Relevancy of Harrod-Domar Model in Nepalese Economy

Transcription

1 IOSR Journal of Economics and Finance (IOSR-JEF) e-issn: , p-issn: Volume 9, Issue 1 Ver. IV (Jan.- Feb.2018), PP Relevancy of Harrod-Domar Model in Nepalese Economy Rajendra Adhikari, Ph.D. Lecurer, Deparmen of Economics Mechi Muliple Campus, Bhadrapur Tribhuvan Universiy, Nepal Corresponding Auhor: Rajendra Adhikari, Ph.D. Absrac: Presen paper aims a examining he relevance of Harrod-Domar model in Nepal hrough economeric echniques by using he daa ses of saving, capial formaion/capial oupu raio and economic growh over he period 1974/ /17. The variables, under sudy are found o be coinegraed, and here appears uni-direcional causaliy running from saving o economic growh; and economic growh is found o be caused also by incremenal capial oupu raio as indicaed by Granger causaliy es. The auo regressive disribued lag models repor ha economic growh of Nepal during he sudy period is affeced posiively by growh rae of saving; and incremenal capial oupu raio has he negaive effec on economic growh, signifying he relevance of Harrod-Domar model in he economy of Nepal. This sudy plays crucial role in policy perspecive ha Governmen of Nepal should creae he environmen as o lead saving o increase, and saving should be convered ino he capial formaion hrough he developmen of capial marke. Addiionally, he capial subsiuion policy is indispensable o formulae hrough developmen of skilled labor force. Keywords: economic growh, ICOR, saving, Harrod-Domar model, ARDL model JEL classificaion: C22, E12, E21, E22, E Dae of Submission: Dae of accepance: I. Inroducion A number of growh heories are available in macroeconomic lieraure o analyze he deerminans of economic growh. According o classical economis Ricardo, labor is he only facor o deermine agriculural oupu, and hence oupu is aken as he funcion of labor force, ha is: Y = f(l). On he oher hand, Neo- Classical heory of Solow-Swan (1956) examined labor, capial and echnology as he deerminans of economic growh, ha is: Y = A f(l, K ). Keynes (1936) analyzed ha level of employmen and oupu are deermined by aggregae effecive demand and aggregae effecive demand is, in urn deermined by level of consumpion demand and invesmen demand. Invesmen is more powerful ool o deermine income in he Keynesian economy. Likewise, Samuelson (1948) and Hicks (1967) also emphasized on he role of invesmen o achieve high economic growh. However, Harrod (1939) and Domar (1946) declared ha economic growh is deermined by growh rae of saving and produciviy of capial. There is a long debae in macroeconomic field wheher saving and capial oupu raio (COR) are deerminans of economic growh. Differen views are found in he economic lieraure wheher saving and COR are he sources of economic growh. Verma (2007) argued ha rae of saving in an economy is he main booser of he seady-sae growh han wha invesmen does. The researchers such as Houakker (1965) and Modigliani (1970) explored empirically he relaionship beween saving and economic growh and concluded ha rae of saving causes higher seady sae growh ransiorily. Classical economiss believed ha saving plays crucial role ha i is necessary as well as sufficien condiion for he foundaion of invesmen. The saving via invesmen promoes economic growh (Najarzadeh, Reed & Tasan,2014). This view clearly implies ha saving plays influenial role for capial formaion o achieve seady-sae growh in he economy. Harrod (1960, 1963a) revealed ha he naural growh rae is exogenous in respec wih he rae of saving. According o him, he opimum saving which is essenial for implemening naural growh rae is an insrumen o formulae macroeconomic growh policy in he conex of he poswar welfare sae. The Harrod- Domar growh Theory is he inegraion of Harrod (1939) and Domar (1946) growh models based on he experience of capialis economies. This heory examines he requiremen of capial oupu raio and growh rae of saving for he seady-sae growh in he economy. This model laid is foundaion o neo-classical heory and uilized he Keynesian echnique of muliplier assuming no excess capaciy in he economy ha planned saving equals planned invesmen (Grabowski & Shields, 2000). According o his model, here is direc link beween economic growh and invesmen and saving is he main source for capial accumulaion. Harrod-Domar model assumes ha invesmen plays dual role: demand side role and supply side role. The Harrod-Domar model conspicuously emphasized he role of rae of saving and capial oupu raio for enhancing economic growh. DOI: / Page

2 Harrod-Domar model is one of he prominen models o analyze an economy s growh rae as a funcion of growh rae of saving and produciviy of capial. This model was independenly developed by Harrod in1939 and Domar in Though Harrod and Domar developed heir growh models independenly similar conclusions were drawn ha seady sae growh could be aained a Razor s Edge. Harrod-Domar model has he following proposiions: Saving is some proporion of naional income S = sy (1) Where, s = proporion of he income devoed for saving (0 < s < 1) Capial Oupu Raio (COR) is defined as he raio capial sock o oupu/income, ha is: K/Y = k (2) K/ Y = k k Y = K k Y = I (3) ( I = K) The change in sock of capial is he invesmen I. There is no excess capaciy in he economy, ha is: I = S (4) k Y = sy Y/Y = s/k (5) Growh rae of oupu is deermined joinly by COR and rae of growh of saving. Thus, i can be concluded ha economic growh is he funcion of growh rae of saving and COR, ha is: G = f(s, k) (6) Main objecive of he presen paper is o observe saving and invesmen/capial oupu raio as deerminans of economic growh of Nepal wih reference o Harrod-Domar model hough his model is proposed for developed counries. I is because relevance of Harrod-Domar model canno be undermined in he developing and underdeveloped counries like Nepal. Capial formaion is an indispensable facor for enhancing high economic growh for boh developed and underdeveloped counries; and saving is he main source of capial formaion. II. Lieraure Review A number of empirical sudies are available in economic lieraure regarding he deerminans of economic growh. Some sudies found saving and invesmen/cor as imporan deerminans of growh. Saving can be aken as one of he main sources of capial formaion in he economy. As rae of capial formaion increases, high economic growh can be aained in he economy. According o Abu (2010), increase in saving causes capial formaion o increase and hereby high economic growh can be aained. Jappelli and Pagano (1994), using Ordinary Leas Square, found ha higher growh rae of saving resuls higher economic growh. Mehanna (n.d.) aemped o explore he relaionship beween invesmen and economic growh in 80 developing counries for he period and esablished a srong posiive impac of invesmen on economic growh. In anoher sudy Moreira (2005) using Generalized Mehods of Momens found economic growh is caused by saving. In he sudy of Drisakis, Varelas, and Adamopoulos (2006) was observed he long run relaionship among expors, gross capial formaion, FDI and economic growh for Greece over he period The auhors employed mulivariae vecor auoregressive and found unidirecional causaliy running from gross fixed capial formaion and economic growh. Meha (2011) in he case of Indian economy examined he impac of capial formaion on economic growh using coinegraion es and vecor error correcion model (VECM) es over he period 1950/ /10. The auhor concluded ha here was shor run and long run relaionship beween capial formaion and economic growh ha capial formaion was found causing economic growh posiively. Adhikary (2011) examined he linkage beween foreign direc invesmen (FDI), rade openness, capial formaion and economic growh in he economy of Bangladesh over he period by employing coinegraion es and Granger causaliy es. The es proved economic growh in Bangladesh was caused posiively by capial formaion and FDI. Najarzadeh, Reed and Tasan (2014) by using he annual daa over he period and applying Auoregressive Disribued Lags (ARDL) and Error Correcion Modeling (ECM), found a posiive impac of saving on economic growh in he economy of Iran. Jagadeesh (2015) also examined he impac of saving on economic growh for Boswana applying ARDL Bound es for coinegraion and found Harrod-Domar being applied. Conversely, Nwanne (2014) in his sudy for Nigeria found adverse effec of saving on economic growh as repored by coinegraion es, VECM and Granger causaliy es, and hence inapplicabiliy of Harrod-Domar model. However, economic growh was posiively affeced by capial formaion in he sudy. Osundina and DOI: / Page

3 Osundina (2014) in heir sudy observed he posiive effec of saving and capial accumulaion on economic growh in Nigeria, and hence he Harrod-Domar model was applicable in Nigerian economy. III. Research Mehodology The presen sudy seeks o analyze wheher saving and invesmen/cor are he deerminans of Nepalese economic growh hrough he economeric mehodology. Augmened Dickey Fuller (ADF) uni roo es, Johansen sysem of coinegraion es, Granger causaliy es and auo regressive disribued lag (ARDL) models are he main economeric ools used in he presen sudy. The ADF uni roo es has been performed o idenify he saionariy or non-saionariy of he variables and use he models accordingly. Afer ADF uni roo es is performed, Johansen sysem of coinegraion es is applied o examine he long run relaionship among he variables. Once he variables under sudy are coinegraed, Granger causaliy es has been used o observe he causal linkage beween he variables. Finally, ARDL models are used o examine he impac of saving and COR on economic growh.. ADF Uni Roo Tes The esimable equaion proposed for ADF uni roo es is: p y y 1 jy j j1 (7) Where, y follows an AR (k) process. The consan erm is said o be drif erm. In he equaion (7), he noaion denoes he ime rend, and p is he lag lengh of he ime series and ε is defined as whie noise error erm. The null hypohesis for ADF uni roo es is The variable has a uni roo. Wheher he variable has uni roo or no can be confirmed wih he help of -saisic and corresponding probabiliy value. Johansen Sysem of Coinegraion Tes Johansen (1988) proposed a es for idenifying he long run relaionship beween and among he variables and number of coinegraing vecor. For his, Le X be a vecor of N ime series, each of which is I(1) variable, wih a vecor auoregressive (VAR) represenaion of order k, X 1 X 1... k X k (8) Where, i are (N N ) marices of unknown consans and is an independenly and idenically disribued (i.e. iid) n-dimensional vecor wih zero mean and variance marix e i.e. N (0, e ). The esimable equaion for he coinegraing relaionship is as follows: X 1 X 1... k 1X k1 X i 1 j, i 1...( k 1) j1 k 1 j j1 k Two likelihood raio ess have been proposed for he deerminaion of he number of coinegraed vecors, which are: Maximim Eigen value es, and Trace saisic. max T ln( 1 r 1) (10) Where, r 1... n are he n-r smalles squared canonical correlaions and T = he number of observaions. given by, race T ln(1 ) i (11) Granger Causaliy Tes Granger (1969) proposed Granger causaliy es ha can be applied o find he one way or wo way linkage beween he saionary variables of he same order. The esimable equaions for Granger causaliy es are: y = x = n n i=1 α i y i + j =1 β j x j + u 1 (12) n n i=1 γ i y i + j =1 δ j x j + u 2 (13) (9) DOI: / Page

4 Auo Regressive Disribued Lag (ARDL) Model The ARDL model has been recommend in he spiri ha variable y is affeced by no only he value of x a he same ime bu also by is lagged values plus some disurbance erm, x, x 1, x 2.., x k, ε.his can be wrien in he funcional form as: y = f(x, x 1, x 2.., x k, ε ) In linear form, y = α + β 0 x + β 1 x 1 + β 2 x β j x k + ε (14) The Ad Hoc approach popularized by Al (1942) and Tinbergen (1949) has been used o idenify he lags o be included in independen variable. Daa and Variables The presen sudy employs secondary daa on GDP, saving and gross fixed invesmen over he period 1974/ /17. The daa are aken from Economic Survey, Minisry of Finance of Nepal (various issues) and Cenral Bureau of Saisics. The nominal daa ses are convered ino real erms wih he help of GDP deflaor wih base year 2000/01. The GDP and saving in real erms have been ransformed ino logarihmic form and hese are differenced as per he requiremens. The variables used in he presen sudy are: Economic growh and saving. The GDP and gross domesic saving in real erms and ransformed ino logarihmic form are represened by Y and S heir firs differences by dy and ds respecively. The variables: dy and ds are aken as he proxy for economic growh and domesic saving of Nepalese economy respecively. Capial formaion. Gross invesmen/ Gross Capial Formaion in real erms convered ino logarihmic form is represened by I and is firs difference by di, which is aken as he proxy for capial formaion of Nepalese economy. Incremenal capial oupu raio. The produciviy of Nepalese capial is measured by incremenal capial oupu raio (ICOR). Fixed capial formaion and GDP in real erms are used o calculae ICOR. ICOR = (GFC /GDP )/(GDP GDP 1 /GDP ) (15) = GFC /GDP GDP 1 where, GFC = Gross Fixed Capial in Real Terms a ime GDP = Gross Domesic Produc a ime GDP 1 = Gross Domesic Produc a ime preceding IV. Daa Analysis And Discussion Of Resuls ADF Uni Roo Tes Firs i is imperaive o check he saionariy of he variables o apply he economeric models. Presen sudy has proposed ADF uni roo es for checking he saionariy. Hence, he resuls from ADF uni roo es have been presened hrough Table-1. Table 1 Augmened Dickey Fuller Uni Roo Tes on Variables Variable ADF es saisic Prob. value Lag lengh Tes Criical Values 1 % 5 % Y dy S ds I di ICOR From Table 1, i is observed ha he null hypoheses The variable has uni roo are no rejeced a level form for he variables: Y, S and I while considering consan as exogenous as repored by he -saisic a boh 5 percen and 1 percen level of significance and corresponding probabiliy values. On he oher hand, here is no reason o accep null hypoheses for all variables: dy, ds and di in heir firs differences including he variable ICOR. Thus, he variables excep ICOR are non-saionary a level; whereas hese variables including ICOR are saionary a heir firs differences. Johansen s Coinegraion Tes Before employing he Johansen s Coinegraion es, i is necessary o selec suiable lags o be used for he endogenous variable in regression. As repored by LR, SC and HQ crieria, lag 1 is suiable for each endogenous variable while applying Johansen s coinegraion es. The Johansen mehod of coinegraion is DOI: / Page

5 based on Maximum-Eigen and Trace saisic values. Table- 2 and Table-3 reveal he resuls from Johansen s coinegraion es among he variables: Y, S and I. Using firs order VAR of he variables under invesigaion, he hypoheses of r = 0 and r 1 are uniformly rejeced in favor of he alernaive hypohesis r = 1 and r = 2 respecively employing he maximum Eigen-value es as repored by 4 h column of Table-2, indicaing wo coinegraing vecors. Thus, on he basis of maximum Eigen-value es, he variables: Y, S and I are found o be coinegraed. Table 2 Tes Based on Maximum Eigen Value ( ) max H 0 H a λ i λ max 5% Criical Value Probabiliy r = 0 r = r 1 r = r 2 r = Table 3 Tes Based on Maximum Eigen Value ( max ) H 0 H a λ i λ race 5% Criical Value Probabiliy r = 0 r = r 1 r = r 2 r = Turning o he race es as repored by Table-3, he null hypohesis r 2 canno be rejeced while he hypohesis r = 0 and r 1 can be rejeced a 5 percen level of significance. The race saisics agains he criical values clearly indicae r = 2 coinegraing vecors among he variables. Thus, boh maximum Eigen value and Trace saisic specify ha here appears coinegraion among he variables under sudy. Granger Causaliy Tes Since he variables: Y, S and I are found o be coinegraed as repored by Johansen s coinegraion es, he causal linkage beween he variables are observed hrough Granger causaliy. Under Johansen s coinegraion es, he non-saionary daa ses of he variables: Y, S and I were used. However, under Granger causaliy es he saionary variables dy, ds and ICOR are used. The ICOR is used insead of I in Granger causaliy es in he spiri of Harrod-Domar model. Table-4 porrays he resuls from Granger causaliy es. In which, he null hypohesis ICOR does no Granger Cause dy a lag 1 and lag 3 is rejeced a 10 percen and 5 percen level of significance respecively as repored by F-saisics and corresponding probabiliy values. Similarly, he null hypohesis ds does no Granger Cause ICOR is also rejeced. However, oher null hypoheses are no rejeced even a 10 percen level of significance. The resuls imply ha here appears uni-direcional Granger causaliy running from ICOR o dy. There is also a lile economic significance beween ds and ICOR for uni-direcional Granger causaliy running from ds o ICOR. Table 4 Pair Wise Granger Causaliy Tes Endogenous variables: dy, ds and ICOR Null Hypohesis (H 0 :α = 0) Lags F-saisic Probabiliy ICOR does no Granger Cause dy dy does no Granger Cause ICOR ds does no Granger Cause dy dy does no Granger Cause ds ICOR does no Granger Cause ds ds does no Granger Cause ICOR ICOR does no Granger Cause dy dy does no Granger Cause ICOR ds does no Granger Cause dy dy does no Granger Cause ds ICOR does no Granger Cause ds ds does no Granger Cause ICOR The resuls from Granger causaliy could no esify in accordance wih he spiri of Harrod-Domar model. I is because he saving could no Granger cause economic growh hough capial oupu raio could cause economic growh. Hence, he presen sudy considers he raw daa of GDP and saving in real erms and heir firs differences o apply granger causaliy. Real GDP and saving in heir firs differences are represened DOI: / Page

6 by GDP r and S r respecively. However, incremenal capial oupu raio of fixed capial ICOR is same as before. Table-5 depics he resuls from Granger causaliy among he endogenous variables: GDP r, S r and ICOR. From his able i is observed ha he null hypohesis ICOR does no Granger Cause GDP r a lag 1 and lag 3 is srongly rejeced a 1 percen and 5 percen level of significance respecively. Whereas he null hypohesis GDP r does no Granger Cause ICOR is no rejeced a boh lag 1 and lag 3. This implies uni-direcional causaliy running from incremenal capial oupu raio o GDP. The saving is also found causing GDP a boh lag 1 and lag 3 a 5 percen level of significance. Whereas, GDP is no found causing saving a any lag. This also esifies uni-direcional causaliy running from saving o GDP. Finally, ICOR is found saving causing a 10 percen level of significance a lag 1. Table 5 Pair Wise Granger Causaliy Tes Endogenous variables: GDP r, S r, ICOR Null Hypohesis (H 0 :α = 0) Lags F-saisic Probabiliy ICOR does no Granger Cause GDP r GDP r does no Granger Cause ICOR S r does no Granger Cause GDP r GDP r does no Granger Cause S r ICOR does no Granger Cause S r S r does no Granger Cause ICOR ICOR does no Granger Cause GDP r GDP r does no Granger Cause ICOR S r does no Granger Cause GDP r GDP r does no Granger Cause S r ICOR does no Granger Cause S r S r does no Granger Cause ICOR Auoregressive Disribued Lag Model For ARDL model, he presen sudy has used he variables dy, ds, and ICOR in which dy is aken as dependen variable, and ds, ICOR independen variable. Firs dy is regressed on ds applying Ad Hoc approach. The dependen variable dy is regressed on he independen variable ds a lag 0 and 1. The coefficien of ds a lag 1 is no found saisically significan. Moreover, he coefficien of ds is found negaive, agains he priory assumpion. Hence, he dependen variable is regressed on wih independen variable ds dropping i a lag 1 wih ARDL model as represened by equaion (14). dy = α + βds (16) Afer replacing he parameers α and β by heir values, equaion (16) is convered as: dy = ds (17) [ ] [2.8176] 1 (0.0000) (0.0075) 2 In equaion (17), he coefficien of ds is posiive and significan a 0.01 as repored by -saisic and corresponding probabiliy value implying 1 percen increase in growh of saving causes economic growh o increase by %. Nex, when dy is regressed on ICOR a lag 0, he coefficien of ICOR is negaive and saisically significan. Similarly, dy is regressed on ICOR a lag 0 and lag 1. The coefficien of ICOR a lag 1 is saisically significan bu algebraic sign changes from negaive o posiive. I is no suppored by heory. The heory saes ha economic growh varies inversely wih ICOR. Hence, in accordance wih Ad Hoc approach, dy should be regressed on ICOR a lag 0 as represened by equaion (16). dy = γ + θ ICOR (18) Subsiuing he values of he parameers, equaion (18) can be expressed as: dy = ICOR (19) [ ] [ ] (0.0000) (0.0050) In equaion (19), he coefficien of ICOR is negaive and significan a 0.1 level. This implies ha higher incremenal capial oupu raio is accompanied by lower economic growh. The robusness of he esimaed OLS equaions (17) and (19) under ARDL model has been esified hrough applying serial correlaion es and heeroscedasiciy es. Moreover, he sabiliy of he esimaed coefficiens in he equaions is esified by Ramsey s RESET es. Breusch-Godfrey approach and Breusch- Pagan-Godfrey (B-P-G) approach are used o check he serial correlaion and heeroscedasiciy respecively in 1 Value in [ ] represens -saisic. 2 Value in ( ) represens probabiliy. DOI: / Page

7 he residuals of he esimaed OLS regression equaions. Table-6 and Table-7 presen he resuls from Breusch- Godfrey LM es, Breusch-Pagan-Godfrey heeroscedasiciy es and Ramsey s RESET es for equaion (17) and (19) respecively. From Table-6 i is observed ha F-saisic, value of 3 (T R 2 ) and probabiliy value of χ 2 (1) under Breusch-Godfrey Serial Correlaion LM es imply ha he null hypohesis of no serial correlaion is no rejeced. Hence, he residuals of esimaed equaion (17) are no serially correlaed. Likewise, he residuals are also free from heeroscedasiciy problem as accouned by F-saisic, value of (T R 2 ) and corresponding probabiliy value of χ 2 (1) under B-P-G. Finally, as repored by -saisic, F-saisic and Likelihood raio of Ramsey s RESET es, he esimaed equaion (17) is correcly specified bearing he propery of lineariy and hence i is sable equaion Table 6 Residuals Diagnosic and Sabiliy Tes of Esimaed Equaion of Equaion (17) Tes Breusch-Godfrey LM B-P-G Heeroscedasiciy Ramsey s RESET F-saisic Degree of freedom (1,39) (1,40) (1,39) Probabiliy T R Tes Probabiliy χ 2 (1) sa DF Probabiliy (39) Likelihood Raio Sa. DF Probabiliy (1) As in Table-6, Table-7 also implies ha he residuals of esimaed equaion (19) are no serially correlaed and free from heeroscedasiciy problem as specified Breusch-Godfrey serial correlaion es and B-P-G heeroscedasiciy es respecively. Finally, he esimaed equaion (19) is found o be correcly specified bearing he propery of lineariy and hence i is sable equaion. Table 7 Residuals Diagnosic And Sabiliy Tes Of Esimaed Equaion Of Equaion (19) Tes Breusch-Godfrey B-P-G Heeroscedasiciy Ramsey s RESET LM F-saisic Degree of freedom (1,39) (1,40) (1,39) Probabiliy T R Tes Probabiliy χ 2 (1) sa DF Probabiliy (39) Likelihood Raio Sa. DF Probabiliy (1) Conclusions and Policy Implicaions Presen sudy confirms he long run relaionship among he variables economic growh, saving and fixed capial formaion, and here appears uni-direcional Granger causaliy running from growh rae of saving o economic growh and incremenal capial oupu raio o economic growh in he economy of Nepal during he sudy period. Growh rae of saving has posiive impac on economic growh, which suppors Harrod-Domar model and oher subsequen researches. On he oher hand, he incremenal capial oupu raio has he negaive impac on economic growh. This also suppors Harrod-Domar model and oher successive sudies. Hence, Harrod- Domar model is found o be relevan in Nepalese economy. Presen sudy gives imporan feedback in policy perspecive ha economy should give focus on saving. However, saving alone canno bring posiive impac in he economy unless i is convered ino he effecive capial formaion. For his, Governmen should give aenion in he developmen of capial marke. The incremenal capial oupu raio for he economy during he sudy period is found o be 6.9 4, which is very high as compared o World s incremenal capial oupu raio around 3.0 for developed counries. So, his high incremenal capial oupu raio of Nepalese economy is essenial o reduce hrough subsiuion of capial by labor. However, Nepalese labor marke lacks skilled labor dominaed by unskilled and semi-skilled labor, no supporive o subsiue capial and enhance high economic growh. Tha is why, Governmen of Nepal should 3 T = Number of observaion, R 2 = Coefficien of deerminaion 4 The ICOR on average during he sudy period based on auhor s own calculaion is found o be 6.9. DOI: / Page

8 give special aenion in he developmen of skilled labor force in such a way as o subsiue capial, and hereby reduce high incremenal capial oupu raio. Reference [1]. Abu, N. (2010). Savings and economic growh nexus in Nigeria: Granger causaliy and coinegraion analysis. Review of Economic and Business Sudies, 3(1), [2]. Adhikary, B.K. (2011). FDI, rade openness, capial formaion, and economic growh in Bangladesh: a linkage analysis. Inernaional journal of Business Managemen, 6(1), [3]. Dickey, D. A., & Fuller, W. A. (1979). Disribuion of esimaes of auoregressive ime series wih uni roo. Journal of he American Saisical Associaion, [4]. Domar, E. (1946). Capial expansion, rae of growh and employmen. Economerica, 14, [5]. Drisakis, N., Varelas, E., & Adamopoulos, A. (2006). The Main Deerminans of Economic Growh: An Empirical Invesigaion wih Granger Causaliy Analysis for Greece. European Research Sudies,9 (3-4), [6]. Grabowski, G., & Shields, M. (2000). A dynamic, Keynesian model of developmen. Journal of Economic Developmen, 25(1), [7]. Granger, C. W. J. (1969). Invesigaing causal relaions by economeric models and cross-specral mehods. Economerica, 37(3), [8]. Gujrai, D. N., Porer D.C., and Gunasekar, S. (2009). Basic economerics. New Delhi, McGraw Hill Educaion Pv. Ld. [9]. Harrod, R. F. (1960). Second essay in dynamic heory. Economic Journal, 70, [10]. Harrod, R.F. (1963a). Themes in dynamic heory. Economic Journal, 73, [11]. Harrod, R. F. (1939). An essay in dynamic heory. Economic Journal,49: [12]. Hicks, J. R.(1967). Mr. Keynes and he Classics: A Suggesed Inerpreaion. In: Readings in Macroeconomics. Edied Mueller, M. G., (ed), New York [13]. Houhakker, H.S. (1965). On some deerminans of saving in developed and underdeveloped counries. In E. A. G. Robinson, (Ed.), Problems in economic developmen, (pp ). London, Macmillan. [14]. Jagadeesh, D. (2015). The impac of savings in economic growh: an empirical sudy based on Boswana. Inernaional Journal of Research in Business Sudies and Managemen, 2 (9), [15]. Jappelli, T., & Pagano, M. (1994). Savings, growh and liquidiy consrains. Quarerly Journal of Economics, 109, [16]. Johansen, S. (1988). Saisical analysis of coinegraion vecors. Journal of Economic Dynamics and Conrol, 12, [17]. Keynes, J. M. (1936). The general heory of employmen, ineres and money. London, Macmillan, (Reprined 2007). [18]. Mehanna, R.A. (nd). The emporal causaliy beween invesmen and growh in developing economies. Journal Of Business And Economics Research, 1(3), [19]. Meha, R.A. (2011). Shor-run and long-run relaionship beween capial formaion and economic growh in India. UMT, 19(2), [20]. (Minisry of Finance [ MOF], 2017). Economic survey (2016/17). Kahmandu. [21]. Modigliani, Franco, (1970). The life-cycle hypohesis and iner-counry differences in he saving raio. In W. A. Elis, M. FG. Sco, and J. N. Wolfe, (Eds.), Inducion, growh, and rade: essays in honor of Sir Roy Harrod (pp ). Oxford, Oxford Universiy Press. [22]. Moreira, S. B. (2005). Evaluaing he impac of foreign aid on economic growh: A cross-counry sudy. Journal of Economic Developmen, 30(2), [23]. Najarzadeh, R., Reed, M., & Tasan, M. (2014). Relaionship beween saving and economic growh: he case for Iran. Journal of inernaional Business and Economics. 2(4), [24]. Nwanne, T.F.I, (2014). Implicaion of saving and invesmen on economic growh in Nigeria. Inernaional Journal of Small Business and Enrepreneurship Research, 2 (4), [25]. Osundina, J.A., & Osundina, K.C. (2014). Capial Accumulaion, Saving and Economic growh of a Naion- Evidence from Nigeria. Global Journal of Inerdisciplinary Social Science, 3(3), [26]. Samuelson, P.A. (1948). Economics: an inroducory analysis. New York: McGraw-Hill Book Company, Inc. [27]. Solow, Rober M. (1956). A conribuion o he heory of economic growh. Quarerly Journal of Economics, 70, [28]. Sunny, I. O., & Osuagwn, N.C. (2016, April). Impac of capial formaion on he economic developmen of Nigeria. Fifh Inernaional Conference on Global Business, Economics, Finance and Social Sciences. GB16 Chennai Conference, Chennai-India. Rerieved from [29]. hp://globalbizresearch.org/chennai_conference_2016_april/docs/conf%20paper/1.%20global%20business,%20economics%20 &%20Susainabiliy/CF610.pdf [30]. Swan, T. (1956). Economic Growh and Capial Accumulaion. Economic Record. 32, [31]. Verma, R. (2007). Savings, invesmen and growh in India: an applicaion of he ARDL bounds esing approach. Souh Asia Economic Journal, 8(1), Rajendra Adhikari "Relevancy of Harrod-Domar Model in Nepalese Economy IOSR Journal of Economics and Finance (IOSR-JEF), vol. 9, no. 1, 2018, pp DOI: / Page