The Relaionship beween Money Demand and Ineres Raes: An Empirical Invesigaion in Sri Lanka R. C. P. Padmasiri 1 and O. G. Dayarana Banda 2 1 Economic Research Uni, Deparmen of Expor Agriculure 2 Deparmen of Economics and Saisics, Universiy of Peradeniya Keywords: Money Demand, Ineres Raes, Baumol-Tobin model Inroducion The ineres elasiciy of he demand for money is an imporan indicaor in considering an effecive ani-inflaionary moneary policy (Hossain and Younus, 2007). Idenifying he empirical relaionship beween money demand and ineres raes is imporan in effecive formulaion of moneary policy in Sri Lanka. The relaionship beween money demand and ineres raes has been explained by various heories including classical quaniy heory approach, Cambridge approach, Keynes s liquidiy preference heory, Baumol-Tobin money demand heory and Friedman s modern quaniy heory of money. In liquidiy preference heory, Keynes posulaed ha here are hree moives behind he demand for money: he ransacions moive, he precauionary moive, and he speculaive moive. Baumol (1952) and Tobin (1956) independenly developed similar models for demand for money, which demonsraed ha even money balances held for ransacions purposes are sensiive o he level of ineres raes. As ineres raes increase, he amoun of cash held for ransacion purposes will decline, which in urn means ha velociy will increase as ineres raes. A few sudies for money demand relaions have been done for Sri Lanka since 1990 (Wijewardena, 1985). However mos of hese sudies explain fiscal and moneary policy issues and behavior of ineres raes. There is a dearh of empirical sudies which examined he relaionship beween money demand and ineres raes using Baumol- Tobin Model for Sri Lanka. Some sudies examined he relaionship beween money demand and ineres raes using Keynesian heory. 11
Objecives The objecive of his sudy is o examine he naure and exen of he relaionship beween money demand and nominal ineres raes in Sri Lanka on he basis of Baumol-Tobin heory of he demand for money. The sudy aims o find ou he impac of nominal ineres raes on money demand using Baumol-Tobin Model. This paper has developed hypoheses o es he empirical relaionship and his hypoheses have been esed by empirical ess, Uni roo es, Coinegraion ess wih he ARDL approach o evaluae he overall impac of ineres raes on money demand in Sri Lanka. Furhermore his paper also invesigaes he implicaion of money demand and ineres raes relaionships on effecive implemenaion of moneary policy. Mehodology The Baumol-Tobin model specifies ha.. (1) rae - Real money demand; Cos of holding money; ineres Equaion (1) can be re-specified as a regression equaion as follows: l Md Md og = a + a og + a l log ( Q) + a 3 logi + U p 0 1 p 2 1... (2) Md Md l og = log of real money demand a ime, l og = Log of p p 1 lagged real money demand a ime, l og Q = Log of real GDP a ime, l ogi, ( ) = Log of nominal ineres raes a ime, U = disurbance erm 12
This sudy used quarerly daa from 1977, 1 s quarer o 2007, 4 h quarer. The daa was aken from annual repors of he Cenral Bank of Sri Lanka for differen years. Quarerly daa for M 1 is deflaed using he GDP deflaor o obain he real M 1. Six alernaive ineres raes are uilized in he esimaes. These variables have been convered ino a naural log form o sandardize empirical analysis. Since quarerly published daa for GDP and GDP deflaor are no available, he sudy ransformed annual ime series daa ino quarerly series using six disaggregaion echniques. (They are he NAIVE procedure, he LS procedure, he BFL-FD procedure, he BFL-SD procedure, he WS procedure and Chow-Lin procedure). Augmened Dickey Fuller (ADF) and he Phillips-Perron (PP) Tess are used o check saionariy of daa. The Auoregressive Disribued Lag (ARDL) approach (Pesaran and Shin, 1996) conains hree seps. The firs sep is ARDL bound es procedure. The second sep of he analysis is o esimae he coefficiens of he long-run relaionship. The hird is he esimaion of he shor-run elasiciy of he variables wih he error correcion represenaion of he ARDL model. According o his sudy, here are wo seps o analyze he relaionship beween money demand and ineres raes. These are he relaionship beween money demand and deposis raes and he relaionship beween money demand and lending raes. Resuls The resuls of he uni roo es show ha only he variable LIBCLR (Iner bank call loan raes) is saionary in level from (I(0)). SR (saving raes), FDR 3, FDR 6, FDR 12 (3 monh, 6 monh and 12 monh fixed deposi raes), TBR (Treasury bill raes), BR (Banking raes) and WLR (loan raes wih securiies) variables were saionary in he 1 s difference. According o he bound es resuls, here is a join long-run coinegraion relaionship among he variables of money demand and deposis ineres raes in Sri Lanka. There is a negaive relaionship among he saving deposi raes, fixed deposi raes and real money demand, while here are posiive relaionships among he real GDP, reasury bill raes and real money demand in long run. The resuls of his analysis saisfy he implicaion of Baumol-Tobin model. This model saes ha an increase in income leads o increase ransacion money demand and a decrease in ineres raes leads o increase in money demand. For he breviy of his paper, resuls ables are no included here. 13
This paper concludes ha here are join long-run coinegraion relaionship among he variables of money demand and lending raes in Sri Lanka. The esimaed coefficien of he long run relaionship shows ha real GDP has a posiive relaionship and significan impac on real money demand. According o Baumol-Tobin hypohesis he value of esimaed coefficien of equaion 2 should be 0.5 (negaive for ineres rae and posiive for GDP). However he esimaed coefficien in his sudy was 0.64 which did no agree wih Baumol-Tobin hypohesis. The lending raes variables are negaive and significan a he 1% level. A 1% increase in lending raes leads o approximaely 0.599 decrease in real money demand, all oher hings being equal. The lending ineres raes variable is suppor his hypohesis. The resuls of he shor-run dynamic coefficiens associaed wih he longrun relaionships obained from he ECM shows ha he speed of adjusmen coefficien is saisically significan. The coefficien suggess ha abou 70% of he disequilibria of he previous year s shock is adjused back o equilibrium. Conclusion and Policy Recommendaions There are negaive relaionships beween money demand and ineres raes while here is a posiive relaionship beween money demand and real GDP in he long-run in Sri Lanka. This analysis also concludes ha he behavior of money demand and ineres raes saisfies he Baumol-Tobin model. Increase in he GDP will increase he income level of people and due o heir ransacion and precauionary moives hey will uplif he demand for money. This sudy finds ha he coefficien beween money demand and income deposi ineres raes is 0.34 which is lower han he heoreical predicion 0.5. In he lending ineres rae analysis GDP coefficien is 0.64. The value of esimaed coefficien of saving rae is 0.26. Boh 6 monh and 12 monh fixed deposi raes are 0.22. On he oher hand he value of esimaed coefficien of lending rae is 0.59. Lending ineres rae allies wih he invenory heoreic model. I is clear ha in addiion o he ineres rae, here were oher facors which affeced he money demand srucure in Sri Lanka. Hence i is difficul o obain he values of coefficien and income and ineres elasiciy value proceeding o he Baumol-Tobin Model. 14
References Baumol, W. (1952) The Transacion Demand for Cash: An Invenory Theoreic Approach. Quarerly Journal of Economics. 66. p.545-556. Pesaran, M,. H., Shin, Y. and Smih, R. J. (1996) Tesing for he Exisence of a Long-Run Relaionship. DAE Working Paper. No. 9622. Deparmen of Applied Economics, Universiy of Cambridge. Available from: hp://ideas.repec.org/p/cam/camdae/9622.hml Tobin, J. (1956) The Ineres Elasiciy of Transacion Demand for Cash. The Review of Economics and Saisic. 38(3). p.241-247. Wijewardena, W. A. (1985) Ineres Elasiciy of Money Supply in Sri Lanka. Saff Sudies. Cenral Bank of Sri Lanka. 15(1). p.43-57. 15