College of Business, Hospitality &Tourism Studies

Transcription

1 College of Business, Hospialiy &Tourism Sudies Working Paper Series No. 05/13 Tile Does Purchasing Power Pariy Theory hold in Fiji? Auhors T. K. Jayaraman, Fiji Naional Universiy and Chee-Keong Choong Universii Tunku Abdul Rahman, Kampar, Malaysia Dae: June This paper presens work in progress in he College of Business, Hospialiy and Tourism Sudies, FNU. Commens are welcomed from all sakeholders and should be addressed o he corresponding auhor. Copyrigh 2013 by he auhor. All righs reserved. 1

2 Does Purchasing Power Pariy Theory hold in Fiji? T. K. Jayaraman* School of Economics, Business and Finance, College Of Business, Hospialiy and Tourism Sudies, Fiji Naional Universiy, Nasinu Campus, Kings Road, Nasinu, Fiji. Chee-Keong Choong Cenre for Economic Sudies, Faculy of Business and Finance, Universii Tunku Abdul Rahman, Jalan Universii, Bandar Bara, Malaysia Absrac Fiji, one of he six Pacific island counries wih independen currencies, has a fixed exchange rae regime ever since is independence in This paper invesigaes wheher he purchasing pariy power heory holds in he case of Fiji. The findings, based on JJ coinegraion and bounds esing procedures indicae only a weak presence of a long-run relaionship among he exchange rae, domesic and foreign prices. Apparenly, here are oher facors a work, including produciviy differenials, in deermining he exchange rae. Keywords: Purchasing Power Pariy, coinegraion, Bounds es, symmery resricion, proporionaliy resricion * Corresponding auhor. 1

3 I. Inroducion Fiji among he 14 Pacific island counries is one of he six counries wih independen currencies and hence has an exchange rae of is own as a policy ool. However, i has adoped a fixed exchange rae regime. The domesic currency, Fiji dollar is pegged o a baske of currencies of major rading parners, namely Ausralia, European Union, Japan, New Zealand, Unied Kingdom and Unied Saes and he weighs adoped are based on averages of Fiji s direcion of rade. The fixed exchange rae regime has provided a nominal anchor for Fiji, a small open economy jus as oher PICs wih he main moneary policy objecive o mainain sabiliy of he domesic and exernal value of he currency. Such a regime has been srongly recommended by economiss for delivering low inflaion (Corden 1994, 2002) as well as for self-imposed fiscal and moneary discipline (Fischer 2001). Targeing he nominal exchange rae has is own problems. The problems arise ou of misalignmens of exchange rae, which are caused by foreign and domesic shocks. These shocks are refleced in changes in world and domesic price levels. If misalignmen is subsanial, adjusmens are made hrough deliberae devaluaion of he currency. Fiji had is adjusmens are made hrough hree devaluaions 1987, 2000 and In beween he periods, he exchange rae was allowed o flucuae wihin a narrow band of 0.07 percen. The heory of purchasing power pariy (PPP) heory is based on law of one price ha under he assumpions of absence of or low ransporaion coss and absence of rade barriers, prices in wo counries of goods of similar qualiy should be idenical when expressed in erms of he same currency. The PPP heory saes ha exchange raes beween any wo counries will adjus over ime o reflec changes in heir respecive price levels (Mishkin 2012). This paper invesigaes wheher PPP heory holds in Fiji. The paper is organized along he following lines. The second secion presens a brief review of heoreical and empirical legislaure; he hird secion deals wih daa and modelling and mehodology; he fourh secion presens he resuls; and he fifh and final secion presens a summary and conclusions wih policy implicaions. II. A Brief Review of Theoreical and Empirical Sudies The PPP heory, which is supposed o be a long run phenomenon, is he cornersone of he moneary models of exchange rae deerminaion (Dornbusch 1976, Musa 1982). Empirical sudies noe deviaions from PPP occur in he shor-run (Anorou e al 2005, Dornbusch 1980, Frenkel 1978). However, no all sudies focusing on he long run validiy of PPP heory had come ou wih unanimous resuls. While, for example, Abuaf and Jorian (1990) and Meef and Rogoff (1988), found evidence of PPP heory holding in he long run, Cooper (1994) and Ahking (1997) obained evidence o he conrary. The urning poin in he invesigaion came abou wih he finding ha nominal exchange rae has uni roo, which would indicae ha nominal exchange rae follows a random walk and is impac is no mean revering (Anorou e al 2005). Considering changes in nominal exchange 2

4 rae are likely o be permanen and hence, he long run PPP heory could only be confirmed, if he uni roo ess are underaken and exisence of uni roos is rejeced (Adler and Lehman 1983; Manzur and Ariff 1995). Various auhors used differen ess. While Mansur and Ariff (1995) and Whi (1992) used Sims ess, Ahking (1997) used Bayesian uni roo approach. Huang and Yang (1996) employed wo-sep approach of Engle-Granger (1987), and Johansen (1988, 1992) maximum likelihood procedure as well as Mone-Carlo simulaions and obained differen resuls. Lee (1999) used a generalised error correcion model for 13 Asia pacific counries. As regards differen exchange rae regimes, sudies by Derodan, e al (1999) found ha in he long run PPP is valid under a floaing exchange rae regime. In heir sudy of exchange raes of 11 developing counries, Anorou e al (2005) conduced uni roo ess using boh Augmened Dickey Fuller ess and Phillipe-Perron procedures o deermine he order for inegraion. Furher, hey adoped a dynamic error correcion model (DECM) o examine he exisence of long run PPP because of he special propery ha DECM relaxes he resricions implici in he radiional uni roos procedures and reas boh nominal exchange rae and price raio as endogenous variables. Empirical sudies done in respec of boh developed and developing counries a differen imes and for differen periods have shown ha he PPP heory has lile predicive power in he shor run, despie he fac ha heory provides some guidance o movemen in he exchange raes over a period. Tha is, policy makers are aware ha in he long-run if a given counry s price level has been rising relaively higher level o ha of anoher counry s price level, is currency depreciaes. Daa III. Daa, Modelling and Mehodology The daa employed for he sudy on Fiji are quarerly daa covering a period of hree decades: 1980Q1-2011Q4. The nominal exchange raes are unis of domesic currency per uni of he US dollar. The price level daa series of Fiji is consumer price index, while he foreign price level daa series is he consumer price index of he US. The daa series are drawn from Inernaional Financial Saisics, an Inernaional Moneary Fund publicaion (IMF 2012). Table 1 presens a summary of daa employed in he sudy. Modelling: PPP Relaionship The PPP heory proposes ha he exchange rae relies on relaive price levels, as follows: * r p p (1) where r is he log of he nominal exchange rae (unis of domesic currency per uni of he * US dollar), p and p are he logs of domesic and foreign prices, respecively. Equaion (1) suggess ha weak PPP relaionship exiss if here is evidence of coinegraion among r, p 3

5 * and p. If we impose he symmery condiion 1 2 on prices, hen his resricion shows a new PPP relaionship. Table 1: Fiji: Nominal Exchange Rae and Inflaion Year Exchange Rae Unis of F$ per US$ (F$/US$) Inflaion in Fiji (%) Inflaion in USA (%) (Ave) (Ave) (Ave) Source: IMF (2012). * r ( p p ) (2) 0 Since we impose he symmery condiion in Equaion (2), his equaion can only have a single coinegraing vecor. In our furher esing, if we impose he proporionaliy condiion on he relaive price coefficien in Equaion (2), hen we can examine he exisence of srong PPP relaionship in PICs. In his sudy, we invesigae he validiy of boh symmery and join symmery and proporionaliy assumpions in PICs. IV Empirical Resuls A necessary condiion for PPP hypohesis o hold is ha he relaive price is saionary; oherwise, disequilibrium or dispariy from PPP would be persisen. Therefore, we firs employ four ypes of individual uni roo es saisics, namely Augmened Dickey-Fuller es saisic, Phillips-Perron es saisic, Ng-Perron es saisics (Mza) saisic and Kwiakowski-Phillips-Schmid-Shin (KPSS) es saisic. Based on he four es saisics repored in Table 2, i is concluded ha exchange rae (ER), domesic price (DCPI) and foreign price (FCPI) are non-saionary a level; bu hey are saionary afer firs differencing, ha is, hey are I(1) sochasic processes. Since he exchange rae is saionary a I(1), his implies ha exchange rae exhibis a high degree of persisence and does no suppor he mean reversion hypohesis. This finding is consisen wih some exising sudies such as Papell (1997), O'Connell (1998), Cerrao and Saranis (2002) and Coakley, e al. (2005). On he basis of hese uni roo es resuls, we coninue o examine he validiy of PPP hypohesis in Fiji by using Johansen and Juslius s (1990) coinegraion es and Bounds esing approach. The resuls are repored in Tables 3 and 4. The es resuls show ha he null hypohesis of no coinegraion can be rejeced, using Johansen and Juslius s (JJ) coinegraion es. For example, based on he Trace and Maximum Eigenvalue saisics, we rejeced he null hypohesis of r=0 a he 1% significance level, which signifies ha PPP did hold for Fiji in he long-run. The validiy of PPP hypohesis is furher confirmed by he 4

6 Bound esing approach es (Table 4). Hence, we conclude ha exchange rae, domesic and foreign prices are coinegraed in he long-run. Table 2: Uni roo ess Uni ess roo Exchange Rae (ER) Domesic Price (DCPI) Foreign Price (FCPI) Level 1 s Difference Level 1 s Difference Level 1 s Difference Augmened Dickey-Fuller es saisic ** *** ** Phillips- Perron es saisic ** *** ** Ng-Perron es saisics (Mza) *** ** *** Kwiakowski- Phillips- Schmid-Shin es saisic 0.216** ** ** Noes: ** and *** indicae significance a 5% and 1% levels, respecively. Noes: Table 3: Johansen and Juslius s mulivariae coinegraion es Counry Trace Saisic Maximum Eigenvalue Saisic r= *** *** r= r= The criical values for he race es a he 95% significance level are (r=0); (r=1); 3.76 (r=2) (r=0); (r=1); 6.65 (r=2) a 99%. The criical values for he maximum eigenvalue es a he 95% significance level are (r=0); (r=1); 3.76 (r=2) (r=0); (r=1); 6.65 (r=2) a 99%. 5

7 ** and *** Significan a he 5 and 1 percen levels, respecively. Table 4. Resuls of Bound Tess Dependen Variable Compued F-saisic ER 6.292*** DCPI FCPI Pesaran, e al. (2001) a Narayan (2005) b Criical Value Lower bound value Upper bound value Lower bound value Upper bound value 1 per cen per cen per cen a Criical values are obained from Pesaran e al. (2001), Table CI(iii) Case III: Unresriced inercep and no rend, p b Criical values are obained from Narayan (2005), Table case III: unresriced inercep and no rend, p. 10. *, ** and *** indicae significance a 10%, 5% and 1% levels, respecively. Equaions (4) and (5) are wih he esimaed coefficiens (domesic and foreign prices) of he coinegraing ess for Fiji. I is found ha he esimaed coefficiens of boh JJ coinegraion and Bounds ess have he expeced signs and hey are significan a he 5 percen level. JJ coinegraion es: ER DCPI * 2.728FCPI ** 0.038TREND ** (1.874) ( 1.968) (2.184) (4) Bounds es (ARDL model): ER DCPI *** 1.768FCPI ** 0.020TREND * (2.838) ( 2.615) (2.017) (5) Since he resuls of boh JJ coinegraion es and bounds es suppor he weak PPP hypohesis, which are in conras wih he findings for he PPP repored by uni roo ess of exchange rae. I is likely hese conrasing findings migh be due o join symmery and proporionaliy resricions imposed on he uni roo ess of he exchange rae (Frankel and Rose 1996). To examine he robusness and validiy of he resuls, we apply he likelihood raio es (Johansen, 1995) o examine he validiy of he join symmery and proporionaliy resricion, namely This indicaes ha one of he coinegraing vecors is (1, 1, 1). The resuls are repored in Table 5. We rejec he join symmery and proporionaliy resricions for Fiji a 1% significance level. The rejecion of hese condiions indicaes ha he join symmery and proporionaliy resricion may be oo resricive. Our finding is 6

8 consisen wih some sudies applying ime series coinegraion ess such as Cheung and Lai (1993). Table 5 Coinegraion es: LR-es Counry Join Symmery and Proporionaliy Resricion (β1=1 β2= 1) Join Symmery Resricion (β1=β2) Likelihood Raio Tes 8.341*** [0.0005] 9.112*** [0.0035] Noes: Numbers in brackes are p-values. ** and *** Significan a he 5 and 1 percen levels, respecively. We also examined he validiy of he symmery condiion, 1 2. The resuls are repored in Table 5 (las column), which rejec he null hypohesis 1 2. Alhough resuls sugges ha domesic and foreign prices are crucial deerminans of he exchange rae in he long run, he esimaes do no necessarily comply wih he resricive condiions (join symmery and proporionaliy resricions) imposed by he srong PPP heory. In conclusion, our evidence ends o suppor only he weak form of he long-run PPP relaionship in Fiji. V. Summary and Conclusions Our sudy examines he validiy of PPP heory in regard o he exchange rae of Fiji for he period of 1980Q1-2011Q4. We find supporing evidence only for a weak form for long-run PPP hypohesis, since he join symmery and proporionaliy resricions are rejeced. There are a large number of sudies and plausible explanaions for such a finding. For example, ariffs, nonariff barriers, ransporaion coss, inflaion differenials in raded and non-raded goods (he Harrod Balassa Samuelson effec) amongs ohers which influence exchange rae. The findings, hus, open up more research possibiliies. The policy implicaions are clear: Even hough here is evidence in suppor of he validiy of PPP heory in Fiji, hey are no sufficien o saisfy he PPP resricions on he size of he parameers. Policy makers should carefully monior oher deerminans besides exchange rae policy in sabilizing domesic prices, monioring high inflaion raes and enhancing expor compeiiveness. References Abuaf, N. and Jorion, P. (1990). Purchasing power pariy in he long run. Journal of Finance, 45, Adler, M. and Lehman, B. (1983). Deviaions from purchasing power pariy in he long run. Journal of Finance, 38, Ahking, F. (1997) Tesing Long-Run Purchasing Power Pariy wih a Bayesian Uni Roo Approach: The Experience of Canada in he 1950s. Applied Economics, 29,

9 Anorou, E. Habu, B. and Yusuf, A. (2002). Purchasing Power Pariy: Evidence from Developing Counries, Inernaional Advances in Economic Research, 8(2), Cerrao, M. and Saranis, N. (2002). The cross secional dependence puzzle, Discussion Paper No Cenre for Inernaional Capial Markes London Guildhall Universiy: Deparmen of Economics. Cheung, Y.W. and Lai, K. S. (1993). Long-run purchasing power pariy during he recen floa. Journal of Inernaional Economics, 34, Coakley, J., Kellard, N. and Snaih, S. (2005). The PPP debae: Price maers! Economics Leers, 88, Cooper, J. (1994), Purchasing Power Pariy: A Coinegraion Analysis of he Ausralian, New Zealand and Singaporean currencies. Applied Economics Leers, 1, Corden, W.M. (1994). Exchange rae policy: choices for developing counries in C. Bah and C.H. Wong (eds). Approaches o Exchange Rae Policy: Choice for Developing and Transiional Economies, Washingon, D.C.: Inernaional Moneary Fund: Corden, W.M. (2002).Too Sensraional: On he Choice of Exchange rae regimes, Cambridge, Massachuses: MIT Press. Derodan,, K. Jung, C. Boyd, R. (1999).Tesing he Law of One Price Under he Fixed and Flexible Exchange Rae Sysem. Applied Economics Leers, 6, Dornbusch, R. (1976). Expecaions and exchange rae dynamics. Journal of Poliical Economy, 84, Dornbusch, R. (1980). Purchasing Power Pariy, in J. Eawell, M. Migare, and P. Newman (Eds.), The New Palgrave Dicionary, New York: Sockon Press. Engle, R.F. and Granger, C.W.J. (1987). Co-inegraion and error correcion: Represenaion, esimaion and esing. Economerica, 55(2), Fischer, S. (2001). Exchange Rae Regimes in he Bipolar View Correc? Journal of Economic Perspecives, 15, Frankel, J.A. and Rose, A.K. (1996). A panel projec on purchasing power pariy: Mean reversion wihin and beween counries. Journal of Inernaional Economics, 40, Frenkel, J.A. (1978). Purchasing power pariy: docrinal perspecive and evidence from he 1920 s. Journal of Inernaional Economics, 8, Huang, B. and Yang, C. (1996). Long-Run Purchasing Power Pariy Revisied: A Mone Carlo Simulaion. Applied Economics, 28, IMF (2012). Inernaional Financial Saisics, CD Rom November 2012, Washingon, D.C.: IMF. 8

10 Johansen, S. (1988). Saisical analysis of coinegraion vecors. Journal of Economic dynamics and Conrol, 12, Johansen, S. (1992). A represenaion of vecor auoregressive processes inegraed of order 2. Economeric Theory, 8, Johansen, S., Juselius, K., Maximum likelihood esimaion and inference on coinegraion wih applicaions o he demand for money. Oxford Bullein of Economics and Saisics, 52, Lee, D. (1999) Purchasing Power Pariy and Dynamic Error Correcion Evidence from Asia Pacific Economies. Inernaional Review of Economics and Finance, 8(1), Manzur, M. and Ariff, M. (1995) Purchasing Power Pariy: New Mehods and Exensions. Applied Finance Economics, 5(1), Meef, R.A. and Rogoff, K. (1988). Was i real: The Exchange Rae-Ineres Differenial Relaion Over he Modern Floaing-Rae Period. Journal of Finance, 43, Mishkin, F.S. (2012). Economics of Money, Banking and Financial Markes, (10 h Ediion), New York: Addison Wesley. Musa, M. (1982). Model of Exchange Rae Dynamics. Journal of Poliical Economy, 90(2), Narayan, P.K. (2005). The saving and invesmen nexus for China: evidence from coinegraion ess. Applied Economics, 37(17), O'Connell, P. (1998). The overvaluaion of purchasing power pariy. Journal of Inernaional Economics, 44, Papell, D.H. (1997). Searching for saionary: purchasing power pariy under he curren floa. Journal of Inernaional Economics, 43, Pesaran, M.H., Shin, Y. and Smih, R. (2001). Bounds esing approaches o he analysis of level relaionships. Journal of Applied Economerics, 16(3), Whi, J.A. (1992) The Long-Run Behavior of he Real Exchange Rae: A Reconsideraion. Journal of Money, Credi, and Banking, 24(1),