Demand for money in China

(1)

Greenwich Academic Literature Archive (GALA)

– the University of Greenwich open access repository

http://gala.gre.ac.uk

__________________________________________________________________________________________ Citation:

Zhang, Qing (2006) Demand for money in China. PhD thesis, University of Greenwich.

__________________________________________________________________________________________

Please note that the full text version provided on GALA is the final published version awarded by the university. “I certify that this work has not been accepted in substance for any degree, and is not concurrently being submitted for any degree other than that of (name of research degree) being studied at the University of Greenwich. I also declare that this work is the result of my own investigations except where otherwise identified by references and that I have not plagiarised the work of others”.

Zhang, Qing (2006) Demand for money in China . ##thesis _type## , ##institution##

Available at: http://gala.gre.ac.uk/8251/

__________________________________________________________________________________________ Contact:gala@gre.ac.uk

(2)

D E M A N D F O R M O N E Y I N C H I N A

Qing Zhang

A thesis submitted in the partial fulfilment of the requirements of the

University of Greenwich for the Degree of Doctor of Philosophy in

Economics

(3)

DEC LARATION

1 certificate that this work has not been accepted in substance for any degree, and is not currently submitted for any degree other than that of Doctor of Philosophy (PhD) of the University of Greenwich. 1 also declare that this work is the result of my own investigations except where otherwise stated.

06 December 2006

Signed.. V l lA, I Qing ZHANG, PhD student

SignedTX^S^N^^k. \ .^.AAA^\_J3A[LuDr. Ferda HALICIOGLU, Supervisor

(4)

Acknowledgements

I would like to express my deepest appreciation and gratitude to Dr Ferda Halicioglu for his great advice, help, support and guidance. His valuable comments improved this study tremendously. I would like to express my deepest thanks to him.

I would like to express my deepest appreciation to Professor Anita Ghatak I have benefited immensely from her advice, support, and encouragement over years. Very sadly, she passed away and did not see the completion of the study. I would like to express my deepest thanks to her

Many thanks to Professor Ron Ayres for his continue support, understanding and encouragement.

My deepest appreciation and special thanks goes to Professor Subrata Ghatak for his continuous encouragement and insightful comments of this study.

I wish to thank my husband, who has been very understanding and encouraging Thanks to Mr NG Wing Yuen, Professor Janet Mcdonne, Professor Mervyn Dobbin, Mrs Pamela Elms, Mr.Yan Tao and Mr.Luo Jia for their help during my study.

(5)

DEMAND FOR MONEY IN C HINA

ABSTRACT

This research investigates the long-run equilibrium relationship between money demand and its determinants in China over the period 1952-2004 for three definitions of money - currency in circulation /wo,narrow money m\ and broad money m 2 The

appropriate dummy variable has been added into the functions to assess and evaluate effects of economic reform in China. The additional influences on money demand in China, such as wages, monetization process and saving effects are explored The real wage index w, the ratio of urban population to total population

ROP,

the ratio of total deposit to currency

DCR,

and the ratio of total deposit to income

RDG

have been considered as additional variables in the same money demand functions. To test the stable long-run money demand functions, the Engle-Granger two-stage cointegration method (EGTS), Phillips-Hansen cointegration approach, Pesaran

et al.

(2001) ARDL cointegration procedure along with CUSUM and CUSUMSQ stability tests and Johansen Multivariate Cointegration procedures are employed Granger Causality Test is applied to indicate either uni-directional or bi-directional causality exists in the variables. Wald tests for homogeneity and parameter constancy tests are employed in this study as well.

The estimation results, especially the cointegration analyses show that the real money demand functions perform better than the nominal money demand functions. Narrow money demand m\ presents more satisfactory coefficients than other two in terms of economic theory and econometric diagnostics The stabilization policy should primarily aim at the narrow money m\ This study reveals that the economic reform did bring significant changes to the Chinese economy. Income is shown to be the most important determinant of money demand The other additional variables also have significant effects on the money demand The wage index influence on money demand models is important The raise of monetization process made the money plays a more vital role The impact of ratio of total deposit to income is significant

(6)

CHAPTER

1 Introduction 1

2 Literature Review 8

2.1 Introduction 8

2.2 Theories of Money Demand 8

2.3 Econometric Literature 9

2.4 The Literature on Money Demand in China 24

2.5 Summary and Conclusions 31

3 General Outline of the Chinese Monetary System 33

3.1 Introduction 33

3.2 Economic Reforms 34

(7)

3.2.1 Centralized Planning System before Reforms 34

3.2.2 Various Economic Reforms 35

3.3 Analysis of the Determinants of Money Demand in China 47

3.4 Summary and Conclusions 55

4 Tests of Money Demand Functions in China, 1952-2004:

Data, Methodology and Test Results 57

4.1 Introduction 57

4.2 Methodologies and Equations 58

4.3 Tests for Stationarity 61

4 4 Engle-Granger Two Stage Cointegration Method 64

4.5 Pesaran et al (2001) ARDL Cointegration Procedure 70

4.6 Phillips-Hansen Cointegration Approach 94

4.7 Johansen's Cointegration Approach 95

4 8 Granger Causality Tests 104

4.9 Chow Parameter Constancy Test 105

4 10 Summary and Conclusions 107

5 Models of Demand for Money with Other Influences 112

5.1 Introduction 1

5.2 Money Demand with Real Wages 113

5.3 Monetization Process 113

5.4 Money Demand and Saving Effects 114

5.5 Methodology and Equations 115

5.6 Tests for Stationarity 1' 7

5.7 Pesaran et al (2001) ARDL Cointegration Procedure 119

5.8 Phillips-Hansen Cointegration Approach 138

(8)

5.9 Johansen's Cointegration Approach 5.10 Chow Parameter Constancy Test

5.11 Summary and Conclusions 6 Summary and Conclusions

6.1 Introduction

6.2 Main Empirical Findings

6.3 The Main Conclusions and Findings 6.4 Policy Implications

(9)

LIST OF TABLES

2.1 Summary of Recent Demand for Money Studies in Developing and Developed

Countries 1'

2.2 Summary of Coefficients of Long-run Demand Money Studies in Developing

and Developed Countries 21

2.3 Summary of Demand for Money Studies in China 27

2.4 Summary of Coefficients of Long-run Demand for Money Studies in China 30

3.1 Banking System in China in 2004 37

3.2 Financial Assets Structure 1978-2004 42

3 .3 Comparison of Deposit and Loan Rate during 1994-2004 45

3 .4 China and World Wide GDP Growth Rate Comparison 52

4.1 Unit Root Tests ^3

4.2 Summary of Results of Unit Root Tests 64

4.3.1 Engle-Granger Two-stage Method (first stage- real & nominal m0)

4.3.2 Engle-Granger Two-stage Method (first stage- real & nominal m\) 67

4.3.3 Engle-Granger Two-stage Method (first stage- real & nominal mi) 68

4.4 1 Engle-Granger Two-stage Method (second stage - real & nominal /Wo) 69

4.4.2 Engle-Granger Two-stage Method (second stage- real & nominal ni\) 6)

(10)

4 4.3 Engle-Granger Two-stage Method (second stage- real & nominal mi) 70

4.5 F-statistics for Testing the Existence of a Long-run Money Demand 72

4.6.1 ARDL Estimations (real & nominal mo) 74

4 6.2 ARDL Estimations (real & nominal //?i) 75

4.6.3 ARDL Estimations (real & nominal m2) 76

4.7.1 Error Correction Representation of the ARDL Model (real& nominal

m

{ )without

dummy) 78

4.7.2 Error Correction Representation of the ARDL Model (real & nominal m{) with

dummy) 79

4 7 3 Error Correction Representation of the ARDL Model (real & nominal

ni\

without dummy) 80

4 7 4 Error Correction Representation of the ARDL Model (real & nominal m\ with

dummy) 81

4 7 5 Error Correction Representation of the ARDL Model (real & nominal

m

2

without dummy) 82

4.7 6 Error Correction Representation of the ARDL Model (real & nominal mi with

dummy) 83

4 8 Summary of the Long-run Results of ARDL Estimations 93

4.9 Hypothesis Testing of Income Coefficients via Wald Test 93

4 10 Philips and Hansen Cointegration Technique 94

(11)

4.11 The Order of VAR Model 96

4 12.1 Johansen Cointegration Tests and Results (real m0) 97

4 12.2 Johansen Cointegration Tests and Results (real

n i \ )

98

4 12.3 Johansen Cointegration Tests and Results (real mi) 99

4.13 Granger Causality Tests 105

4.14 Summary of Results of Demand for Money Functions in China 1952-2004 109

5.1 China's Saving Level in 1952-2004 115

5.2.1 Unit Root Tests 118

5.2.2 Summary of Results of Unit Root Tests 118

5.3 F-Statistics for Testing the Existence of a Long-run Money Demand 119

5.4.1 ARDL Estimations (Eq (5 1)) 120

5.4.2 ARDL Estimations (Eq.(5.2)) 121

5.4.4 ARDL Estimations (Eq (5.4)) 123

5.5 Hypothesis Testing of Income Coefficients via Wald Test 125

(12)

5.6.2 Error Correction Representation of the ARDL Model (Eq (5.2)) 127

5.6.3 Error Correction Representation of the ARDL Model (Eq (5 3)) 128

5.6 4 Error Correction Representation of the ARDL Model (Eq.(5 4)) 129

5.6.5 Error Correction Representation of the ARDL Model (Eq.(5 5)) 130

5.7 Philips-Hansen Cointegration tests 138

5 8 The Order of the VAR Model 139

5.9.1 Johansen Cointegration Tests and Results (Eq (5 1)) 140

5.9.2 Johansen Cointegration Tests and Results (Eq (5.2)) 141

5.9.3 Johansen Cointegration Tests and Results (Eq (5.3)) 142

5 9 4 Johansen Cointegration Tests and Results (Eq (5 4)) 143

5.9.5 Johansen Cointegration Tests and Results (Eq (5 5)) 144

5.10 Summary of Estimation Results of Long-run Money Demand in China with

Dummy and Additional Other Variables 1952-2004 149

6.1 Summary of Cointegration Tests (Long-run with dummy) 153

6.2 Summary of Cointegration Tests with Additional Variables (Long-run with

dummy) 154

Appendix

Table 1 Total Wages Bill in China 1952-2004 157

(13)

Table 2 Exchange Rate 1952-2004 157

Table 3 Financial Deepening Indicators 1952-2004 158

Table 4 Real Estate Investment in China 1991-2004 158

Table 5 GDP Per capita in China 1952-2004 159

Table 6 Savings Level and Structure 159

(14)

LIS T OF FIGURES

Figure 3.1 Price Index Chart 38

Figure 3.2 Total Wages Bill in China 1952-2004 40

Figure 3.3 Exchange Rate in China 1952-2004 41

Figure 3.4 Inflation Rate in China 1952-2004 44

Figure 3.5 The Ratio of Money Supply to GDP 49

Figure 3.6 Growth Rate of Money and GDP 1979-2004 51

Figure 3.7 Real Estate Investment in China 1991-2004 51

Figure 3 8 Per Capita GDP in China 1952-2004 52

Figure 3.9 Savings Level and Structure 54

Figure 4 1CUSUM and CUSUMSQ Plots for Stability Tests m^/p without dummy 84

Figure 4.2 CUSUM and CUSUMSQ Plots for Stability Tests mo without dummy 85

Figure 4 3 CUSUM and CUSUMSQ Plots for Stability Tests

mjp

with dummy 86

Figure 4.4 CUSUM and CUSUMSQ Plots for Stability Tests /;?«with dummy 86

Figure 4.5 CUSUM and CUSUMSQ Plots for Stability Tests

mjp

without dummy 87

ni\

without dummy 88

(15)

Figure 4.8 CUSUM and CUSUMSQ Plots for Stability Tests m\ with dummy 89

m

2

/p

without dummy 90

Figure 4 10 CUSUM and CUSUMSQ Plots for Stability Tests m2without dummy. 90

Figure 4 11 CUSUM and CUSUMSQ Plots for Stability Tests

m

2

lp

with dummy 91

Figure 4 12 CUSUM and CUSUMSQ Plots for Stability Tests m2with dummy 92

Figure 4.13 CUSUM and CUSUMSQ plots for Stability Test (Wp without dummy) 100 Figure 4.14 CUSUM and CUSUMSQ plots for Stability Test (m,Jp with dummy)

100 Figure 4 15 CUSUM and CUSUMSQ plots for Stability Test (mj/p without dummy)

1 0 1

Figure 4 16 CUSUM and CUSUMSQ plots for Stability Test (m,/p with dummy) 102 Figure 4 17 CUSUM and CUSUMSQ plots for Stability Test (/wVp without dummy)

102 Figure 4 18 CUSUM and CUSUMSQ plots for Stability Test (/wVp withdummy)

103

Figure 4 19 Chow Test for y (wi) 106

Figure 4 20 Chow test for r (/Wi) 106

Figure 4.21 Chow test for p (awi) 106

Figure 5.1 CUSUM and CUSUMSQ Plots for Stability Tests Eq.(5.1) without

Dummy 131

Figure 5.2 CUSUM and CUSUMSQ Plots for Stability Tests Eq (5.1) with

Dummy 132

(16)

Dummy 132

Figure 5.4 CUSUM and CUSUMSQ Plots for Stability Tests Eq.(5.2) with

Dummy 133

Dummy 134

Figure 5.6 CUSUM and CUSUMSQ Plots for Stability Tests Eq (5.3) with

Dummy 134

Dummy 135

Figure 5.8 CUSUM and CUSUMSQ Plots for Stability Tests Eq (5 4) with

Dummy 136

Figure 5.9 CUSUM and CUSUMSQ Plots for Stability Tests Eq.(5.5) with

Dummy 136

Figure 5.10 CUSUM and CUSUMSQ Plots for Stability Tests Eq (5.5) without

dummy 137

Figure 5.11 Chow test for y Eq. (5.1) 146

Figure 5.12 Chow test for r Eq.(5.1) 146

Figure 5.13 Chow test for w Eq.(5.1) 146

Figure 5.14 Chow test for ROP Eq. (5.1) 147

Figure 5.15 Chow test for RDG Eq. (5.5) 147

(17)

ABBREVIATIONS

ABC - The Agricultural Bank of China

ACF -Autocorrelation Function Approach

ADF-Augmented Dickey-Fuller test

AlC-Akaike Information Criterion

ARDL -Auto Regressive Distributed Lag methods

BOC - The Bank of China

CITIC - China International Trust and Investment Company

CUSUM - Cumulative Sum tests

CUSUMSQ- Cumulative Sum of Squares tests DC - Developed Countries

DCR - Ratio of Total Deposits to Currency

ECM - Error Correction Model

EGTS - Engle - Granger Two-stage Cointegration test

GP1 - General Price Index HQC - Hannan-Quinn Criterion

ICBC - Industrial and Commercial Bank of China

LDC - Less Developed Countries

MSI - Monetary Service Index

OLS- Ordinary Least Square

OPI - Official Price Index

PBC - The People's Bank of China

PCBC - The People's Construction of China

(18)

PP- Philip and Perron test

RDG- Ratio of Total Deposit to GDP

ROP - Ration of Urban Population to Total Population.

SBC-Schwarz Bayesian Criterion

SOB-State-Owned Bank

SOE - State-Owned Enterprise

SYBC-Statistical Year Book of China VAR-Vector Auto regression

(19)

Chapter 1 Introduction

The study of money demand is always useful and important and has been a subject for public debate in many countries. In practice, the estimation of the demand for money is a measurement of the demand for real balances. Theories suggested that the demand for real balances is proportional to the level of income and has an inverse relationship with the rate of interest (Laidler, 1993). Demand for money plays a major role in macroeconomic analysis. As commonly agreed, the effectiveness and success of a monetary policy crucially depends on a stable money demand function which ensures the money supply would have predictable impacts on some other economic variables (Driscoll and Ford, 1980). The stability issue in money demand functions becomes an interesting area for researchers to test the effectiveness of monetary programmes.

Numerous studies on relationship between money, financial development and economic growth in China have been well demonstrated by many economists. Because of its special characteristics, China is selected for careful analysis. It is well recognized that Chinese economic reforms feature a dual frame work, in which market-oriented reform is developed progressively whist the traditional planned system continues to function (Yi, 1994). In the two decades prior to economic reform, Chinese economic development was practically static but economic and financial reforms since the late 1970s have brought about major changes in China The use of monetary policy is being revised and the monetary authorities have much greater responsibilities for keeping a macroeconomic balance. A potential implication of the financial reforms is the increased importance of monetary aggregates in policy making decision (Hafer and Kutan, 1994). China's economic development in the last three decades has not only had great impact on its own economy and development, but also has great effects on the world economy. Although China is still underdeveloped compared with developed countries, monetary control has started to

(20)

play an increasing role and the research into the demand for money in China is being seriously pursued

A few previous studies have empirically examined the relationship between money and its determinants. Many have paid some attention to the centralized planned economy in China The studies from Chow (1987), Yi (1993), Qin (1994), Hafer and Kutan (1994), Huang (1994), Girardin (1996), Arize (2000) and Gu (2004) from different viewpoints demonstrate that a long-run equilibrium relationship exists. However, many studies of money, prices, inflation and other variables still pose a lot of questions Existing researches are still limited and the empirical results are controversial Therefore, this study will employ advanced econometric methodologies and improve current literature on money demand in China.

With rapid liberalization of the financial markets and institutional changes, some special features of China's monetary system since reform have become apparent and a few examples are listed below:

• The monetization process accompanied by rapid income increase of both individuals and enterprises has boosted money demand

• Wages as part of income has changed significantly during the reform

• Household savings have been sensitive to changes in price levels and interest rates

These features are of increasing importance on the determinants of money demand and could be considered as possible additional causes of the influences on the money demand

The main aim of this study is therefore two fold

1) To determine if there exists a long-run stable relationship between money demand and its determinants in China.

(21)

2) To investigate the wider context and find other variables which influence the demand for money in China so as to provide some useful points for analysis and policy making.

The major objectives of this study are as follows: 1) To survey the existing theories.

2) To review some existing empirical works on money demand function in both developed countries (DCs) and less developed countries (LDCs).

3) To analyze some of the existing studies on money demand in China

4) To offer a general discussion of the Chinese monetary system so as to provide a better understanding of the Chinese monetary system and its institutional changes in recent years.

5) To estimate various money demand functions by using time series techniques.

6) To test the stability of the demand for money in the long-run through a comparison of various estimations in following ways:

• Using real and nominal money demand

• Using three definitions of money: currency in circulation /Mo.narrow money m\ and broad money mi

• Using dummy variable to check stability of demand for money in the period before and after economic reform.

7) To explore the role of other variables, some of which were neglected in the previous studies. The influences of real wages, the monetization process, and saving effects will be considered as additional influences on the same money demand functions.

(22)

8) To explore models of demand for money not previously studied in China and to further investigate the link between money demand and economic activities in China

The whole thesis is arranged as follows, during which the overall argument is developed.

• Chapter 2 provides a review of relevant literature with a brief analysis of existing theories and empirical studies on money demand in different countries. Empirical research on money demand requires a background on monetary theory This study will involve a brief review of the theoretical literature of money demand The simple versions of the quantity theory will be followed by the Keynesian theory of liquidity preference and then by more modern variants. As theory evolved, so did empirical research. In recent years, further empirical investigations of the demand for money functions for the long-run and short-run have been provided by many economists. Sriram (2001) provided a comprehensive survey of the empirical literature on demand for money in various countries. Sriram has summarized the major features of previous studies and presented relevant information in a comparable framework to promote easy understanding of the approaches used. Our study intends to extend Sriram's early work in the same format to survey more recent studies in both developed and developing countries for comparison

• Chapter 3 consists of an analysis and discussion of the monetary system before and after economic reforms in China. The discussion will concentrate on various aspects of monetary performance and indicate the possible factors affecting money demand in China. The demand for money plays an important role in the literature of the monetary theories. The historical outline about Chinese monetary system is necessary. However, because China was largely locked into a planned economy in the past, some of the theories employed in developed countries will not suitable for use in the Chinese economy. Therefore, a brief survey of the

(23)

factors which influenced monetary control in China over the period 1952-2004 is necessary and a review of more recent developments in the economy will be included.

• Chapter 4 estimates various money demand functions by using time-series econometric techniques on the annual data over the period 1952-2004 in China. Empirical estimation of a money demand function requires choosing explicit variables measuring both money and its determinants. The theoretical approach defines money by using economic theory to decide which assets should be included in its measure. Many previous studies cover the sample period only to the 1990s and monetary aggregates mostly considered were narrow money wo or broad money or both However, in the last decades, there have been profound changes in money demand and monetary policy in China. Therefore, this study intends to employ more definitions of money including m0, m\ and m2 to check all

the influences. The time period over 1952-2004 was chosen entirely because of availability of relevant data on the Chinese economy. The econometric advances in the last two decades, especially in the case of cointegration techniques, have enabled researchers to test more vigorously the stability issue of money demand functions. The stabilization policy should aim at those components of money which are cointegrated with all the explanatory variables. To achieve this objective, the current study will employ both single and multivariate cointegration procedures along with the stability tests. Recent advanced technique will be employed. To determine the order of integration, three procedures suggested by Autocorrelation Function Approach (ACF), Augmented Dickey-Fuller (ADF) test (Dickey and Dickey-Fuller, 1981) and Phillip-Perron (PP) test (Phillip and Perron, 1988) will be applied. Each of these will be considered in turn. To test the stable long-run money demand functions, the Engle-Granger two-stage cointegration method (EGTS), Phillips-Hansen cointegration approach, Pesaran el al. (2001) ARDL cointegration procedure along with CUSUM and CUSUMSQ stability tests, Johansen Multivariate Cointegration (Johansen and Juselius, 1990)

(24)

procedures and Error Correction Model (ECM) will be employed Granger Causality Test will be applied to indicate uni-directional or bi-directional causality in the variables. Wald tests for homogeneity and parameter constancy tests will be used as well. Through various estimations, the aim is to find out a more appropriate money demand function in China and provide useful indication for policy making.

• Chapter 5 constitutes the core estimations of new models using the additional variables. The conventional study of money demand in China is based on a frame-work which suggests that money demand is generally determined by some scale variables such as income and opportunity cost like the domestic interest rate. However, many economic factors have now become very important in the determination of money demand. Therefore, this study intends to ascertain the influences of other variables and try to provide greater insights regarding the nature of money demand behaviour in the Chinese economy. The real wage index

w,

the ratio of urban population to total population

ROP,

the ratio of total deposit to currency

DCR

, and the ratio of total deposit to income

RDG

will be considered as additional variables in the same money demand functions. The methodologies will be using the same techniques mentioned above in Chapter 4.

• Chapter 6 provides a summary and conclusion of the study.

The contributions and innovation of our research are listed below.

1) Our study extends the data of the previous research to 2004, which is the most update information that one can gather from the Chinese economy.

2) Our study makes various estimations and extends the existing empirical works on the money demand function in China.

(25)

3) Our study employs one of the most recent single cointegration technique, namely . Pesaran

el al.

(2001) which has not been used in the previous money demand

estimations in China.

4) Our study explores and uses a few additional variables like real wages, monetization, urbanization and saving effects to establish the empirical suitability of the demand for money function for the Chinese economy.

5) Our study provides helpful information for appropriate monetary policy making.

Unlike previous research, this study provides more consideration in terms of different variables, models and methodologies in order to arrive at more accurate estimations of money demand functions in China The full empirical results are shown in the tables of each section in Chapter 4 and 5. The detailed reports of the estimates of these models are available in the Appendix section of the attached CD-ROM

(26)

Chapter 2 Literature Review

2.1 Introduction

The objective of this chapter is to provide an analysis of the existing theories and empirical studies of money demand. Macroeconomics defines money as a medium of exchange, a store of value, a unit of accounting and a source of deferred payment There have been engaged vast econometric estimations and discussions of money demand relationship in many countries in the world. The significance of money demand study is well recognized by both economists and researchers when designing monetary policy. The relationship between the demand for money balances and its determinants is a fundamental building block in most theories of macroeconomic behaviour. Indeed, a stable demand function for money has long been perceived as a prerequisite for the use of monetary aggregates in the conduct of monetary policy.

This chapter is divided into five sections. The second section will provide a brief review of theories of money demand The third section will outline a review of empirical literature on the studies of money demand of recent years in both developing and developed countries. The fourth section will make comparison of various money demand studies in China to investigate the further relationship between money demand and economic activities. The fifth section will provide a summary and conclusion.

2.2 Theories of Money Demand

The classical theory of demand for money of Quantity theory shows that changes in the level of the stock of money would change the general level of prices (Fisher, 1911). The Cambridge equation from the work of Pigou (1917) develops and

(27)

improves Fisher's early approach and considers the various factors which influence an individual's expectations to hold money. However, classical theories only emphasize the transactions demand and assume that the economy is at a full employment level Two main approaches to the theory of demand for money were developed by Keynes (1936) and Friedman (1956). Keynes (1936) looks at the functions of money and suggests three motives - the transactions motive, the precautionary motive and the speculative motive. Keynes' analysis of demand for money leads to conclusions which are opposite to those of Fisher's and represents a development of the earlier Cambridge theory. Quantity theory by Friedman (1956) specifies various roles of money. Money is a representative asset. Friedman's work moves attention away from the motives of people to hold money. Friedman needs a theory for the determination of nominal income. The quantity theory is in the first instance of 'the demand for money' (Friedman, 1956). While the original Keynesian theory emphasizes the distinction between the transactions demand depending on real income and speculative demand depending on the rate of interest; Baumol (1952) and Tobin (1958) attempted to link the three motives and show that transactions demand also depends inversely on the real rate of interest. Earlier evidence suggested that a few variables, essentially income and interest rates were capable of providing a plausible and stable explanation of money demand Mundell (1963) provided an introduction of the exchange rate into the money demand relationship and introduced an open economy model to assess the effects of fiscal, monetary and exchange rate polices in a world with ever increasing capital mobility.

2.3 Econometric Literature

Numerous studies in the past have empirically examined the money demand functions in both developed and developing countries. In recent years, further empirical investigations of the demand for money functions for the long-run and short-run have been provided by many economists. The empirical literature on money demand function in less developed countries provides little that is new in the way of

(28)

approaches to the problem of estimation compared with the amount of equivalent work undertaken for developed countries (Ghatak, 1995). Sriram (2001) has provided a comprehensive survey of the empirical literature on demand for money in various countries Sriam's work surveys a selected number of papers that applied the error correction model (ECM) approach to analyze the demand for money of various definitions Sriram has summarized the major features of previous studies and presented relevant information in a comparable framework to promote easy understanding of the approaches followed, variables included and coefficients derived (Sriram, 2001). However Sriram examines the work produced until the end of the 1990s. In this study, we will extend Srirairfs survey to examine more recent money demand studies in both developed and developing countries using the same framework as Sriram "s

Table 2.1 presents details relevant to modelling and estimating the demand for money of recent year's studies Table 2.2 summarizes the long-run coefficients from those studies listed in Table 2.1

(29)

V3 c 3 © U •o_cu — o > 4J

a

t*5 '•3 3 *•> C/3 4^ S O £ "O s «s E Q -*•< s_ CS s E 3 cn n — -Q H m 00 ,3 "5_s CJ w IS -2 5 <8 t/DH s o '•3 « s §_00g. o .3 .2 1 <38 s D c 03 C a> "S Q m P i-i in_<L>_<D 3 t3 t/5 <«3 -SiIs_{re 'C} <~>re en > Q rero o g> O & «5 00_re o •Si^ S o- o S-2 "C o_{re o C.} (/} (X, (I, O 6 3 < ! ' C 3 O O J -—! m—- u}£ <D 3 o m -Si5 3 ^ re >-tn (2 X'"w « "S <->2 :S s ex-o E S 3 § < ^ C/5 CD C/3 O (L>Uh O <L> 5 .2 •O r-> S <Do W) c o "o S £ .2 a re ££= ex .2 is <— M ° g S2 c 2 — l s | S 2 2 0 C3_ >1-CX© « u 2 °-.3 3 o « § 1 S 1 1/1 _{o e} S"3 re o JH C! >-.T3 >->-' >> >- 2 ."3 £ o ~ S if «s i H CU ^5 oo x ON Uh Q < c il 51 5* C/3 Di CL Q 0 "re 3 01 re o Qi <L>CJ (Uo oo y inO p.2 o 0) &• 0 1_{.W O (L>}₀₃S to(/} tA> Q h U H ON ON I-u c o

§

2 o Z G Cure Q •-- ex_re o ?3 >-3 u »S a. a> c£ O J re <L> e o re 2 5 2 -§ 3 T: O O o 1 § I H exft ^: -I g.# •- a. o 2 P > C/D <D C3 a £ o c co o o -o73 <DC3 cS <L> II_ex3 C/3 D O in 3 O fc <L) CL T3 3 - O O .> 3 N < ON ON

I

₃ _O2 o o Z .2 re £' S ^ o Z 5 o r-CN GN ONTH ON 1 <D_€c5 s i i₁ oc

§

vo <D €c5 s ""t CN <D €c5 s ONcc r^i O O_<N <Sl c>2 u ca <D4^ o a_{>- C3} O Nd

(30)

•o_CJ 3 S -w c © U — X! cs H 1/3 00 _s •3 s O UJ x> •= ;/) 2 o C/3H OJjo3 w-. S" c = .5 .2_{o o} <_>3 s D c '3 C O "5 Q CD X3 o ^ C/J O <D

s

co co C_<->CO cn > o COSv ~ 00_{O y} C H O W) T-«00 ^ CO O ^ ^ S O-o 3 3 .a cr 3 u. o C3<L)I-C/3CUtt, 3 < 3 3 O o <L> 3 <L> 3 3 « -a ^ o j5-§ CJ s_o c o o o o T3

§

e £ <l>-T Q 2 5/5 o CO T3_{O d)} II 3 ^ -8 O.1/2 _o uo3 -P u O 3 T3 CJ — O ON ON 1/3r-D. O GO = 2 J S « O n. Q < £ SJ o_r -- s C3§ * .S re <u c£ NO O ON nC3 oc § "S_COOo_o fN o CO ^ o

1 1

<D c 'cl m CON O <L> CO D-.o ^ rNi c/3U; C ^ i c o_ON ON o_= Q.cj 33 e/5 — 3 32 « CLhX C3 3 '3 O Z o re o o y a: .S e a 3 ° = ₊ O C I Z 5 § £_o _{t :} <n cj 00 "E *n ~i ^ a-o o <N 3_en CO C CO^ » S 00

^ z

'Aj ON On tU Q < O e z S O 1^-' o t; 6 |g 00 3

ON5-m

O

o

rsj

1

c s

CO

Cu

CO

•—5

•--5

"O

(Da

>

CO

[J-4

2

5

-o

_a

CO

x>

co

to

C/5 <D

TD

(L>

O

c

_£

•g

<D

V

_h

c

>—>

<L>

o

H <D

-o

£ CN| o o >-co on P -a 3_<u X en h~ O CL CL U, Q < oc oc ON C3 3 £ o

z

a, D O "re_o a: CO o a: (N O o fN r~-ON _re o <N OC 3 cZ T3 s S re ° £ OtJ

(31)

T3 o 3 C *-w S o U X! es H CO CUD •5_c CJ cu S -_{is co} 2 o c/of— c o -5 c/> £ 2_oX)5* o .ST o o O H C </) "1 P <u •s

o

o o Q 1/5_U VhC/5 O o (/} o S -2 re'C o TO

<S>

>

C<2

TO« 3 g? o S> 00 #4TO 6' ^ ^ § £ - ° § • = u u J3 <L>u, (•30. U, 0 13 < 1 ' 3 3 O CJ aj c/i g .5 .= .£ -o

"5 £ —TS

_{u 3 o?} > -, £ < ->= (D (Dw CXT3^ *> c3 T3 S-<Qi <2 £ " 3_O 3_O en O o „„ i/i o >-TO <UT2_£ _™ 3 D 3 o _ _ £ cCCJ

<y

„

<S)o

3 2 £-2 ^ s £ D •§ g c/5CJ co "c 3 a a-T3 3 3 5 •2-g| E ~ H so~ -i o «J C t- ^3 o .£ ° •§ o £ "> u _{"5 to} .3 <s w •SO " ? uj 2 co C CO iJ V? o -Ji VJ o ^ CD(Ui *S C3 CJ «_{\ (L>}

P

£ -1T3 00 £ 3 O o ^ —H CO C o £ o tn 3 3 a 52

*6

p § C/3</3 D D O CJ C <L> CO G_rr

Os

o u. Q < 3 O 0 oo \c s TO TO C2-3 1 * ' UJH =srfE CJ TO5 O O Qi.3 o OS r-ONO

? !

_ONO

Os£

CXOQOQ

P

<D <— o O sc c (L> o 13 COc

J3

•»-H_g

7j

a

_o CO_.

<D

_Xj £ ^ o, —_{pr o} .5 ^ a o' TO3 i- o CD3 oo-3 <N CO W <D O ~ c a ^ <L>_—,coOTOCO<DCSrs c M 2 2 o o G u O to o XJ y <D 2 -5 2 3 £ a £ c •+-Hl-H S_C/)CJO

v

c _c C3g •3 iS_{O O (N} X

£1

<L>cy-) O > , 3 A.o s ^ o r-1/1 ₃ 3 2 i 3? o_ON ON LL. a < o_oo £ -c «» o ^ S 2 u 2 cq c £ O Z <D £ 13 R_{o o}

cC

. £ 03

o CJ

rsj

Os

OS

<u

=3 cr

(32)

•o

4J

3 C "•« c o U fN — j5 « H C/D

a

•5_c

£

2 CJ tu JOw c/> 5 o co H c o -C C/D 2 O & §• O .X o o CJ H C o D H o

o

C/3 03 '3 E o c3 Q i/> <D *-»c/3

O CJ

C/3 o S -2 o co CO> C' | S at)_{o <u} c •— o -r 00 «"C03 o "D-_r3"So 'C ?3 o

W)

Cu O c o 3 cr o !-. u_ 3 < c 3 o O •a_o 03 S I = c/) J2 c/i G o o 'Z2"^3 <Dc/5 i2 H o (D C o £ g ₅ C/)i-_<L>₃ IE O .g c3 5-.X) c "Oo "5•c<L>_OH1o•—i in -*—» (D 3 GO a>_l-H •_»,0 O f„ wrf «g v© | S_C3 w C's o o oo£ O 3 fc S3 a x it CL5i OS oc Os 3 C o 2 z -5 CL, Q O o a; E 03_<L> OS ©_<N<D S § os £•

<r,

o o •a Ci 3 ~ s S _^ I is "g £ CS03 £ K ffi 03 "3 & *₃ _c O.O 2 \3 O 3 H '"§ O 3 o 2 & T 03 3 ja' o ^ O O3 t-a> '$•3 .2 O H <L> X) C3 T3 c3_r- _{03 c/>} § 1 1 « a- 3 § Q ^ b 3 O - « .2 -a _ g b o 8 3 ° > « Si o >- — cC« -1 jS S S i l l s O .Es c £ .Es x C3 S o ,.. 03 ^_{& T3£>} C _{"9 n} 2 = 8 5 J b

U-§

a- ^

£

a-

in

cd

hJ

Oi

3

_Ph cx Q

a

13_o a: <L> &

C~)

as ON r •o ONcr T3oo_{C on} C3ON G o c_Q-) o <L> OQUh

ZJ

'£

_c 3 O

u

OJD #C o SJOC3 ^ Oij O <L> _{d) O} o <D >-"3 o 2 1 « i>! § s s b_{° o} Z -3 y a: .S ,0 3 a.

C3

O u Q. 03 (L> Di <r, Os Os iri r-Os _ea o o -3 O_« (Uw T3 « 'J2 fg 5 "S OQen

(33)

"O o 3 _C e o U r^j

o

3

K

H in Ml 3 •5_s

£

o LU '3 i£ •2 cfl_{2 o} C/5f— s o •3 - §«_OX) r^-O o o CJ H •—c/5_{c o} D f—

CJ

•C

cj

_J—CO O o q = e o O C3 CD> o « W>_{O o} c *-o ^ OX) ^ cs CJ c

CJ

3 O" -H>-.Q C3

CJ

(/) DhUH S" O 3 < 3 3 o u .2 p c c « 3 .2 .2P .ts -ti "S 2 ^ 75 1 1 £ <2g w

~~*

CJ

CO« g-jS*S3 .H *•* ^ -a t: tj o s i s - s ® 3 a. c E .2 CJ£j MlS 33 — £ £>

C3to

t|_l 3 OJ ^ 0X1_— o OfiiJ O r/l "2 3 O ^_{CQt/5 J—i} <D OX) e

! <

o (/) O >-oc oc L_ Q < "o o J2 00 f | 2 ra i: 7i o « s y 3 o£ E o 2 re £ — e « u E w»"2 ° 5 o Z 2 2

o.

Q O <L> CtJ -o Z On ON «/~i ONC3 O 3 -2 "o = 3C3•—_{O o} O 3 >-u. ,o

U

o c/) C3'-S V 2 3 £! - 1/1 _{~ T-l} •2 _S >.'£ 3 ^ ? i 3 » 5 Q ^ a« u J CJ >~ 8 ^ 3 £ O T3 3 ^ 0335 3C Uh Q < § b 3 <* I E O G *-*o o <3 -* 3 S ;S CL Q O "E_o &. *MTD_{R O} CO

CJ

_O O C co C c3 w _{3 1 1} a! ON^ ON^ ont:

CJ

• "C_c3 ^ =3 on g-Q <N hJ ^ ^ <D f -c w <

(34)

•o

<U

3

C

-M S o U

£

3 cs 00

c

•3_s o w <2 •S <«_{S o}

in

H (/! re cj ' 3_cr Sb cj °

9.

U 3

1/5 « = g P 2 C g c 1_E CJ «3 Q in_u. cj -s

o

OJ 1-1C/5_{O CJ} 3 re (/3 _oj </2o J=2ca re"C O CO C/}> a « o g> g 6b <5 00 «S 03 JJ ^ s 2 o a S t ?_{s u e} C/>CXPi 3 < C 3 O CJ (flCJ oxiy o (L>w OJCJ RE <D c/3 cj >* 3 ^ <L> r<-1/1 _' 3 ^_WON SCw r- W o y .£' !•) 3 ^ to G -2_{<ure 5} h cu (/) tL. Q (J c C

fl

o o £ cS o Q _i o 3_OX) O

CJV-H

_p _<D i-H o ^ a a Q o a . s e 3 O l 1

z .1

<s "re CJ

-cC

& o a* •3 ~ 00₃ ₃ O _ 3

z s

x

0° — gf — 3 ONO ON3 0 CU ts cr « © Ci- 00J-3 o 3 3 1 CD <L> 3_O X)_re 55 re CO -O £ S 1 ^ 1_7/5 »n _* ON^_r—(t/5 C ^ <L> CD CD H

CJ

CD "Jrt^ X) w m 00 ON ^ ON £ r-\£ ON c/> ' T3 CJ) 1 - g 2 -o "O!£, re g S Q._<L>_CJ "O~ > H ON ON g o Z cx Q O "re_o o

C

re 0i m O o <N 'Hb c iy5 •o c 03 03 •3 O_C3 0^ O C3 0^ X>r*-._re *—-•a °

^ P

_rera — Q. •3 '-2_{X co} o c <u2 I- 'Z-cjre C "P •O_{_o} ni"O ^ " = P 2 oo ^ — CJ -D o _ u £ . Q- a s "o = _.2

I I I

B,ts.e a o -§ - S > (D c/) C/5 <D V-i_o O c j NO On ON o_ON ON ^ c

C' a

_{o CD} cd £_uire

O X

a,

CL

3 3₃ O re ^3 g O '•n

z s

o o <N_I «Ti ON o o 2

-a

a

rz

73_a3

P

C3 1

£ cC

^

(35)

cn 00 3 •3 3 u UJ 5 —*_—l/l is o c/) H C3CJ 3 a-£0 o .a c E _ o o CJ £! (/} CJ o o cj "5 Q <D to <D »-«c/) CJO 2 C/2 O o I w 'G O TO on > o 3 c3 »— OXj o CJ_c u-o 0X) 2 0D03 & o £ 'C re o co Cu o

a

cj 3 a-35 i-b i/l u. O 3 < 3 3 O <j 1-,o

zij-r.

cj 3 cj r® a- aJ cj H O .O T3 00 cj > *2_{^ C/2}& 3 D C/0CO D D U <J 5C DC CN c>CN Li. Q < s £ 3 £> g • o Z cj ao £ c o o — — (/I 0-Q U

~7Z

o OS re re cj o

a: Pi

r-ON ON_{i I} I Ci r-cj tr_re 3 cr <N O O_<N 00 re e 2 « o > ^ SC re l . § <UX CUDCJ i < s C/l .o 3 CJ-3 O <L»3 j± O re o £3 CJ § a X

cj

3 m "J _o 6' E C c/5 QJU* y

s

t3 CJ o § C/)

cj

la T3_CJ s .£_C3c/) O,'C S ° -+ <UON G ^ cj ^ C ° a- 5 aQ

<S)

o ^ o_{t/5 D} p, c/) « B « § oc oc ON c/) U-Q < CJ "^r_{S c} II I y I Z w 5 3 O —'° .i 3 "tore 'S S2E o B ^ « Z .E re re

CJ

cz g

cS.S C3 o cc: § £_<N 1 t-vOg oc o Ong re S f> u ( j o C -a o o >, r-i 3 O 3 CO

.s-3

« p ao 3 TO1—-5 J d , < CJto .re_y VI CJ > i? 6'_CJ3 s I^ V) 1-^H w >? (J o c£ s 0COn 0C

Os

ON^

U-§

it,

a

0H ' ^ f e O PQ

Pi

3 2: co u 3 J D Q Q

<%

Pi

a: s m Q CQ H H en (L> ON ON rl ^ 9 ON £J| 3 § ^ (L> T3_O ob2 o o S s "o ^_CJo 22 E o on CJ ^ O' ^ CO § 2 D D

c/3ir>

O

D CJ O =3 C/5 2 CJo H 3₃ o 2 CJ CJ 0JDT3 G cj «—] T_Xj

7Z

CJ>_CJ co D c o cn o C C3^ X O (•3 (SiCi d) C3OX) .s s

i f s

>7 !*; 3 re 3 'S o

z

CJ 3 re o o y a: .S re

Pi

re •c_CJ 60 >r, O o r-l O 3

(36)

OJO _c -3_c CJ UJ -£ is) 5 o co H

a

o w c/5 g a 5£2. cj. ? c £ O Tj

U 2

o o :: t/j_o a iu_cj cj

a

p

'- c/>_{cj cj} £ G C/} </io — 2_C3C

O C3

<y>

> c_{' £} fi 00 2 cj c o s bt) OJj re ii S* £• o »-Hi—» re cj c/da. 6'

a

o 3 cr CJ o 5 3 < £ • c 3 O CJ S/5OJj re ao Q-C/3o Q ^ g O ^T3^ l/l cj > S!

o

2 2 D P on c/3 D D U U cn * a C 00 o e 5 s_{re — o} x re : w

aL

CJ re p <L>y a: .5 03 o o a: a: -a_G C3 <D <D O o o_cyo_o O H ' 5Cu -E re1 _^ CQ ir>

CJ

>

o C r-

x

U

O UJ S3 -*•»» r— c3 H 03 C/l <U cu CN r^4 U-. 9 ^_{< CL} eg c E ^ § s i * CJ #—' "re p_{o y} a: .S g O m Z 5 O c<-£ s <1rs O « c/) Z "§ 59 ' "3 S r2 _o 2 H a H ~ .5 5 •H c CJ W ^ OS O c a. c 0 5 }-. O c/o 1 I 8 ep o =-».g CJ CJN 3 •a ~ re M)"2 c S CJ •—c/l 00 ^ — s ° "o i2 x_re CJ b-Q < g o Z CJ oo e re o « 5 G re c g o Z CJ g re p o y ct .g re oi o Os ^-. ON7U •—d> rl B r- § av g.

(37)

•o

iJ

3 C • s o U fN — 3 C5 H C/3 00 -5_c E s cj w 5 -_>a1/5 S -^ CO 2 >5 COCO D D CJCJ G n l-H CO CO o CL C i/) D 3 G 2 E o o Q C/5 o_{U. (/)} O O = s C/5 COX) —!s_I— O a CO> £• -c5 cn <L><D OX) OX) 03 6* j* ^ s &

O

S-S r_{ca d>^}« C/)OHPM C/3 Wh O r* "3 < ! •_d 3 o CJ e c3 C/D 03 <L> O T3_O c/)_=s.3 T3_<L> a_E o o a. a> X> 2 o o S S H° 3 _ o g g) 5 ^ ° 2 £ Z w 2 o Z o £5 if •** cs _ o

3 S

_a. 03_<L> 0£ <L» c3ir, 3 © —,° Q£>_{O ^} O 5 "c3 5t> H SCD nj • o r— 0 OCON 0Con U, D < o_OJO c o 3 5 s o Z CQ H a: — 3 w S'5 S_Q._rr _{_} E g ^ g Q-O x a, o ^ ^ § DC. o 3 G_<L> S3§ -<73 «2i

cCx>

O -£ r-l o ri ^ oo 3 on 5-r^i O <N s_—' •>> . _< <DCJ b .ss> P H

(38)

•o

o s

s

'•*-> c o U fN — x: ee H OX) c 2 O UJ •—c/i -2 on

2 CJ

m

£— rrt C/)_O ob o c CT J o CJ73o CD O o o (/} <L>_V-C/) <L>o O - [3 O ~ CDJO ii w o3'C O C3 c/} > b a « « r, mi_<L>_O feb CUD TO 6" — t3 ^ 2" 5 g-G £ O TO<Dt-, CO0, (JL, G < ! • G G o U "J C/5 TOCJ w -a

<Po

to o >-OC 3CO a> _ U-Q < <—'

K

C§ •d o£ o G o o c TO O o£ •a o u. X5 "TO <l> Qi ^ ~ c>

jC

1

7-

_{oc 9}1 ONc ci o o_<^l 53'is o H O _f-"03G •r- *±2•r" Hi£ 00 aa> C/l a>_»— c_ a> a>.52 s ^ I b o 00 <*>ON - CQ c £_<y^ a> C <L> 3 r; © -5 o_y. 03 £ T3 o3 -C^_{> O} _ o

u &

i § § J 5 J= oot: <u.X i- ti 1)c3 •£ 3 .b

t-*

_a3 ^ "O E « r3t? 3 c ^ o 00 £ 00 e3 O 00C a>i> T3 ci i> •2 £ e3 ^ cu^ "OCQ_C 03H C/2C/3 a. on rG c 8"°r •a eu C/50) JT c

C6c

₀₀ II o_0> <L>ed C-.S J £

<—

QV2

O CD

&>

J '•§ 3_—03 C ^ o -c C4_, tS o S o_c r_->

nj

OX) ao3 _C_CJ X 1) u — Cr! 5 u. rC V 2 *_{5 2 °} C c 1> O > J3 t- w S 03 -C

a.

o —° a>

i „

_{o '} X ^ O a> i> 03 = ,p c-D aj

v)

Q O 5 ^ « .a 5"_{fl>O 0>} ^

Cu

"~

^ 03'to i- o <UQ-Oi) C ^3 ^ cd 0) 00 c03 J= O_X i> ca_{c 13'"} 5 "S 5 J

onc

c

o - ^_{J] ?3} Cci "O

'j

« S £ « fc 3 £ g « i l l caQi^ C =-c Wij 5 25 .; ss <oI| ^ 1)^ C V-P. 5 04 •u Q, S-g_^3 03 -O"o3 03 00 T3 OC OS C in_E_| _E. C b 3 o - 'J E ^ § ^ o xs_{03^ ^}

P4

v 03 03 ;-*C_a. c a. E 00 > c v-o co CQW2 ^ lZ w_{L J—•} £ c/J r=D c VJ a> X C/5 uj a> ^ o i>§ Z '?. 03 _c V O O c •- a> cu O tj ^ . •-^ 3'C ^_{«J-O3 2} qq o o o c U 0) <u -S Cuc«

(39)

Table 2.2 Summary of Coefficients of Long-run Demand Money Studies in Developing and Developed Countries

Authors ^priustrial countries German Beyer(1998) Sample Period/ Frequency Money Elasticity

Real income Interest rate Inflation

1975-1994 Realm3 (quarterly) 0.936 1.61 (RS) -3.279 (RL) -1.78 Others German Lutkepohl, Terasvirta andWolters (1999) 'Greece" jCarfakis (2002) Italy" MuscatcUiand Spinelli (2000) Japan Nagayasu(2003) 1961-1990 (quarterly) Real w, 1 . 1 2 -5.11 0.98 form. Nominal 1948-1997 mi & m2 (annual) 0.349 1.813 DOLS Real run 1861-1996 (annual1₎ -0.003 0.235 1958-2000 (quarterly) Nominal m2 Nominal rth + CD 1.791 Phillips-Hansen [.937 DOLS 7958-1985 1247 1958-1989 1.270 1958-1993 1.299 1958-2000 1.299 -14.10 (rl) 9.94 (rs) .08 (rl) 10.17 (rs) -19.01 (rl) 12.80 (rs) -0.006 0.01 -0.009 -0.015 Japan Raoand Singh (2004) 1972-2001 (annual) Real mx 1.13 J(1988) -0.041 Russia Bahmani-Oskooee andBarry (2000) 1991-1997 (monthly Real m2 2.46 Sweden Hacker and Hatemi-J (2005) 1969-2001 (quarterly) Real ni] before break, 0.273 -0.008 after break, 0.441 -0122 -4.09 -0.18 (exchange rate) Developing countries Bangladesh Siddiki (2000) 1975-1995 Realw2 (annual) 3.26 0.08 (r") -0.145 (r1₎ -0.047 Dominican Public

Carruth and Sanchez-Fung (2000) 1950-1996 (annual) Nominal /«< 1.043 -0.547 0.648 (black market exchange rate) 2 1

(40)

Table 2.2 (Continued)

Authors Sample Money Elasticity

Inflation Others

Period/ Real income Interest Inflation Others

Frequency rate

'fighfLDCs 1973-1999 Real m2* First system (m, y,r mt)

Arize (2005) (quarterly) (desired real Indonesia 2.29 -0.090 -0.66

money Kenya 1.41 0.043 -1.09 Balances) Malaysia 1.47 -0.053 -12.31 Mexico 1.92 -0.010 -0.45 Philippines 2.52 -0.029 -0.55 South Africa 1.05 -0.030 -0.47 Singapore 1.10 -0.020 -4.20 Thailand 1.46/1.38 -0.05/- -0.78/0.53 0.08

^igWLDCs 1973-1999 Real m2* Second system (m, v, rcrn )

Arize (2005) (quarterly) (desired real Indonesia 2.33 -0.08 -0.32

money Kenya 0.91 0.030 -0.34 Balances) Malaysia 1.60 -0.029 -0.06 Mexico 2.27 -0.010 -0.24 Philippines 2.72 -0.030 -0.30 South Africa 1.25 -0.018 -0.07 Singapore 1.70 -0.040 -0.003 Thailand 1.56 -0.010 -0.26

Hungary and Poland 1991-1998 Real Poland

Buch (2001) (monthly) ni\ & m2 1.14 (Real rn2) 0.003

0.95 (Real aw,)

Hungary

0.95 (m2) -0.002 -0002

0.55 (m2) -0.001 -0.04

India 1953-2002 Real ni\ 1.185 -0.23

Raoand Singh (annual) (2003)

India 1951-2001 Nominal m3 1.06 ( for output) 1.53

Ramachandran (annual) .(2004)

Korea 1973-1997 Real mi 0.69 (nil) -0.08 (m2)

_Hwang (2002) (quarterly) Real m2

Malaysia 1973-1995 Real m2 TD3R INF TB3MR

Sriram (1999) (quarterly) 1.036 J (1988) 4.884 -4.745 -5.391

1.130 JJ (1990) 2.510 -4.891 -1.834

(41)

Table 2.2 (Continued) Authors Sample Period/ Frequency Money Real income Elasticity Interest Inflation rate Others Nigeria ikinlo (2005) 1970-2002 (quarterly) Real m2 1.095 -0.097 0.008 (exchange) Seven-Asian Countries

Bahmani -Oskooee & Rehnian (2005) 1973-2000 (annual) Real ni\ Real rn2 ni\ / m2 India 0.55 / -3.83 Indonesia 1.29/3.92 Malaysia 1.20/1.25 Pakistan 0.86 / 0.58 Philippines 0.30/1.13 Singapore 0.73/1.48 Thailand 0.14/0.90 EX ni\ / m2 0.27/3.49 0.65/0.32 -0.64/-0.75 -0.15/0.34 -0.04/-1.09 -0.52/0.46 2.57/-0.76 "South Africa Nell (2003) 1968:1998 nominal m} 1.48 EGC 1.58 Pesaran et al. 1.56 JJ (1990) 0.94 0.92 0.93 Thailand

Bahmani-Oskooee &

Techaratanachai (2001) 1977-1990 (quarterly) Real m2 1.20 -0.12 0.62 Turkey Halicioglu & Ugur (2005) 1952-2002 (annual) Real ni\ 0.939 -0.13 -0.064 (for ER) Turkey Akinci (2003) 1987-2003 (quarterly) Real cash balances 0.85 (PCONr) -0.284 (RTB) -0.051 Turkey Civcir (2003) 1987-1999 (monthly) Real broad money 1.009 -13.393 (Rb) 17.613 (Rd) -31.333 ( 7T ) -1.461 <se₎

Notes:r'1_{represents domestic interest rate, r}1_{represents foreign interest rate. TD3R} _{Interest on 3 month time deposits. TB3MR- discount rate on 3month}

treasurybills. ER is exchange rate. PCONr represents private consumption expenditure. FR is foreign interest rate. EX is exchange rate. RTB is the interest rateon government securities. Rd- is interest rates on government securities, Rb is the interest rates on lira deposits

The preceding literature review shows that the money demand functions in many countries have been estimated using various aggregates. Scale variables which represent the transactions or wealth effects are positively related to the demand tor money. The opportunity cost variables such as foreign interest rates and expected exchange depreciation are widely considered and expected to have a negative effect on money demand. Sriram (2001) concluded that the extensive literature underscores two major points relevant to modeling and estimating the demand for money: variable selection and representation and the framework chosen. According to previous studies, the independent variables in the demand for money function could normally fall into three groups - the scale variables, the opportunity cost variables and other

(42)

variables. In the real sector of the economy, other variables play a role in the demand for money function and some but not all of these influences have been considered in a number of studies above. In terms of the methodologies, many studies employed Johansen (1988) and Johansen and Juselius (1990) cointegrating techniques to exam the long-run relation between demand for money and its determinants. A number of studies have applied Pesaran

et al.

(2001) and stability tests. In applying the cointegration technique, it is necessary to determine the order of cointegration of each variable As noted in the literature, depending on the power of the unit root tests, different tests will yield different results In view of these problems, the single cointegration method proposed by Peseran et al. (2001) appears to emerge more in empirical studies of money demand functions This method has a number of econometric application advantages over the previous cointegration techniques. (See examples of Bahmani-Okskooee and Ng 2002, Akinlo, 2005; Halicioglu and Ugur, 2005). In general, studies of demand for money functions for LDCs indicate that stability functions can be estimated in the same lines as constructed for DCs. From the above analysis, the studies relating to LDCs have yielded additional evidence on the role of money demand and its determinants and provided a useful framework for research in China

2.4 The literature on Money Demand in China

China has special features compared with above different countries. Since China was largely locked into a planned economy in the past, some of the theories and examples may not have been employed in the Chinese system From a general point of view, Chinese economic development can be separated into two periods: before and after economic reform. During the two decades prior to economic reform, the Chinese economy was under the control of central planning system. The third meeting of the Eleventh Central Committee held in 1978 can be regarded as the starting point of Chinese economic reform. After which many reforms of the whole economic system

(43)

were carried out. China's financial sector reform comprise mainly of two important and interrelated aspects: reforming the existing financial institutions and creating new financial institutions, instruments and markets (Shang, 1998) In the 1990s, a more comprehensive reform was implemented focusing on the commercialization of state-owned banks (SOBs). China has made dramatic changes in the macroeconomic and monetary conditions in recent years and experienced several stop-go and economic fluctuation. Each economic fluctuation was mainly brought about by economic overheating and retrenchment during these market reforms.

The empirical work on money demand in China was started by Chow (1987). During 1983-1984, the currency in circulation in China increased about 50% as a result of the monetary policy allowing individual banks to extend credit without establishing a mechanism of monetary control by the central bank. The possible effect of this increase in money supply on inflation became an important issue for Chinese economic reform in 1985 Chow's work on money and price level determination in China attracted theoretical interest over the world as well as being of relevance to price reforms in China. Chen (1989) investigated causal relationships between three alternative monetary aggregates and four indicators of macroeconomic performances - economic development, budget deficit, trade deficit and price stability in mainland

China. In Chen's analysis, a causal relationship existed between currency and nominal income, currency and budget deficit and currency and trade deficit. One way causality existed between currency and total inflaton. Hence, it can be concluded that around the 1980s, currency was the best target for monetary policy (Chen, 1989). Yi (1993)'s empirical evidences indicated that the demand for money in China changed significantly before and after economic reform. In Yi's study, the measure of the opportunity cost of holding money was omitted This was because during the period of 1952-1988 in China, interest rates had been strictly controlled and virtually fixed at a level far below the equilibrium level. A few studies also omitted the interest rate for the same reason. It would be natural to include the inflation expectation in Yi's model as an explanatory variable. According to Yi, the official retail price level was virtually

(44)

frozen during the period 1952-1978 and the market price level was also relatively stable before the economic reform, so it is fair to say that the inflation rate was extremely low during that period Yi (1993) suggested that the growth rate of money supply should accommodate both GDP growth and the monetization process, as each important reform step is associated with a monetization consequence (Yi, 1993). Qin (1994) surveyed the money demand function in China and was particularly concerned by the effects of economic reform Qin (1994) tried to find an economically meaningful money demand function with relatively constant coefficients on the basis of the belief that any empirical results worthy of economic interpretation must exhibit certain degrees of parameter constancy Girardin (1996) discussed the question about 'Is there a long-run demand for currency in China 9 _{The record of Chinese monetary}

authorities at targeting m() in the late eighties and early nineties was rather poor

Girardin (1996)'s research aims at determining whether the instability of currency demand is responsible for this. The results show that using adequate economic econometric techniques a long-run demand for currency did exist over the period of 1988-1993 with quarterly data. Another objective of Girardin's paper was to test for the robustness of the result Because most previous studies concluded that income elasticity of currency demand was very high, Girardin's work showed that the income elasticity is unity when proper account is taken of institutional variables representative of the transition process. Yu and Tsui (2000) constructed a monetary Services Index (MSI) for China. Compared to the traditional simple money aggregates, this index has solid microeconomic foundations and has consistent variables. More recent study from Gu (2004) revealed that a stable long-run money demand function for both narrow and broad money exists in China over the period 1952-2000 However, Gu (2004) found there is a couple of weaknesses in structural break test. To arrive at a better understanding about the money demand in China, Table 2.3 and 2 4 summarizes the results of some previous studies of demand for money in China. Table 2.3 presents details of relevant to modeling and estimating the demand for money in China. Table 2.4 summarizes the long-run coefficients from those studies listed in Table 2.3.

(45)

C5 ^3 xi U 5/5 '"5 3 -M C/3 C o u c® •o c «S E 4> fi «*. o

t>

a

E E 3 (Z3 c4 X! ee H in OJO C "5 C Li-UJ •—1/5 •S W S o c/3E-c o '•3 1/3_{03 O} tv 3 &X)r— o .£T E Jl "o o O

2

a o

D 2 2 c/>_u. O O X O O 'C ^3 0-CO <D O O -5 2 •p f-« C/io S -21 « u S a c/3 > £ °o O CJ § 6 b 00 CT5 C/}tt. o_r-I -5 3 < H £ • Q

,

o o Q- o> .c •—< y S-sa>_C/5o_o •a •n s COo-2 O c a- _Is o >_{O a.} H a.i—oX a.

aL

O 15 a Q

i

re o

o£

r-oc on o

xt

CJ t/3 03 _</) -a

v

•Ss •i 1 .5 ^ .2| 1 1 Q. X o ca —3 _S "? a. !2 CL 3 £ oi j3 D Q a. a£ O SE Q; O "re_o

CC

a "re 0

01

i> ai' X 00 y C 2 X) 00 ON Ov 00 00r

^i'

00 00 ON 113g ONCD ON1"5DON1

u*

u

s <N <NC

c,

re •Ai 00 «r>CH00 ON

<

ON

o

ON< ONa r_~H 1—1 t-H oc> on c «J O I 3 q CQ§ r»l ON ON >~ — -a u y c w _{o C «} ts oJj-o > o O 0COn DC o

a

w u. Q < •a_o re

S

E £ a: a.

2

"red> 5 £ — OS_{w -a S} s s i (N <N •—•(N 5 S 13_o 13_<L> 13 13_<L>_a> 0^ OH oo

m m

00O CN ON ON ON T2

dj

•c

m

Onm 00 00 00 ONONONQ ON ON OX) G — <L>^ CO -4-i H W

(46)

-o iJ s o u r4 J2 OJD c Lu 2 u U-l -£ u?_{2 o} c/}t— c o w to re o ab3. a ET o <J o a c o 3 D 2 2 O o * •c_{o_.=}V o_i—in O o £ E

§

00 0 o 1 &_ofj 03 c/) Lu o 5 a < T3 c 7Z^ P oc CJ > X X Ll. tu Q Q < _J C/2 K *i 2 C3 o OS c C/3_{o X>} o o X CU 0-C/3 o CS_O C/D C 03 > Q z Q C/DCSO C 03 > a a T3_o C 1-1 T5Z s? >. a: X> On OS C a in aj >< o G CJ o UJ u. Q < CL U ao a: CU

z

o "re_o a: 3OS as OS 00 c re 3 E to o >-3COs 3CCs U-Q — CLUJ OSQ re o oS 0-Z o "re_o OS re re o <L) OSOS ON Os <l>re t3 3 ffi NS X X u. a < Q_ K oc rsi g E o Z a re o OS o ONsON as OC^ CN ON 113g CNi tn cmr- 3}00re ONc_CH <-Nj 00 •/^ 00

ON< ONa av < ON< ON

1 1 s^ .—4 \D_as Os e -3 O

(47)

•o 5

J

3

C

•*» s o

U

ro c4 — 3 C5 H c/i cjq c O W —(Z> "2 on S o c/3E— c o 2 " OX)g. o .ST 5 ji o o O a C O to D £ Si C/i •— O •r- "O ^ j-CL.5 C/3 <L> V-iC/3 <L><D C cS (•) E_C3 C C/D '5_r _^ _n <3J*.S J-H

ri. y «

Q c/5 > C' 3 2 a I g> *2 00 ^ re o 'C o >, a, o i2 5_{a. 3} a cr S P_{re kH} C/)tu o -3 < M) O 2 C/3 CJ > o _{(/) c} Q.o ;s (/i— = 51 :r a.

K

co

Z C3 <D c3 cd c3

0 CJCJ

01QCOl

o-CN1) (/} <D > C/3 -1

o

_Q _ cu uj OSQ re o

CC

cu Q O

cC

re re <do

OL

o£

-a £

« 2

c- a

o o MiS£ <uc >- re O I -1* o ON o ONIS 2 1IS2 <Ni <Nc fN s «/^i ON< ON 1—1 o -z: o (N 2 (N_{. C}'—<D <nre 00 3 O 3

o

2 c Ml o o < 03^ ,1 "2 e 3 .2 C.~ o i-D,JH 3 2 " & 2 C3 r-o | 0 Je 1 & s. »•

°L

S Ct!—1_.5 + i: ^ -1 _£ & c/j O C c/> o « P —OA <u?-T3.= X5 5£fc o

2

dS

•4°

= SS 1>eB •= 3 td ^ S T3_etf_c o o • § « v C ? •c a c_{3 . O} C 3 o>^2_<uC£ u tu,^ ^ a. £; SO T3_{C 3} - TD 1> ±1 •Si e s O

B-• E

X 5 a>T3 1 -s a>lu c ^ ^ o fe a> £ § 2 f § §

U

_{. ^}c —.o ^ o U-2 . & 73O .S osj 0) U.* CJ O E.3 u 5 .2 u O u V. C O <D O C Ci. "S4>_o}"O o o &_ohl c_«J JH_0/) c w u O_pj —.S 1 ₃ _ O o-. ^ -S S 1 O-O Z .£ C3PH

(48)

Table 2.4 Summary of Coefficients of Long-run Demand for Money Studies in China

hors Sample Period/ Monev Elasticity

Frequency Real income Interest Inflation others

rate >w (1987) 1952-1983 Real m0 1.162 - - -(Annual) ton 1983-1988 Household 1.66 - -1.0 -Ha (1990) (Quarterly) Monev balances (1993) 1952-1989 Real m2 1.25 - - -(Annual) 0.719 1983-1989 0.949 (

(Quarterly) Real m2 1.152 UP)

0.530 (UP) ;ng 1983-1988 Real ni] 1.53 - -1.51 -il. (1994) 1989-1993 Real m2 1.81 -2.21 1983-1993 (Quarterly) lang (1994) 1979-1990 Real m2 2.12 -0.29 - -(Annual)

ifer and 1952-1983 Real m0 1.13 0.13 2.48(DEF)

-itan (1994) (Annual) Real rn2 1.33 0.15 1.52

(DEF) rardin (1996) 1988-1994 Real m0 1.391 - -0.826 (Quarterly) rize (2000) 1952-1994 Real m0 1.326 -0.169 0.721 -(Annual) Real m2 1.510 - 0.334 Real m} 1.134 -0.319 0.306

u (2004) 1952-2000 Rail m„ By Johansen Model -

-(Annual) Real m2 1.3165 (RM0) AN 0.4426 1.5354 (RM2) AB 0.4082 1982-2000 1.5369 (RM0) QN 0.1155 (Quarterly) 1.5094 (RM2) QB -0.1503 By DOLS 1.336 (RM0) AN 0.219 1.556 (RM2) AB -0.002 1.335 (RM0) QN 0.002 1.485 (RM2) QB -0.016

otes: UP is the urban population ratio account for monetization process. Where two annual models: AN is the model with narrow money. AB is the lodel with broad money, two quarterly models, QN is the model with narrow money, QB is the model with broad money. See Gu. (2004).

Tables 2.3 and 2.4 above, which reflect the findings of previous studies of money demand in China demonstrate that a long-run stable relationship exists between real balances and its determinants. These studies cover the sample period from 1952 to 2000 and monetary aggregates considered are narrow money mo and broad money ni2

though Tseng's work also considered narrow money m\. The long term interest rates are considered widely by the researchers as opportunity cost of holding money. These

(49)

studies also present evidences on the effect of the inflation variability on the demand for real money balances in China In terms of methodologies, some studies construct an error correction model (ECM) to evaluate the dynamic adjustment process of money demand in China in the reform period and a few studies employed Johansen (1988) and Johansen and Juselius (1990) cointegrating techniques to examine the long-run relation between demand for money and its determinants. There is no previous empirical evidence on studies of money demand in China by using Pesaran

el at

(2001) method There have been profound changes in money demand and monetary policy in the recent years in China There are still some unanswered questions which need further research Therefore, this study intends to make some contributions towards filling in the gaps and improve current literature.

2.5 Summary and Conclusions

In this chapter, major money demand theories, related issues and the more recent developments of the demand for money function have been reviewed The theoretical developments on money demand began in the classical tradition. Keynes and later economists developed a number of models to provide alternative explanations to confirm the formulation relating real money balances with real income and interest rates. The medium of exchange function of money led to the inventory theoretic formulation This emphasized the transactions costs under certainty and the precautionary demand for money models introduced the concept of uncertainty. In recent years, many empirical investigations on money demand function for the long-run and short-long-run in both developed and less developed countries provided more examples of the analysis of the money demand function in China The role of monetary policy and the financial system has seen significant changes since China's economic reform. The performance of Chinese monetary authorities accelerated the reforms of the banking system and the financial sector. It moved to a system of monetary control through indirect, market based instruments. There are a number of empirical studies about the money demand functions in China However, after

(50)

decades, many studies of money, prices and inflation, there still remain un-resolved issues. The further study and estimation of money demand is important for policy purposes.

To develop a better understanding of other influences, a review of the Chinese monetary system will be provided in next chapter which will include a brief survey of the factors influenced monetary control in China The discussion in the next chapter will concentrate on various aspects of monetary performance and indicate the possible factors affecting money demand in China. The link between money and economic activity in China will be investigated

(51)

Chapter 3 General Outline of the Chinese Monetary System

3.1 Introduction

The objective of this chapter is to offer a general discussion of the Chinese monetary system. From early discussions, it can be seen that the dramatic monetary expansion is a prominent feature of the Chinese economy after many reforms. The performance of real economic sectors, rapid economic growth, structural changes and monetary expansion have been of major interest to economists and provide useful explanations of money demand in China. Many economic factors have now become very important in the determination of money demand. In most societies, the ultimate purpose of economic growth is to raise the consumption and welfare of the people. For decades after economic reform, China evidenced dramatic changes in the savings and money demand of the Chinese household sector Chinese economic growth led to a tremendous rise in the standard of living of Chinese people and the average wage rate changed significantly. This chapter consists mainly of an analysis and discussion of the monetary system before and after economic reforms and contains an exploration of various factors which influence monetary control in China

This chapter is organized in four sections. The second section will provide a review of major economic reforms in Chinese monetary system The third section will analyse various determinants of money demand in China. The additional influences of other variables will be explored so as to provide more consideration regarding the nature of money demand behaviour in the Chinese economy. The fourth section will provide a summary and conclusion.

(52)

3.2 Economic Reforms

3.2.1 The Centralized Planning System before Reforms

To understand China's financial sector reforms, it is necessary to take a brief review of the prevailing conditions. Under the old monetary system, Chinese econ