GD 2243 Introduction to Econometrics (Assignment) Structural change in Export and economics growth: Analysis for Spain ( ) Econometric

Transcription

1 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) Economeric Projec Paper Tile: Srucural change in Expor and economics growh: Analysis for spain (198-1) *Shahrul Nizam Haron *Mohd Zahari Ma Amin *Mohd Fauzi Azmi 1

2 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) 1. Absrac The Spanish expor-led growh hypohesis is re-examined from he rade liberalizaion process iniiaed four decades ago. For his purpose boh he expor expansion and he progression from radiional expor o semi manufacures expor are aken ino consideraion. Alongside a feedback beween personal Disposable Income (PDI), Personal consumpion Expendiure (PCE), Corporae Profis Afer Tax (PROFITS) & Expor (X),are aken o be proved ha i also become a key facor for Spain s economic developmen.. Inroducion A disincive feaure of he Spanish economy has been he rapid and susained growh of heir expors during he las four decades. This phenomenon conracs wih he evoluion of expors in he preceden period in which he economy was characerized by a proecionis and auarkic siuaion. The process of growh in expor has is origin in an ouward-looking regime known as Sabilizaion and Liberalizaion Plan of Since hen, Spain has carried ou a coninued process of liberalizing is economy finally leading o he inegraion of he counry in he European Union in The effecs of a open rade regime and a policy of promoing expors on domesic real oupu was he fundamenal concern deal wih in Balaguer and Caavella-Jordai (1). A comparison beween he auarkic period and he rade liberalizaion period hrough he use of hisorical series for he las cenury was carried ou. In conras wih he exising evidence for he preceding period, he aggregae expor evaluaion was a significan facor in order o explain economic growh in he las four decades. Afer

3 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) highlighing he imporance of expor expansion in ha period, i would be ineresing o re-examine ha cycle aking ino consideraion some addiional facors such as PDI, PCE and PROFITS will be aken. Thus, bearing in mind ha a grea par of he hisorical ime series was esimaed, new daa series will be used now and, wha is more imporan, informaion abou he change in expor composiion will be analyzed. The empirical lieraure on GDP (Gross Domesic Producs) and macroeconomic insabiliy deals eiher wih ime series sudies based on Spain average over ime or averages of panel daa. We include relaion beween GDP growh and macroeconomic indicaors such as Personal Consumpion Expendiure (PCE), Personal Disposable Income (PDI), Corporae Profi afer Tax (profi), and Ne Corporae Dividend (dividend). GDP is he oal marke value of a counry s oupu. I is he marke value of all final goods and services produced wihin a given period of ime by facors or producion locaed wihin a counry. The cenraliy of GDP as a working concep canno be overesimaed. Jus as an individual firm needs o evaluae he success or failure of is operaions each year, so he economy a whole needs o assess iself. GDP, as a measure of he oal producion of an economy provides us wih a counry s economic repor card. Because GDP is such an imporan concep we need o ake some ime o explain exacly wha is definiion means. A large par of GDP consiss of personal consumpion expendiure (PCE). There are hree main caegories of consumer expendiure; here is durable goods, nondurable goods, and services. Durable goods, such as auomobiles, furniure s and household appliances, las relaively long ime. Nondurable goods, such as food, clohing, gasoline and cigarees, are used up fairly quickly. Paymen for services is hose hings ha we 3

4 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) buy ha do no involve he producion of physical iems. I is include expendiures for docors, lawyers and educaional insiuions. PDI or Personal Disposable Income is he subrac personal ax paymens from personal income, and personal income is he naional income accouns measure of he income received by persons from all sources. Naional income is a measure of income earned from curren producion of goods and services. For some purpose, however, i is useful o have a measure of income received by person regardless of sources. For example consumpion expendiure by households would be influenced by income. The relevan income concep would be one of all income received by persons. Also we would wan o measure of income needed o make ax paymens could no be used o finance consumpion. Alhough, GDP have been widely used in order o analysis he demand effec on he economic developmen, he use of his informaion may be insufficien when here exis a deep change in expor srucure. In fac, alongside he srucure change in expor composiion has also been frequenly suggesed as an imporance facor of a counry s economic growh. 4

5 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) 3. Lieraure review Researches showed repeaedly ha in he expendiure approach shows ha he Personal Consumpion Expendiure (PCE) is use in accouning GDP. I also includes invesmen expendiures (I), governmen expendiure (G) and ne expor (NX). To ge he GDP, hese four of spending mus be added up. Gross Domesic Produc or GDP for a given economy: PCE( + ) I ( + ) G( + ) NX = PQi i = GDP Naional Income represens he reurn o labor and he remaining one-quarer is known as he reurn o capial. Hisorically, he percenage of naional income allocaed beween labor and he oher facors of producion like corporae profi and ne ineres. The difference beween gross domesic produc and naional income is due o depreciaion expense, indirec business axes, and business ransfers o individuals. Profi can give effec in calculaing he GDP. Profis are he amoun of money a company has over from he sale of is producs afer i has paid for all he expense of producion. In such counries, profi belongs o he owners of companies or he sockholders of corporaions; one of he chief reasons for operaing a business is o make a profi. The adven of imperfec compeiion and sraegic rade heory has led o he invesigaion of pass-hough a indusry level. Such analyses examine if esimaes of incomplee pass-hough provide suppor for models of imperfec compeiion. Recen lieraure, which deals wih he impac of demand schedules in responding o GDP 5

6 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) changes, e.g. parial pass-hough occurs if demand becomes more elasic as price increase (Dornbusch, 1987; Marson, 199). Variaion across indusries in he exen of pass-hough has been aribued o facors such as he degree of indusry compeiion, produc, subsiuabiliy and relaive domesic and foreign shares in he marke. Honohan and Flynn (1986) found ha expors changes were fully refleced in aggregae domesic prices in he long run using an error-correcion framework. O connel and Frain (1989) examined if price and wages fully responsive o gross domesic produc changes. In he model, he change in he price of domesic goods (GDP deflaor) is explained by he personal disposable income, personal consumpion expendiure, and corporae profis afer ax and expor in semi manufacures and capial goods. Wagner (1883) was he firs who observed he diachronically endency o increase public spending, which he managed o combine wih he posiive raes of economic growh. While an economy is developing, public spending as a percenage of gross naional produc-increase a he expense of he coss of he privae secor. The fac ha he percenage of public spending on he naional producs increase wih ime means ha elasiciy consumpion for public goods is greaer han one. More specifically, one can examine if he expansion of public secor had a favorable effec on economic growh or if public spending, as dicaed by Keynesian Theory, simulaed economy growh. The previous secion proved exisence of a posiive relaionship beween PDI, PCE, PROFITS, and EXPORT, bu his does no allow conclusions o be drawn abou he direcion of he underlying causaliy. The daa employed can be used o es boh Wagner s heory, according o which he direcion of gross domesic produc increase owards public spending, and also Keynesian heory in which he direcion is reverse. 6

7 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) One of he approaches ha has suppored he expor-led-growh hypohesis and provided much influence in subsequen papers can be found in Feder (1983). On he one hand, he auhor argues ha developmen can be boosed hough resource reallocaion of less efficien secors owards hose hose are more efficien. On he hand, he, igher wih oher auhors, claims ha exernal demand-oriened secors experience more efficiency. Following he laer argumen, i can be inferred ha resource allocaion and producion efficiency can be improved hrough he dynamics expors. In order o capure he laer i would be sufficien o consider he aggregae evoluion of expors if heir srucure held relaively consan in ime. Excep for a few papers ha address he change in expor srucure such as Ghaak e al. (1997) and Amin Guiérrez de Piňeres and Ferrenino (1997), his is he implici assumpion in mos empirical lieraure on expor-led growh. Neverheless, i should be borne in mind ha he resource allocaion owards a new exernal demand may also conribue o increase a counry s general efficiency in a very significan manner. Thus, even hough aggregae expor expansion was sagnaed, he expor srucure change migh generae a beneficial effec on he economic aciviy. In his paper, he Spanish expor-led growh hypohesis is re-examined. For his purpose, no only expors and economic growh as such are aken under consideraion bu also he srucure change in PDI, PCE and PROFITS. The empirical approach used in his analysis allows, in a simple way, o sudy he GDP growh effec caused by he disribuion of resources among differen expor, PDI, PCE and PROFITS. 7

8 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) 4. Daa, Mehodology and Empirical Resuls The empirical analysis uses annual daa and he sample period spans from 198 o 1. This projec paper considers five variables; gross domesic produc (q), personal disposable income (PDI), personal consumpion expendiure (PCE), profis and expors (X). The X variable has been consruced by including semi-manufacured and capial goods divided all by oal expors. 4.1 Analysis The Gross Domesic Produc (GDP) deerminaion funcion of Personal Disposable Income (PDI), Personal Consumpion Expendiure (PCE), Corporae Profis afer Tax (PRO) and Expors (X) is defined as: q α PDI PCE PRO X + ε = + α1 + α + α 3 + α 4 where q, PDI, PCE, PRO and X sand for Gross Domesic Produc, Personal Disposable Income, Personal Consumpion Expendiure, Corporae Profis afer Tax and Expors respecively. α s are parameers o be esimaed. Meanwhile denoes ime-subscrip, whereas, ε is he error erm. 8

9 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) The below able shows our esimaion of our sudy: Dependen Variable: Y Mehod: Leas Squares Dae: 1/8/4 Time: 3:6 Sample: 198:1 1:4 Included observaions: 88 Variable Coefficien Sd. Error -Saisic Prob. C PDI PCE PRO X R-squared Mean dependen var Adjused R-squared S.D. dependen var S.E. of regression Akaike info crierion Sum squared resid Schwarz crierion Log likelihood F-saisic Durbin-Wason sa Prob(F-saisic). 4. Our Repor q α α PDI α PCE α PRO α X + ε = q = PDI PCE PRO X S.E =(99.833) (.97583) ( ) ( ) (.6465) -s = (7.5537) ( ) ( ) (7.173) (.35469) P.v= (.) (.341) (.) (.) (.9) R = = n = 88 Sample= 198:1-1:4 R F-saisic= As our resuls owards our sudy, afer an esimaion examinaion we find ha all he sign are mee wih our earlier expeced equaion. Therefore, we are surely o say ha all he sign of independen variables is in posiive sign. Mean ha increasing in independen variables will increase he value of GDP. 9

10 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) 4.3 Hypohesis Tes We have conduced an appropriae hypohesis es o es wheher our values of α 1 is significanly posiive a 5% level or no. The resul of our hypohesis es is as followed: Sep 1: Define he H and H A H : α < ( ˆα 1 is significan no posiive) H A : α ( ˆα 1 is significan posiive) Sep : Decision Rules choose α =.5 df = n-k-1 = = 83.5, 83 = Rejec H if calculae > Oherwise, do no rejec H Sep 3: Compue es- saisic = = ˆ α α SE( ˆ) α =

11 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) Sep 4: Decision Since calculaed =.9573 < criical = 1.671, H is no rejeced a 5% significan level. Sep 5: Conclusion Since we have rejeced H a 5% significan level, we have enough saisical evidence o infer ha ˆα 1 is no significan posiive We have conduced an appropriae hypohesis es o es wheher our values of ˆ α is significanly differen from zero a 1% level or no. The resul of our hypohesis es is as followed: Sep 1: Define he H and H A H : α ( ˆ α is no significan differen from zero) H A : α = ( ˆ α is significan differen from zero) Sep : Decision Rules choose α =.1 α / =.5 df = n-k-1 = = 83 11

12 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1).5, 83 = Rejec H if calculae > Oherwise, do no rejec H Sep 3: Compue es- saisic = = ˆ α α SE( ˆ) α = Sep 4: Decision Since calculaed = criical = 1.671, H is rejeced a 1% significan level. Sep 5: Conclusion Since we have rejeced H a 1% significan level, we have enough saisical evidence o infer ha ˆ α is significan differen from zero. 1

13 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) We have conduced an appropriae hypohesis es o es wheher or no Corporae Profis afer Tax (PRO) and Expors (X) should be excluded from he esimaion. The resul as i followed: Dependen Variable: Y Mehod: Leas Squares Dae: 1/11/4 Time: :37 Sample: 198:1 1:4 Included observaions: 88 Variable Coefficien Sd. Error -Saisic Prob. C PRO X R-squared Mean dependen var Adjused R-squared S.D. dependen var S.E. of regression Akaike info crierion Sum squared resid Schwarz crierion Log likelihood F-saisic Durbin-Wason sa Prob(F-saisic). q = + α1pro + α α X + ε q = PRO X S.E= (3.815)(.739) (.34313) -s= (16.73) ( ) ( ) P.v= (.) (.) (.) R = ˆ = n = 88 Sample= 198:1-1:4 R F-saisic=

14 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) F-saisic Tes: Sep 1: Define he H and H A H : ˆ ˆ α1 = α = ( PRO and H A : H is no rue. Sep : Decision Rules choose α =. 1 X should no be excluded) The 1% criical F- value for m (number of he consrains implied by H ) and (n-k-1) degrees of freedom is: m= ( PRO and F α, m, d. f = F.1,,83 = X ), d.f=n-k = =83 Rejec H if calculae F > 4.98 oherwise do no rejec H. Sep 3: Compue F- saisic F ( R = (1 R UR UR R R ) / m ) /( n k 1) where where R UR = R value of he unresriced regression model. R R = R value of he resriced regression model. F F ( ) / = (1.997) /83 = F =

15 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) Sep 4: Decision Since calculaed F = > criical F = 4.98, H is rejeced a 5% significan level. Sep 5: Conclusion Since we have rejeced H a 1% significan level, we have enough saisical evidence o infer ha in he model. PRO and X should be excluded 4.4 Inerpreaion of P value Noe ha P value is imporan o show wheher H is rejeced or no in addiion o prove our hypohesis es. If he P value is smaller han α i mean ha H o is rejecing. From he esimae able i proves ha: P value of GDP < α. < Therefore, H is rejecing. P value of PDI > α >.934 Therefore, H is no going o rejec. P value of PCE < α 15

16 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1). < Therefore, H is rejecing. P value of PRO < α 3. < Therefore, H is rejecing. P value of X < α 4.9 < Therefore, H is rejecing. 4.5 Sudy Inerpreaion Our inerpreaions of our sudy are as followed: There are negaive relaionship beween Personal Disposable Income (PDI) and Gross Domesic Produc (GDP). As an example, if PDI is increase wih 1 uni, i may decrease he value of GDP wih.934 unis. There are posiive relaionship beween Personal Consumpion Expendiure (PCE) and Gross Domesic Produc (GDP). As an example, if PCE is increase wih 1 uni, i may increase he value of GDP wih unis. There is posiive relaionship beween Corporae Profis afer Tax (PRO) and Gross Domesic Produc (GDP). As an example, if PRO is increase wih 1 uni, i may increase he value of GDP wih unis. There are posiive relaionship beween Expors (X) and Gross Domesic Produc (GDP). As an example, if X is increase wih 1 uni, i may increase he value of GDP wih unis. 16

17 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) 4.6 R Inerpreaion Noe ha our R is and nearly wih value of 1, i should be he bes o assume ha independen variables are efficienly will affec he GDP as a dependen variable. Therefore, R wih he value of means he variaion in independen variable can explained 99.7% of he variaion on y. R crierion is o find a subse model so ha adding more variables o he model will yield only small increase in R. In pracice, he bes model found by he R crierion will rarely be he model wih he larges R.In addiion, R does no accoun for he number of β parameers in he model. 4.7 Residuals Diagnosis: i. e agains ime gdp 8, 6, 4,,, -, -4, -6, -8, -1, -1, e agains ime ime e 17

18 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) The graf is flucuaing around he mean zero. I means ha i was saisfied he CLRM. As a maer of he fac hey exhibi he disinc behavior, i generally posiive, hen become negaive and urn up o posiive. The e is up down around zero. ii. e agains e -1 char 8, 6, 4,,, e -, e -4, -6, -8, -1, e-1-1, The ploing of e agains e -1 is posiive relaionship owards is oher. I was in sysemaic paern Mos e are bunched in he firs (norheas) ane hird (souhwes) quadrans. 18

19 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) iii. e agains GDP char e e GDP The graph shows ha here are no correlaions because i did no have any sign abou he sraigh line, curve or he sysemaic paern. I doesn have heeroscedasiciy and i covariance is zero. 19

20 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) iv. Hisogram of e As a resul from our hisogram or e, i show ha he residuals are no normally disribued. To es wheher i was normally disribued or no we used he Jarque-Bera es. Define H and H A H = normal disribued H = no normal disribued A If P value < Jarque-Bera, H is rejeced. So,.4967 < 6.478, H is rejeced. As a resul, he hisogram above is no normally disribued.

21 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) 4.8 Limiaions of Sudy Our poins of limiaion in our sudy are, in our model i should include ohers variables ha could give some ohers affec owards our GDP. As examples, i could useful if we include wih impors so ha i should give some differen sign o influence he GDP. The sign of impor should be negaive and i was negaive relaionship. We also should include our model wih Ne Corporae Dividends Paymens. The firs iems subraced from Spain naional income are he porion of he corporae profis iem in he naional income accouns ha are no paid ou as dividends includes corporae profis ax paymens, undisribued profis (reained earnings), and a valuaion adjusmen made by he Spain Commerce Deparmen. The deails of he valuaion adjusmen need no concern us here expec o recognize ha, because his adjusmen was made o he corporae profis enry in he naional income, bu did no acually affec profis ha can be paid ou o persons, i mus be subraced in compuing personal income. 1

22 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) 5. Conclusion This research paper focused on he role of expors as an engine of Spanish growh in he las four decades bu wih he pariculariy of aking under consideraion he exisence of a deep srucural change in expors. The moivaion of his research paper aims a examining wheher he economic growh has been spurred by eiher a general increase of expors or oher secors o hose more efficien or maybe by boh of hem. The inclusion of pas informaion on he expor variable improves he forecas for oupu. Neverheless, i is found a he same ime ha real expors are driven by economic growh, ha is, bidirecional causaliy exiss beween expors and oupu. This resul is compaible wih he fac ha he resource movemen from one secor o anoher has aimed a saisfying he exernal demand of goods wih greaer added value. This resul shows evidence regarding he imporan of ha variable in he explanaion of he exporled growh process. Therefore, i can be said ha even in circumsances in which he evoluion of expors is kep consan, i is possible ha he resource allocaion owards more compeiive secors generaed by exernal dynamics may cause significan growh raes. In his sense, no only expor has a fundamenal role in he economic growh, bu also i has o be suppored by he suiable indusrial policy which favors he mobiliy of resources owards he efficien expor secors.

23 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) 6. References Akaike, H. (1969) Fiing auoregressive models for predicion, Annals of he Insiue of Saisical Mahemaics, 1, Amin Guiérrze de Piñeres, S. and Ferranino, M. (1997) Expor diversificaion and srucural dynamics in he growh process: he case of Chile, Journal of Developmen Economics, 5, Balaguer, J. and Canavella-Jordà, M. (1) Examining he expor-led growh hypohesis for Spain in he las cenury, Applied Economics Leer, 8, Dickey, D. A. and Fuller, W. A. (1979) Disribuion of he esimaors for auoregressive ime series wih a uni roo, Journal of he American Saisical Associaion, 74, Dickey, D. A. and Fuller, W. A. (1981) Likelihood raio saisics for auoregressive ime series wih a uni roo, Economerica, 49, Feder, G. (1983) On expors and economic growh, Journal of Developmen Economics, 1, Ghaak, S., Milner, C. and Ukulu, U. (1997) Expors, expor composiion and growh: coinegraion and causaliy evidence for Malaysia, Applied Economics, 9, Greenaway, D. and Sapsford, D. (1994) Wha does liberalisaion do for expors and growh?, Welwirschafliches Archives, 13, Johansen, S. (1988) Saisical analysis of coinegraion vecors, Journal of Economic Dynamics and Conrol, 1,

24 GD 43 Inroducion o Economerics (Assignmen) Srucural change in Expor and economics growh: Analysis for Spain (198-1) Moreno, L. (1997) The deerminans of Spanish indusrial expors o he European Union, Applied Economics, 9, Oserwald-Lenum, M. (199) A noe wih quaniles of he asympoic disribuion of he maximum likelihood coinegraion rank es saisics, Oxford Bullein of Economics and Saisics, 54, Phillips, P. and Perron, P. (1988) Tesing for a uni roo in ime series regression, Biomerica, 75, Romer, P. M. (1986) Increasing reurns and long run growh, Journal of Poliical Economy, 94, Shan, J. and Sun F. (1998) On he expor-led growh hypohesis: he economeric evidence from China, Applied Economics, 3, William Mendenhall & Terry Sincich, Regression Analysis Sixh Ediion, Pearson Prenice Hall, 3. 4