number of studies that evaluated the demand for money using the error-correction model approach in the 1990s across a range of industrial and developing countries. It briefly presents issues relevant to modeling and estimating the demand for money; and synthesizes information concerning variables, data period and frequency, unit root and cointegration techniques, stability tests, and findings in a tabular form. In addition, it presents estimated long-run income elasticity and elasticities or semi-elasticities for opportunity cost and other variables in a comparable framework. It aims to provide a reference tool for future research on demand for money in various countries.

Abstract

This paper surveys a selected number of studies that evaluated the demand for money using the error-correction model approach in the 1990s across a range of industrial and developing countries. It briefly presents issues relevant to modeling and estimating the demand for money; and synthesizes information concerning variables, data period and frequency, unit root and cointegration techniques, stability tests, and findings in a tabular form. In addition, it presents estimated long-run income elasticity and elasticities or semi-elasticities for opportunity cost and other variables in a comparable framework. It aims to provide a reference tool for future research on demand for money in various countries.

Demand for money plays a major role in macroeconomic analysis, especially in selecting appropriate monetary policy actions. Consequently, a steady stream of theoretical and empirical research has been carried out worldwide over the past several decades. The interest has, however, heightened in recent years, triggered primarily by the concern among central banks and researchers on the impact of the movement toward flexible exchange rate regime, globalization of capital markets, ongoing domestic financial liberalization and innovation, advancement in time series econometrics, and country-specific issues.

The extensive literature underscores two major points relevant to modeling and estimating the demand for money: variable selection and representation, and framework chosen. Failure to provide due consideration to these issues has tended to yield poor results. For the former, proper specification of opportunity cost variables happens to be the most important factor in getting meaningful results. Regarding the latter, the chosen system should be free of theoretical and estimation problems, and should perform well in empirical testing. The error-correction models (ECMs) have shown to meet these criteria.

This paper surveys a selected number of papers that applied the ECM approach to analyze the demand for money (of various definitions) during the 1990s in several industrial and developing countries.1 The objective is to extract relevant information from these studies and provide it in a readily useable and comparable framework. In specific, the paper presents details concerning the techniques followed, variables chosen, periods and frequency selected, and major findings. In addition, it summarizes the long-run income elasticities, interest-rate semi-elasticities (or elasticities), and the coefficients of other relevant variables. It is hoped that the materials presented in this paper provide some reference points concerning the behavior of money demand in various countries, which in turn will help the policy makers in designing appropriate monetary policy actions and the researchers in carrying out further research.2

The paper is organized as follows: Section I briefly specifies the general framework that usually underlies the empirical formulation in estimating the demand for money. Section II carries out relevant discussion regarding the variables and estimation techniques, and summarizes information concerning various studies including the findings and estimated coefficients. Finally, Section III presents the conclusions.

I. General Framework

There is a diverse spectrum of money demand theories emphasizing the transactions, speculative, precautionary or utility considerations.3 These theories implicitly address a broad range of hypotheses. One significant aspect, however, is that they share common important elements (variables) among almost all of them. In general, they bring forth relationship between the quantity of money demanded and a set of few important economic variables linking money to the real sector of the economy (see Judd and Scadding, 1982, p. 993). What sets apart among these theories is that although they consider similar variables to explain the demand for money, they frequently differ in the specific role assigned to each. Consequently one consensus that emerges from the literature is that the empirical work is motivated by a blend of theories.

The general specification begins with the following functional relationship for the long-term demand for money:

M P = f ( S , O C ) ( 1 )

where the demand for real balances M/P is a function of the chosen scale variable (S) to represent the economic activity and the opportunity cost of holding money (OC). M stands for the selected monetary aggregate in nominal term and P for the price. Like in theoretical models, the empirical models generally specify the money demand as a function of real balances (see Laidler, 1993).4

II. Discussion on Variables and Estimation Techniques

Given the above general framework, this section provides a brief overview of issues concerning selection and representation of variables, modeling, and estimation. Sriram (1999c) presents detailed account of these issues, including relevant references justifying various approaches undertaken by the researchers. The literature shows that money demand has been estimated for various aggregates, their components, or certain combination of these components. As definitions of money differ across countries (see Boughton. 1992, and Kumah, 1989). measures considered, including divisia aggregates, also varied across studies. Scale variable is used in the estimation as a measure of transactions relating to the economic activity. It is usually represented by variables expressing income, expenditure, or wealth concept (although a host of other variables is discussed in the literature). The price variable is selected to follow closely the chosen scale variable, although consumer price index is the most commonly used measure.

One of the most important aspects of modeling the demand for money is the selection of appropriate opportunity cost variables. The literature has shown that studies which paid inadequate attention on this matter produced poor results. There are two major ingredients: (i) own-rate and (ii) alternative return on money. The former happens to be very important, especially if the financial innovation has been taking place in an economy (see Ericsson, 1998). The latter involves yields on domestic financial and real assets for a closed economy, and additionally on foreign assets for an open economy. A number of instruments are available to represent the yields on domestic financial assets. The yield on real assets is usually proxied by the expected inflation. And, on foreign assets by foreign interest rate or some form of exchange rate variable. Prior to selecting appropriate opportunity cost variables, careful attention should be paid on evaluating macroeconomic situation and developments in the financial system (including institutional details and the regulatory environment), and degree of openness of the economy.

The economic theory provides some guidance in reference to the relationship between demand for money and its arguments. As the scale variable represents the transactions or wealth effects, it is positively related to the demand for money. The own-rate is expected to be positively related as higher the return on money, less the incentive to hold assets alternative for money. Conversely, higher the returns on alternative assets, lower the incentive to hold money, and hence, the coefficients of alternative returns expected to be negative. The expected inflation generally affects the demand for money negatively as agents prefer to hold real assets as hedges during the periods of rising inflation. The foreign interest rates are expected to exert negative influence as increase in foreign interest rates potentially induce the domestic residents to increase their holdings of foreign assets which will be financed by drawing down domestic money holdings. Similarly, the expected exchange depreciation will also have a negative relationship. An increase in expected depreciation implies that the expected returns from holding foreign money increases, and hence, agents would substitute the domestic currency for foreign currency.5

The economic theory does not provide any rationale as to the correct mathematical form of the money demand function. There is consensus, however, that the log-linear version is the most appropriate functional form (see Zarembka, 1968). While money and scale variables typically enter in logarithms, interest rate variables appear either in levels or in logarithms. Consequently, estimates of the coefficient for the scale variable directly provides the measure of income elasticity, and those of interest rates show either elasticities or semi-elasticities depending on the way they are introduced in the formulation.

The partial adjustment framework was extremely popular in the 1970s. However, it was shown to suffer from specification problem and highly restrictive dynamics (see, for example, Cooley and LeRoy, 1981; Goodfriend, 1985; Hendry, 1979 and 1985: Hendry and Mizon, 1978). To counter these problems, two major solutions were proposed—modifying the theoretical base and improving the dynamic structure. The former led to buffer-stock models (BSMs), which were built upon the theory of precautionary demand for money (see. for example, Laidler, 1984: Cuthbertson and Taylor, 1987; Milbourne, 1988), and the latter to ECMs.6 The BSMs also ran into criticism, especially in their relevance in the empirical estimation (see Milbourne, 1988). Meanwhile, ECMs seem to be promising. An important aspect of these models is that the data characteristics are thoroughly examined before selecting the appropriate estimation techniques. Furthermore, lag structures are selected based on the data generating process of the economic variables and not on a priori based on the economic theory or naive dynamic theory.

The ECM is shown to contain information on both the short- and long-run properties of the model with disequilibrium as a process of adjustment to the long-run equilibrium. Granger (1983 and 1986) has demonstrated that the concept of stable long-run equilibrium is the statistical equivalence of cointegration. When cointegration holds and if there is any shock that causes disequilibrium, there exists a well-defined short-term dynamic adjustment process such as the error-correction mechanism that will push back the system toward the long-run equilibrium. In fact, cointegration does imply the existence of a dynamic error-correction form relating to variables in question (see Engle and Granger, 1987). The major advantage of the error-correction modeling is that the economic theory is allowed to specify the long-run equilibrium while the short-run dynamics be defined from the data.

The earlier ECMs on money demand tended to be based on the single equation cointegrating relationship between money and the chosen scale variables as developed by Engle and Granger (1987). However, further research suggested that multivariate cointegrating vectors encompassing a broader number of variables provided a fuller characterization of the long-run determinants of demand. The specification of such multiple cointegrating vectors between nonstationary variables primarily employs the procedures developed by Johansen (1988) and Johansen and Juselius (1990) which make the original Engle-Granger framework a special case. However, as can be seen from Table 1. a number of other measures available to conduct the cointegration analysis.7

Table 1.

Summary of Demand for Money Studies Involving Cointegration/Error-Correction Modeling in Selected Industrial and Developing Countries

article image
article image
article image
article image
article image
article image
article image
article image
article image
article image
article image
article image
Note: The following abbreviations are used: Monetary aggregates: B = base money; BM = broad money; CHP = currency held by public; CC = currency in circulation; COB = currency outside banks; DD = demand deposits; NM = narrow money; QM = quasi-money; SD = savings deposits; and TD = time deposits. Scale variable: DA = domestic absorption; GDE = gross domestic expenditure; GDP = gross domestic product; GNP = gross national product; IIP = index of industrial production; IO = industrial output; NI = national income; and NNI = net national income. Interest rate: CMR = call money rate; CBDR = Central Bank discount rate; CPR = commercial paper rate; CBR = corporate bond rate; FDR = fixed deposit rate; LIBOR = London interbank offered rate; LTBR = Long-term borrowing rate; MMR = money market rate; CBTD3M = Three-month deposit rates at commercial banks; TDR = time deposit rate; T-bill = Treasury bill; and T-bond = Treasury bond. Exchange rate: DEPR = depreciation; XR = exchange rate; EER = effective exchange rate; and NEER = nominal effective exchange rate. Prices: CPI = consumer price index; RPI = retail price index; and WPI = wholesale price index. Deflators: DAD = domestic absorption deflator; GDED = gross domestic expenditure deflator; GDPD = gross domestic product deflator; GNPD = gross national product deflator; IGDPD = implicit GDP deflator; IGNPD = implicit GNP deflator; IPD = implicit price deflator; and NID = national income deflator. Unit root tests: ADF = augmented Dickey-Fuller; CRDW = cointegration regression Durbin-Watson; DF = Dickey-Fuller; J (1988) = Johansen (1988); KPSS = Kwiatkowski, Phillips, Schmidt, and Shin (1992); P (1987) = Phillips (1987); PO (1990) = Phillips and Ouliaris (1990); and PP (1988) = Phillips and Perron (1988). Cointegration tests: AEG = augmented Engle and Granger; CRDW = Cointegration regression Durbin-Watson; DOLS = dynamic ordinary least squares of Stock and Watson (1993); EG = Engle and Granger; EY = Engle and Yoo (1987); IVT = instrumental variable technique; J (n) = Johansen (n) where n stands for 1988, 1991a, 1991b, 1992a, 1992b respectively; JJ (1990) = Johansen and Juselius (1990); OLS = ordinary least squares; PH = Phillips and Hansen (1990); and PO (1990) = Phillips and Ouliaris (1990). General: avg. = average; CB = corporate bonds; EC = error-correction; Govt. = Government; NCB = nationwide commercial banks; L-T = long-term; and S-T = short-term.

Seasonally adjusted.

Where “it” stands for time deposit rate of deposits between DM 100,000 and DM 1 million.

Spreads between yield on T-bill and net return on time deposits and between yield on T-bill and net return on repurchase agreements respectively.

Own-rate is interest rate on bank deposits, net of taxes; and alternative return is yield on longer-term government debt.

BOT stands for Buoni Ordinari del Tesoro and CCT for Certificati di Credito del Tesoro.

R is defined as the three-month average Gensaki rate minus the average return on holding broad money defined as weighted average of the interest rate on three-month certificates of deposit and the guideline three-month deposit rate.

R = own rate of return for M2 (weighted average of explicit interest rates paid on the components of M2) minus RM2 (four-six month CPR).

mavarπ is annual moving average of changes in inflation calculated as |Δ1n(1+p)|t and mavarr is for interest rates.

Defined as one-year time deposit rate minus the rate of inflation.

FCDS and M2LL stand for U.S. dollar-denominated deposits and Lebanese pound component of M2 respectively.

Table 1 also presents details relevant to modeling and estimating the demand for money from various studies. In specific, it summarizes information for a cross-section of developing and industrial countries, on monetary aggregates (nominal or real), scale variable(s), and the opportunity cost and other variables included; data period and frequency chosen; unit root, cointegration, and stability tests applied; nature of various time series (such as the order of integration and whether seasonally adjusted or not). It also presents the findings. The presentation of information will enable the researchers lo draw some insights into the justification of selecting diverse set of variables and approaches across various countries.

Table 2 summarizes the long-run income elasticities and the semi-elasticities or elasticities of opportunity cost and other variables from those studies listed in Table 1. As the short-run dynamics can be potentially complicated, the table concentrates only on the long-run results. In order to promote comparability, the results are shown only for those studies which reported the long-term relationship (existence of cointegration). If more than one cointegration relationship is found, results are reported only for the preferred cointegration vector(s) as identified by the author(s), which not only meet a battery of statistical tests but also economically make sense with correct signs of the variables and meaningful size of coefficients.

Table 2.

Coefficients of Long-Run Demand for Money Estimated Under ECM Framework in Selected Countries1

article image
article image
article image
article image
article image
article image
article image

Refer to Table 1 for corresponding expansion on abbreviations used in this table.

Semi-elasticities except for those market by *, which refer to elasticities.

Variables in nominal term are shown in upper case letters and in real term in lower case; and all variables are in italics to show that they are expressed in logarithmic term.

Where own-rate or alternative return is not explicitly stated; also refers to the net interest rate measure.

Financial innovation variable.

Elasticities of those variables expressing the income and wealth concepts respectively.

Exchange rate measure.

A measure of foreign interest rate.

Short-term interest rate for alternative return, and the other category includes both NEER and a measure of foreign interest rate.

Long-term interest rate for alternative return, and the other category includes both NEER and a measure of foreign interest rate.

Implicit divisia rental price or user cost index.

Financial innovation variable and volatility measure for yield on long bond.

Measure of price variability.

Exchange rate depreciation.

Foreign exchange risk and a measure of foreign interest rate respectively.

Spread between local and foreign interest rates.

Foreign exchange risk.

Variable expressing foreign influence.

Figures 13 show the distribution of income elasticities for real money as presented in Table 2 for components of narrow money, narrow money, and broad money respectively. The relevant descriptive statistics is shown in Table 3. It is clear from the table, the medians for all three groups are closer to one than to 0.5 thereby indicating that money does not play the role of transaction measure alone. There is no clear guidance from the theory or empirical studies regarding the acceptable magnitude on elasticities or semi-elasticities of the opportunity cost variables. The most relevant information will be the signs of the coefficients—positive for own-rate and negative for alternative return on money and expected inflation. As can be seen from Tables 1 and 2, there are a number of other variables considered to tackle the country-specific issues; in addition, the open-economy type models also employ the foreign opportunity cost variables.

Figure 1.
Figure 1.

Frequency Distribution of Estimated Income Elasticities for Components of Narrow Money

Citation: IMF Staff Papers 2001, 002; 10.5089/9781451973747.024.A003

Figure 2.
Figure 2.

Frequency Distribution of Estimated Income Elasticities for Components for Narrow Money

Citation: IMF Staff Papers 2001, 002; 10.5089/9781451973747.024.A003

Figure 3.
Figure 3.

Frequency Distribution of Estimated Income Elasticities for Components for Broad Money

Citation: IMF Staff Papers 2001, 002; 10.5089/9781451973747.024.A003

Table 3.

Descriptive Statistics for Income Elasticities

article image
Source: Table 2.

III. Conclusion

The study has made an attempt to survey a number of papers that applied the error-correction models to analyzed the demand for money in a number of industrial and developing countries. The major contribution of this paper is that it has summarized the major features of these papers and presents relevant information in a comparable framework to promote easy understanding of the approaches followed, variables included, and coefficients derived. The information presented thus will enable the researchers to compare their own results and approaches with what were undertaken previously in a wide range of countries. Alternatively, it will help identify important factors to be considered before modeling and estimating money demand in other countries exhibiting similar or different economic characteristics. In short, it will provide a starting point to conduct the money demand research using the error-correction approach.

REFERENCES

  • Adam, Christopher S., 1992, “On the Dynamic Specification of Money Demand in Kenya,” Journal of African Economies, Vol. 1 (August), pp. 23370.

    • Search Google Scholar
    • Export Citation
  • Arize, Augustine C., 1994, “A Re-Examination of the Demand for Money in Small Developing Economies,” Applied Economics, Vol. 26 (March), pp. 21728.

    • Search Google Scholar
    • Export Citation
  • Arize, Augustine C., and Steven S. Shwiff, 1993, “Cointegration, Real Exchange Rate and Modelling the Demand for Broad Money in Japan,” Applied Economics, Vol. 25 (June), pp. 71726.

    • Search Google Scholar
    • Export Citation
  • Ashley, Richard, 1984, “A Simple Test for Regression Parameter Instability,” Economic Inquiry, Vol. 22 (April), pp. 25368.

  • Asilis, Carlos M., Patrick Honohan, and Paul D. McNelis, 1993, “Money Demand During Hyperinflation and Stabilization: Bolivia, 1980–1988,” Economic Inquiry, Vol. 31 (April), pp. 26273.

    • Search Google Scholar
    • Export Citation
  • Baba, Yoshihisa, David F. Hendry, and Ross M. Starr, 1992, “The Demand for M1 in the U.S.A., 1960–1988,” Review of Economic Studies, Vol. 59 (January), pp. 2561.

    • Search Google Scholar
    • Export Citation
  • Bahmani-Oskooee, Mohsen, 1996, “The Black Market Exchange Rate and Demand for Money in Iran,” Journal of Macroeconomics, Vol. 18 (Winter), pp. 17176.

    • Search Google Scholar
    • Export Citation
  • Bårdsen, Gunnar, 1992, “Dynamic Modelling and the Demand for Narrow Money in Norway,” Journal of Policy Modeling, Vol. 14 (June), pp. 36393.

    • Search Google Scholar
    • Export Citation
  • Boughton, James M., 1981, “Recent Instability of the Demand for Money: An International Perspective,” Southern Economic Journal, Vol. 47 (January), pp 57997.

    • Search Google Scholar
    • Export Citation
  • Boughton, James M., 1992, “International Comparisons of Money Demand,” Open Economies Review, Vol. 3, No. 3, pp. 32343.

  • Choudhry, Taufiq, 1995, “Long-Run Money Demand Function in Argentina During 1935–1962: Evidence from Cointegration and Error Correction Models,” Applied Economics, Vol. 27 (August), pp. 66167.

    • Search Google Scholar
    • Export Citation
  • Chowdhury, Abdur R., 1995, “The Demand for Money in a Small Open Economy: The Case of Switzerland.” Open Economies Review, Vol. 6 (April), pp. 13144.

    • Search Google Scholar
    • Export Citation
  • Cooley, Thomas F., and Stephen F. LeRoy, 1981, “Identification and Estimation of Money Demand,” American Economic Review, Vol. 71 (December), pp. 82544.

    • Search Google Scholar
    • Export Citation
  • Cuthbertson, Keith, and Mark P. Taylor, 1987, “Buffer-Stock Money: An Appraisal,” in The Operation and Regulation of Financial Markets, ed. by Charles A.E. Goodhart, David A. Currie, and David T. Llewellyn (London: The Macmillan Press Ltd.).