Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Bank Credit and Economic Growth in Nepal: An Empirical Analysis# Neelam Timsina Abstract This study examines the impact of commercial bank credit to the private sector on the economic growth in Nepal from supply side perspectives. The study has applied Johansen co-integration approach and Error Correction Model using the time series data for the period of 1975-2014. The empirical results show that bank credit to the private sector has positive effects on the economic growth in Nepal only in the long run. Nevertheless, in the short run, it has been observed a feedback effect from economic growth to private sector credit. More specifically, the growth in real private sector credit by 1 percentage point contributes to an increase in real gross domestic product by 0.40 percentage point in the long run. The empirical results imply that, policy makers should focus on long run policies to promote economic growth – development of modern banking sector, efficient financial market and infrastructure so as to increase the private sector credit which is instrumental to promote growth in the long run. Key Words: Economic Growth, Bank Credit, Co-integration JEL Classification: E23, G21, C32 # The earlier version of this paper is available at www.nrb.org.np under NRB Working Paper series, NRB-WP-22, 2014. Director, Nepal Rastra Bank, Research Department, Central Office, Baluwatar, Kathmandu, Nepal. Email: [email protected] Acknowledgement: I would like to express my sincere thanks to Mr. Guna Raj Bhatta, Assistant Director of Monetary Division, Research Department for his valuable inputs in setting methodological frameworks. 2 NRB ECONOMIC REVIEW I. INTRODUCTION Economic growth is one of the major objectives of macroeconomic policy. It is the crucial means of uplifting living standards as well as achieving economic development. Economists define economic growth from various perspectives. Some economists view that it is an increase in the national income or the level of production of goods and services by a country over a certain period of time. Generally economic growth is defined as an increase in gross domestic product (GDP). Therefore, GDP is considered as proxy of economic growth in the study. Credit is the aggregate amount of funds provided by commercial banks to individuals, business organizations/industries and government for consumption and investment purposes. Individuals obtain credit for both consumption and investment purposes, business organizations/industries borrow loans to invest in plant and machinery where as government borrows loans to spend for recurrent as well as capital expenditure purposes. More specifically, credit is understood as the provision of resources such as granting a loan by the creditor/lender to the debtor/borrower where the debtor does not reimburse the lender immediately, thereby generating a debt, and instead arranges either to repay or return those resources at a later date (Mishra at all, 2009). Credit is considered as a key to economic growth especially in developing countries as it lubricates the economy. Therefore, the role of bank credit in economic growth has been accepted by many researchers as various economic agents are able to invest money in various investment opportunities. Economic growth has been one of the major macroeconomic objectives of the government of Nepal. Nepal Rastra Bank (NRB) considers that monetary policy should also support growth. NRB always directs commercial banks to flow their credit to productive sector. Credit channel of monetary policy is considered very important and effective in Nepal. In this channel, money supply is expected to affect real variables through the means of bank balance sheet and availability of credit. A large body of evidence suggests that financial sector development plays a huge role in economic development. Okwo (2012) examined the effect of bank credit to private sector on economic growth in Nigeria and found that bank credit to private sectors has a statistical strong positive relationship with GDP as expected. Bank credit to private sector promotes economic growth through capital accumulation and technological progress by increasing the savings rate, mobilizing and pooling savings, producing information about investment, facilitating and encouraging the inflows of foreign capital, as well as optimizing the allocation of capital (World Bank, 2013). One of the major indicators for measuring financial development of a country is private sector credit to GDP ratio. The role of credit provided by banks to private sector is considered more efficient to support economic growth rather than the credit provided to government. Therefore, private sector credit is taken as the proxy of bank credit here in the study. Bank Credit and Economic Growth in Nepal: An Empirical Analysis 3 Bank credit has significant role in economic growth. Especially in developing countries like Nepal, it caters resource need for economic growth. Hence, NRB and the government have adopted many policies and programs to increase economic growth through the use of bank credit. NRB has been playing a leading role to determine the proportion of bank loans and advances to productive sectors (agriculture, energy, tourism, industry). The main objective of this provision is to stimulate economic growth in the country. However, the relationship between private sector credit and economic growth has not yet been assessed properly in the Nepalese context. In this regard, this study attempts to fulfill the gap. Therefore the main objective of this study is to examine the effects of commercial bank credit to private sector on economic growth from supply side perspectives as well as to suggest ways of improving bank credit to private sector so as to achieve better economic growth in Nepal. The rest of the paper is structured as follows. The second section describes the theoretical framework. The third section reviews the related literatures. The fourth section presents the status and trend of bank credit to private sector. The fifth section presents the data and methodology and the sixth section shows results of the study. The last section concludes the study. II. THEORETICAL FRAMEWORK Bank credit contributes to economic growth in several ways. For example, credit is an important link in money transmission; it finances production, consumption, and capital formation, which in turn affect economic activity. The transmission mechanism of monetary policy can be strengthened, and the monetary policy objectives attained to a large extent, if the financial system is well-operated and regulated. Credit extended to the private sector in an environment of banking discipline will be instrumental in tapping the productive potentialities and development prospects of the economy. It thereby ushers to inculcate economic growth, generating employment opportunities, and strengthening the competitiveness of the economy (Basyal, 2009). It is a means of generating self employment opportunities, strengthening informal activities. Ademu (2006) explained that credit can be used to prevent economic activity from total collapse in the event of natural disaster such as flood, draught, disease or fire. By using credit, farmers increase agricultural production by investing money in seed, fertilizers, tractor, and pump set etc. Industrial production can be increased by using credit. Moreover service sectors need credit to flourish. In fact all components of GDP need credit to grow. In performing the financial intermediation role, it has been argued that by virtue of this function that banks generate economic growth by providing needed resources for real investment (Kinnon, 1973).Sustainable economic growth depends on the ability to raise the rates of accumulation of physical and human capital to use the resulting productive assets more efficiently and to ensure the access of the whole population to these assets (Fitzgerald, 2006). This is possible only by having access to bank credit. Banks perform the act of financial intermediation that collect money from the surplus sector in the form of deposits and lend it to various sectors of the economy leading to economic growth. Extension of credit is one of the major functions of banking institutions. 4 NRB ECONOMIC REVIEW Neo-classical growth theory states that labor and capital are the major factors of production. I.e. Y = f (K,L) where Y denotes aggregate output, K denotes aggregate capital stock, and L is the labor force. If technology and human capital are added, then equation becomes : Yi,t = AKα(Lh)1−α. ( Mankiw, Romer, and Weil, 1992). Bank credit facilitates to acquire more capital in this production function. When a new technology is available, the labor and capital need to be adjusted to maintain growth equilibrium. To acquire new technology and thus to increase total factor productivity, the role of credit provided by banks would be of immense help. Private sector credit fosters growth through increasing investment and an efficiency/productivity. The capital accumulation channel is particularly important for underdeveloped and emerging countries, while the productivity channel is mostly relevant for advanced countries. In standard neoclassical theories investment-savings is the engine of growth. In these theories, there are no capital market frictions and thus financial intermediation is not explicitly modeled. However these models assume that savings translate directly to investment and thus one could argue that finance affects growth primarily through capital deepening (investment) (Papaioannou,2007). A different class of theoretical models argues that financial development may foster growth by raising human capital accumulation. In Galor and Zeira (1993) model income inequality and credit market frictions impede growth, since not all individuals can invest in education. They argue thus that financial intermediation can spur growth (and eventually decrease inequality) by fostering human capital accumulation. III. LITERATURE REVIEW A large body of literature is available on the extensive empirical work with regard to the nexus between finance and economic growth. Largely, this task has been performed by King and Levine (1993) and Levine(1997). They showed that financial development has predictive power for future growth and interpret this finding as evidence for a casual relationship that run from financial development to economic growth. Although the literature regarding the role of financial development on economic growth has grown rapidly in recent time, studies that examine bank credit or access to private sector credit and how it affects the economic performance of industries or economic sectors have been overshadowed by the increasing number of empirical studies that largely focus on financial development and growth. Nevertheless, private sector credit is one of the important indicators of financial development. Therefore the literatures on financegrowth nexus are helpful for the study on bank credit –growth nexus. King and Levine (1993) provided the evidence that financial sector proxied by the ratio of bank credit granted to the private sector to GDP, affects economic growth both through the improvement of investment productivity ( better allocation of capital) and through higher investment level. Financial system could impact positively on real economic performance by affecting the composition of savings (Bencivenga and Smith, 1991), providing information (Greenwood and Jovanovic, 1990), and affecting the scope for credit rationing (Boyd and Smith, 1997). Bank Credit and Economic Growth in Nepal: An Empirical Analysis 5 Schumpeter (1971) identified banks' role in facilitating technological innovation through their intermediary role. He argued that the role of bank of channelizing resources from surplus sector to deficient sector plays crucial role in promoting growth. Several others such as (Kinnon 1973), Shaw (1973), Adekanye (1986), Fry (1988), King and Levine(1993), and Adeniyi (2006)have focused on the significance of private sector credit to economic growth. Similarly studies by Gurley and Shaw (1967), Goldsmith (1969), Jayaratne and Strahan (1996), Kashyap and Stein(2000), Beck et al.(2000), Beck et al (2003), Driscoll (2004) etc, found that financial development can foster economic growth by raising saving, improving allocative efficiency of loanable funds, and promoting capital accumulation. In their opinion, well developed financial markets are necessary for the overall economic advancement of less developed and the emerging economies. King and Levine (1993a) said that the banking sector's development in Europe was not only correlated with economic growth but was also a cause of long-term growth. Adekanye (1986) argued, by providing credit; banks are rendering a great social service which leads to increase in production, capital investment and improving living standard. Akpansung (2011) by using two stage least square, found that private sector credit impacts positively on economic growth in Nigeria. However, lending interest rate impedes economic growth. Moreover, that paper recommends the need for more financial market development that favours more credit to private sector with minimal interest rate to stimulate economic growth. A low rate of expansion of the credit volume is not only a symptom of weak economic growth, but can also be one of its causes (Bundesbank, 2005). Bayoumi and Melander (2008) found that a 2.5 percent reduction in overall credit caused a reduction in the level of GDP by around 1.5 percent. Dey and Flaherty (2005) used a two stage regression model to examine the impact of bank credit and stock market liquidity on GDP growth. They found that banking development is significant determinant of GDP growth. However Koivu (2002) found that growth in credit has not always been sustainable and in some cases it may have led to a decline in growth rates. Murty at al (2012) by using multivariate Johansen co integration approach, examined the long run impact of the bank credit on economic growth of Ethiopia and found that bank credit to the private sector affected economic growth through its role in efficient allocation of resources and domestic capital accumulation. Thus the policy makers should focus attention on long run policies to promote economic growth – the creation of modern banking sector so as to enhance domestic investment, which is instrumental to increasing output per capita and hence promoting economic growth in the long run. Ugoani (2013) examined the power of bank credit on economic growth in Nigerian perspective and found that bank credit has significant relationship with economic growth and socio-infrastructural development. He argues that on the one hand, bank credit is the oil on the wheel of economic growth. On the other, there is strong empirical evidence that the development of sound financial markets and institutions has significant relationships with long term economic growth. Financial sector plays a key role in channeling savings into productive investment especially in the formal sectors of the economy. The banking sector is well recognized as 6 NRB ECONOMIC REVIEW a key conduit of financial intermediation in the economy. Access to credit enhances the productive capacity of businesses (Were Nzomi and Rutto, 2012). Private sector credit is considered as proxy of bank credit in many international studies. Beck and Levine (2001) measure bank development as bank credit to private sector divided by GDP. Also the endogenous growth theory sheds light on the role of finance on economic growth. Solow (1956, 1957) in his two factor neoclassical growth model incorporated the role of credit. The supply of credit, both in terms of volume and in terms of credit standards applied on loans to enterprises, have significant effects on real economic activity. In other words, a change in loan growth has a positive and statistically significant effect on GDP (Cappiello, Kadareja, and Sarensen, 2010). In the same way that financial services increase income of poor by expanding the supply of financial services which can be accessed by the poor. It will generate income growth for the poor, thus having a direct impact on poverty reduction (Jalilian & Kirkpatrick, 2001). The role of private sector credit as transmission channel of monetary policy cannot be ignored. Monetary policy may affect real economic activity, and ultimately inflation, via its impact on the banking sector credit through a number of transmission channels (Brunner and Meltzer, 1963 and Bernake, 1983). IV. THE STATUS AND TREND OF BANK CREDIT TO PRIVATE SECTOR IN NEPAL A significant portion of credit in Nepal is provided through the banking system, though there are some institutions such as savings and credit cooperative societies, finance companies, development banks and micro finance institutions. However, availability of time series data for the latter institutions is very limited. Therefore, in this study, private sector credit provided by commercial bank only is taken into consideration. The ratio of bank credit to private sector and nominal GDP has not increased steadily over the study period in Nepal. Before 1980s, such ratio was very low. The financial sector of Nepal witnessed revolutionary changes in 1980s. A broad based program of reforms was launched since 1980s. The banking sector in Nepal had been transformed from a highly dominated inefficient state-owned sector to a dynamic private sector. NRB and the Government of Nepal took a number of steps to further enhance the pace of this transformation process of the development of financial sector in the country. A substantial increase in private sector credit to GDP ratio took place only after implementation of financial liberalization in early 2000s.The last five years from 2009 to 2014 recorded ratios of 44.0 percent, 41.7 percent, 40.3 percent, 41.2 percent , 45 percent and 47 percent respectively. Bank Credit and Economic Growth in Nepal: An Empirical Analysis 7 50.0 45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 percent Chart 1 Private Sector Credit to GDP Ratio (Nominal Term) year Source: Data from Quarterly Economic Bulletin July,2013 (NRB) and figure from Author's calculation. It has recorded significant increases only in 2009, 2013 and 2014. The stringent measures adopted by NRB to limit the real estate and margin lending loan of the banking sector, short term nature of loans, low growth of remittances and resulting liquidity crunch, low growth of government expenditure etc. were accountable for the low private sector credit to GDP growth in year 2010 and 2011. In Nepal, economic growth rate is low compared to other developing countries (Annex 5). Real GDP growth rate is only 3.6 percent in 2013 and 5.2 percent in 2014. Average real GDP growth rate is 4.08 percent over the last twenty years. One of its main reasons is low private sector credit growth. Private sector credit (provided by commercial banks) is only 47 percent in 2014 but it was 21 percent of nominal GDP on average over the sample period. Therefore, to boost the economic growth of the country, private sector credit should be increased to productive sector. 8 NRB ECONOMIC REVIEW Chart 2 Private Sector Credit and GDP Growth (Nominal Term) 50.0 40.0 percent 30.0 20.0 Pvct 10.0 GDP -10.0 -20.0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 0.0 year Chart 2 shows that there is positive relationship between nominal private sector credit and nominal GDP. As private sector credit increased, GDP also increased, but at a lower rate. Except some years their growth rate also seems to move in the same direction. Chart 3 also shows that real private sector credit has positive relationship with real GDP during the period 1975-2014. Chart 3 Private Sector Credit and GDP Growth (Real Term) 40.0 30.0 percent 20.0 GDP 10.0 PVCT -10.0 -20.0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 0.0 year Up to now, the Nepalese banking system does not seem to have the investment bank for long term loans, venture capital for viable projects, which in turn leads to inadequate economic growth. Moreover, Nepal is experiencing still a significant credit transaction in informal sector despite government's efforts to channel credit to the productive sector Bank Credit and Economic Growth in Nepal: An Empirical Analysis 9 through commercial banks, development banks, finance companies and micro credit development banks. Sector wise Distribution of Bank Credit to Private Sector Sector wise distribution of private sector credit has great meaning to economic growth. Generally it is assumed that credit to productive sector caters economic growth where as credit to consumption sector can not contribute in this regard. In Nepal, of the total private sector credit provided by commercial banks in July 2014, wholesale and retail trade constituted 22 percent followed by production (20 percent), others (16 percent), construction (10 percent), finance, insurance & fixed assets (8 percent), service industries (8 percent), transportation, communication & public services (4 percent), agriculture sector (4 percent) and metal, machinery, tools & fitting (1 percent). Though agriculture constitutes 32.40 percent of GDP in Nepal, only 4 percent of total private sector credit was provided in this sector as of 2014. Though Nepal Rastra Bank has made policy provision that banks should provide at least 20 percent of their total loan to productive sector and at least 12 percent of their loan to agriculture, energy and tourism sector, this percent was merely 6 percent only. Commercial banks seem to provide only 5 percent of their total loan to consumption sector (very small figure in total) which helps to establish the relationship between bank credit and economic growth. Chart 4 Sectorwise Distribution of Bank Credit (July, 2014) Agriculture Mines 4% 0% Local Government 0% Consumable Loan 5% Others 16% Productions 20% Service Industries 8% Finance, Insurance & Fixed Assets 8% Construction 10% Wholesaler & Retailers 22% Transportation Communications & Public Services 4% Metal, Machinary, Tools and Fitting 1% Transportation Equipment Production & Fitting 2% V. DATA AND METHODOLOGY Secondary data that captured the whole population of all commercial banks in Nepal for the period 1975 –2014 are used in the study. Secondary data are gathered from Quarterly Economic Bulletin and Quarterly Financial Indicators (NRB). In Nepal, the bank credit is allocated to both the public and private sector of the economy. However, private sector credit is considered to be more effective to stimulate economic 10 NRB ECONOMIC REVIEW growth. Several studies such as Beck et al (2005), Levine(2002), Odedokun (1998), King and Levine(1993), Boyreau-Debray (2003), Liang (2007) and Crowley (2008) have suggested that bank credit to the private sector is more significant for economic activities than bank credit to the public sector. Therefore, in this study bank credit to private sector is taken as appropriate variable. Since the paper attempts to assess relationship between private sector credit and economic growth, variables such as bank credit to the private sector (lnrpvct), economic growth (lnrgdp) are taken as the main variables. Government expenditure (lnrgexp) and interest rate (ir) have been included in the study as control variables. Mathematically, GDP = f( pvct, gexp, ir). Murty et al (2012) suggested this type of variables to examine the effects of private sector credit on growth. Also, Okyo et al (2012) emphasized the interest rate and inflation as control variables in their study. The study has applied co-integration approach error correction model and granger causality for the empirical examination of the relationship between the private sector credit and economic growth. Empirical Model Firstly, we form the following regression equation to estimate the effects of private sector credit on real gross domestic product. Government expenditure and interest rate are taken as control variables. Where, lnrgdpt = α0 + α1lnrpvctt+ α2lnrgexp t + α3ir t+ μ rgdp = real gross domestic product rpvct = private sector credit provided by banks in real terms ir = interest rate rgexp = real government expenditures Unit root test and co-integration tests should be performed first before performing the ordinary least square method. If the variables are found I(I) and co-integrated to each other, then co-integration and error correction test should be run. If the variables are found I(I) but no co-integration found between the variables of interest, then it is better to run OLS. Therefore in this case we first perform the unit root test and co-integration test. Unit Root Tests The pre-requisite of co-integration test is the stationarity test of each individual time series over the sample period. Co-integration analysis has increasingly become the appropriate methodological approach for analyzing time series data containing stochastic trends. Hence before turning to the analysis of the long run relationships between the variables, we should check for the unit root properties of the data, as non stationary behavior is a prerequisite for including them in the co-integration analysis. Bank Credit and Economic Growth in Nepal: An Empirical Analysis 11 Table 1 ADF Test Results (Unit Root Tests) Intercept Variables lnrgdp lnrpvct lnrgexp ir Intercept and Trend Level t-stat p-value First Difference t-stat p-value Level t-stat p-value First Difference t-stat p-value -0.2191 0.9275 -6.2461 0.0000 --1.5065 0.8103 -6.1634 0.0000 -0.5403 0.8722 -4.6902 0.0005 -3.0366 0.1361 -4.738 0.0026 -1.6497 0.4483 -6.1444 0.0000 3.6743 0.0363 -5.9839 0.0001 -1.3167 0.6119 -4.6169 0.0006 -2.7912 0.2091 -4.5509 0.0043 Source: Author's computation ADF statistics in the above table shows that all the variables included found to be I(1) with one variable lnrgexp having deterministic trend. Hence, although it can be modeled at first difference with OLS and extracting trend and cycles for trend-stationary variables, this is possible only if variables are not co-integrated. The Johansen co-integration test has been carried out as follows to identify whether there exists a co-integrated relationships. Table 2 Johansen's Cointegration Test (LNRGDP LNRPVCT LNGEXP IR ) Trace Statistics Hypothesized No. of CE(s) None* Trace Statistic Maximum Eigenvalue P-value 58.3249 0.05 Critical Value 47.85613 0.05 Critical Value 27.58434 P-value 0.0039 MaxEigen Statistic 26.1242 At most 1* 32.2006 29.79707 0.0259 17.1239 21.13162 0.1662 At most 2 15.0767 15.49471 0.0577 14.1104 14.26460 0.0528 At most 3 0.9663 3.841466 0.3256 0.9663 3.841466 0.3256 0.0759 * denotes the rejection of null hypothesis at 5 percent level of significance. Trace test indicates 3 cointegrating equations at o.05 level whereas maximum Eigen Value test indicates 1 cointegrating equation at o.05 level. The Johansson cointegration tests for cointegration shows conflicting results with trace test and maximum eigenvalues test. The trace test indicates a 2 cointegration relation, however, eigenvalue shows none. Hence it is desired to test the cointegration relation only with the variables of interest, credit flow and its impact to GDP. The test results for cointegration between economic growth ( lnrgdp) and private sector credit (lnrpvct) are as follows: 12 NRB ECONOMIC REVIEW Table 3 Johansen's Cointegration Test (LNRGDP LNRPVCT) Trace. Maximum Eigenvalue Hypothesized No. of CE(s) Trace Statistic 0.05 Critical Value P-value MaxEigen Statistic 0.05 Critical Value P-value None* 15.6358 15.4947 0.0476 15.5774 14.2646 0.0308 At most 1 0.05843 3.8414 0.8090 0.0584 3.8414 0.8090 Both the results consistently show a single co-integration relationship, at a highest level of confidence which further confirms the earlier test of co-integration with four variables. Hence, it can be decided that there exists a long term relationship between private sector credit and country real GDP. Causality Test A number of studies have been carried out to examine the direction of causality between bank credit and economic growth. Mishra at al (2009) examines the direction of causality that runs between credit market development and economic growth in India through the application of Granger Causality Test and find that credit market development spurs economic growth. Mukhokadhya and Pradhan (2010) assess the causal relationship between financial development and economic growth of seven Asian developing countries and concluded that no general consensus can be drawn about finance growth relationship in developing countries. Odedokun (1989) find the case of unidirectional causality from the real sector to the financial sector and concludes that money is causally prior to income. Here in the study, Granger Causality Test has been conducted to find out the direction of causality between the bank credit and economic growth. The results show evidence of unidirectional casual relationship from GDP to private sector credit (annex 3, annex 4). With different lag structure at 2 and 5 lags, the estimated F-stat suggests that private sector lending does not Granger causes the real GDP but the other way is true. Hence, the preliminary relationship is something different than expected. Nepalese economic growth is led by feedback effect from the growth, rather than multiplier effect that of investment. Based on these results, a bivariate error correction model is being estimated in the following sections. Using the representation theorem of Engle and Granger (1987) to establish a link between the co-integration and Error Correction Model (ECM), we can show the long-run relation as: t ln rgdpt 1 ln rpvctt ….. (1) Bank Credit and Economic Growth in Nepal: An Empirical Analysis 13 By transforming the equation 1, we can develop an error correction model as: l l ln rgdpt nrgdp ln rgdp t 1 a1h ln rgdpt h b1h ln rpvctt h uln rgdp.t …… (2) h1 h1 l l h1 h1 ln rpvctt ln rpvct ln rpvct t 1 a2 h ln rpvctt h b2 h ln rgdpt h uln pvct .t ..... (3) Where, u ln rgdpt and u rpvct.t are stationary white noise processes for some number of lags l. The coefficients in the co-integration equation give the estimated long-run relationship among the variables and coefficients on the error correction model (ECM) describe how deviations from that long-run relationship affect the changes on them in next period. The parameters ln rgdp and ln rpvct of the equation (2) and (3) measure the speed of adjustment of private sector credit and economic growth respectively towards the longrun equilibrium VI. RESULTS Estimates of the equation (1) for co-integration t ln rgdp t 1 8.153 0.410 ln rpvctt 1 (0.0087)* ….. (4) The error correction estimates of equation (2) and (3) have been presented as follows: ln rgdpt 0.056 0.0553ˆt 1 0.0149 ln rpvctt 1 0.004 ln rpvctt 2 0.0878 ln rgdpt 1 0.215 ln rgdpt 2 . (0.012) (0.07) (0.046) (0.005) (0.188) ….. (5) (0.199) Adj. R2 = 0.054, F-Stat = 0.35 ln rpvctt 0.181 0.916ˆt 1 0.534 ln rpvctt 1 0.0786 ln rpvctt 2 1.39 ln rgdpt 1 1.59 ln rgdpt 2 (0.0329) * (0.215) * (0.125) * (0.134) (0.505) * (0.53) * … (6) Adj. R2 = 0.59, F-Stat = 9.05 Note: * Significant at 5 percent or lower level. Values in parenthesis indicates the standard errors of the respective estimates The estimates of the model show interesting results. By rearranging the estimates of cointegration equation (4), it can be inferred that one percentage point increase in the real private sector credit may cause the increase in real GDP by 0.41 percentage points over the long run equilibrium relationships. Nevertheless, the short run equilibrium effects are more induced by the feedback effects of GDP growth to the private sector lending, not from the private sector lending to GDP growth, which is against our hypothesis. All the coefficients of error correction estimates with the dependent variable as ∆lnrgdpt are found to be insignificant including ln rgdp , very low adjusted R2 value (0.054) and insignificant F-Stat (0.35). In the contrary, with the dependent variable ∆lnrpvctt, the 14 NRB ECONOMIC REVIEW error correction estimate is significant showing that the estimate of ln rpvct is 0.916; significant at 5 percent or lower level. It includes relatively high R2 value (0.593) and FStat (9.05). It indicates that any deviation in real GDP in any given time will affect the real private sector lending by 0.916 in the next period and the effect of such deviation in private sector credit to the real GDP is almost zero. Hence, the finding is that, although there is a long-run relationship can be observed from private sector lending to overall growth of the economy, there is no immediate multiplier effect from investment to growth and such a long-run relationship became only possible through feedback effects. Diagnostics tests shows estimations are valid. Residuals Plots move around zero (annex 6). LM Test for Autocorrelation shows no serial correlation in error terms (annex 7).Since p-value is higher while we include up to three lags, we do not reject null, in favor of this, there is no serial correlation in residuals (annex 7). Spikes of the correlogram graphs are also found to be within the bands (annex 9) and also, all inverse roots of AR Polynomial lie inside the circle (annex 10 ). VII. CONCLUSION Credit is an important link in monetary transmission as it finances production, consumption, and capital formation, which in turn affect economic growth. Especially in developing countries like Nepal, it caters resource need for economic growth. NRB and the government have adopted many policies and programs to increase economic growth through the use of bank credit. However, the relationship between private sector credit and economic growth has not yet been assessed properly in the Nepalese context. Applying Johansen co-integration approach and estimating Error Correction Model, the study found that the banks credit to private sector has positive impact on economic growth only in the long run. Nonetheless the short run equilibrium effects are more induced by the feedback effects of GDP growth to the private sector lending, not from the private sector lending to GDP growth, which is against the proposed hypothesis. The empirical results imply that, policy makers should focus attention on long run policies to promote economic growth such as development of modern banking sector, efficient financial market, infrastructures so as to increase private sector credit which is instrumental to promote growth in the long run. ***** Bank Credit and Economic Growth in Nepal: An Empirical Analysis 15 REFERENCES Adekanye F. 1986. "Elements of Banking in Nigeria." F and A Publishers, Lagos, Nigeria. Ademu, W. A. 2006. "The Informal Sector and Employment Generation." Selected for the 2006 annual conference of the Nigeria Economic Society. Adeniyi O. M. 2006. "Bank Credit and Economic Development in Nigeria." A Case study of deposit money banks, University of Jos. Akpansung A. O. and S. J. Babalola. 2011. "Banking Sector Credit and Economic Growth in Nigeria: An Empirical Investigation." CBN Journal of Applied Statistics, 2(2) : 51-62. Alade, S. O., M. Ajayi, C. I. Enendu and E. Idowu. 2003. The Supply of and demand for loanable funds in CBN contemporary economic policy issues in Nigeria, edited by O.J Nnanna S.O. Alade and F.O. Odoko, Garki, Abuja CBN. Basyal, T. R. 2009. "Role of Finance - Nepal's Relative Position in the Private Sector Credit." Socio-Economic Development Panorama, 1(4). Bayoumi, T. and O. Melander. 2008. "Credit Matters: Empirical Evidence on US Macro Financial Linkages." IMF Working Paper, No 08/169. Beck, T. and R. Levine. 2001. Stock Markets, Banks and Growth Correlation or Causality, Washington DC: The World Bank. Beck, T., R. Levine and N. Loayza. 2000. "Finance and the Sources of Growth." Journal of Financial Economics, 58 : 261-310. Beck,T., A. Demirguc and R. Levine. 2003. "Law, endowments and Finance." Journal of Financial Economics, 70(1) : 137-181. Bencivenga, V. and B. Smith. 1991. "Financial Intermediation and Endogenous Growth." The Review of Economic Studies, 58 : 195-209. Bernake, B. S. 1983. "Nonmonetary effects of the financial crisis in the propagation of the Great Depression." American Economic Review, 73(2) : 257-276. Boyd, John H., and B. D. Smith. 1997. "Capital Market Imperfections, International Credit Market, and Non –convergence." Journal of Economic Theory, 73 : 335-364. Boyreau, D. G. 2003. "Financial Intermediation and Growth: Chinese Style." Policy Research Working Paper 3027, World Bank. Byrns, R. T. and G. W. Stone. 1992. Economics (5th Edition), New York: Harper Collins Publishers. Bundesbank. 2005. "Credit Growth, Bank Capital and Economic DeutcheBundesbank Monthly Report, March. Activity." 16 NRB ECONOMIC REVIEW Cappiello, L., A. Kadareja and C. K. Sarensen. 2010. "Do Bank Loans and Credit Standards Have An Effect on Output?" Working Paper Series No 50, European Central Bank, January. Crowley, J. 2008. "Credit Growth in the Middle EAST, North Africa and Central Asia Region" IMF Working Paper No. 08/184. Dewett, K. 2005. Modern Economic Theory, New Delhi, ShyamLal Charitable Trust. Dey, M. K. and S. Flaherty. 2005. "Stock Exchange Liquidity, Bank Credit and Economic Growth." Paper presented at the Max Fry Conference on Finance and Development, University of Birmingham, The Business School University House, Birmingham B15 2TT. Dickey, D. A. and W. A. Fuller. 1979. "Distribution of the Estimators for Time Series with a Unit Root." Journal for the American Statistical Association, 74 : 427-431. Engle, R. F. and C. W. J. Granger. 1987. "Co-integration and Error Correction: Representation, Estimation and Testing." Econometrica, 55 : 251-276. Fitz Gerald, V. 2006. "Financial Development and Economic Growth: A Critical View." Background Paper for World Economic and Social Survey. Fry, M. J. 1988. Money, Interest and Banking in Economic Development, London: John Hopkins University Press. Galor, O. and J. Zeira. 1993. "Income Distribution and Macroeconomics." The Review of Economic Studies, 60(1) : 35-52. Goldsmith, R. W. 1969. " Financial Structure and Development." New Haven, CT, Yale University Press. Greenwood, J. and B. Jovanovic. 1990. "Financial Development, Growth and Income Distribution." Journal of Political Economy, 98 :1076-1107. Gurley, J. and E. Shaw. 1967. "Financial Structure and Economic Development." Economic Development and Cultural Change, 15(3) : 257-268. Jalilian, H. and C. Kirkpatrick. 2001. "Financial Development and Poverty Reduction in Developing Countries." Working Paper No. 30, IDPM, Manchester University. Jayaratne, J. and P. Strahan. 1996. "The Finance-Growth Nexus: Evidence from Bank Branch Deregulation." Quarterly Journal of Economics, 111 : 639-670. Jhingan, M. L. 1984. Money, Banking and International Trade, Delhi vikash Publication Ltd. Kashyap, A. and J. Stein. 2000. "What Do a Million Observations on Banks Say About the Transmission of Monetary Policy." American Economic Review, 90 : 407-428. King, R. G. and R. Levine. 1993. "Finance and Growth: Shumpeter Might Be Right." Quarterly Journal of Economics, 108 : 717-738. Bank Credit and Economic Growth in Nepal: An Empirical Analysis 17 King, R. G. and R. Levine. 1993a. "Financial Intermediation and Economic Development." In Mayer and Vives(Eds): Financial Intermediation in the Construction of Europe: London Centre for Economic Policy Research, 156-189. Kinnon, M. C. 1973. Money and Capital in Economic Development, Washington: The Brooking Institute. Koivu, T. 2002. "Do efficient banking sectors accelerate economic growth in transition Countries." Bank of Finland, Institute for Economies in Transition, BOFIT Discussion Papers 14. Levine, R. 2002. "Bank –Based or Market Based Financial Systems: Which is better?" Journal of Financial Intermediation, 11 : 398-428. Levine, R. 1997. "Financial Development and Economic Growth: Views and Agenda." Journal of Economic Literature, 35: 688-726. Mankiw, G., D. Romer and D. N. Weil. 1992. "A Contribution to the Empirics of Economic Growth." Quarterly Journal of Economics, May, 107(2) : 407-437. Mishra, P. K., K. B. Das and B. B. Pradhan. 2009. "Credit Market Development and Economic Growth in India." Middle Eastern Finance and Economics. Mukhopadhya, B. and R. P. Pradhan. 2010. "An Investigation of the Finance Growth Nexus: Study of Asian Developing Countries Using Multivariate VAR Model." International Research Journal of Finance and Economics, 58 : 134-140. Murty, K. S., K. Sailaja and W. M. Dimissie. 2012. "The Long-Run Impact of Bank Credit on Economic Growth in Ethiopia: Evidence from the Cointegration Approach." European Journal of Business and Management, 4(14) : 20-33. Papaioannou, E. 2007. "Finance and Growth A Macroeconomic Assessment of The Evidence From A European Angle." European Central Bank, Working Paper Series, No. 787. Odedokun, M. O. 1989. “Causalities Between Financial Aggregates and Economic activities in Nigeria: The Results from Granger’s Test.” Savings and Development, 23(1) : 101-111. Okyo, M. I., M. Blessing and U. D. Okelue. 2012. "The Effect of Deposit Money Banks Credit on Nigerian Economic Growth." International Journal of Current Research, 4(12) : 555-559. Saw, E. S. 1973. Financial Deepening in Economic Development, London: Oxford University Press. Schumpeter, J. A. 1911. The Theory of Economic Development, Oxford: Oxford University Press. Solow, R. M. 1956. "A Contribution to the Theory of Economic Growth." Quarterly Journal of Economics, 70 : 65-94. 18 NRB ECONOMIC REVIEW Ugoani, J. N. N. 2013. "Power of Bank Credit on Economic Growth: A Nigerian Perspective." International Journal of Financial Economics, 1(3) : 93-102. Valpi, F. 2006. "Financial Development and Economic Growth: A Critical View." Background Paper for World Economic and Social Survey, Oxford University, London. Were Maureen, J. Nzomoi and N. Rutto. 2012. "Assessing the Impact of Private Sector Credit on Economic Performance." International Journal of Economics and Finance, 4(3) : 182-190. Bank Credit and Economic Growth in Nepal: An Empirical Analysis 19 APPENDICES 1. Co-integration Test (Four Variables) Date: 10/20/14 Time: 14:39 Sample (adjusted): 1977 2014 Included observations: 38 after adjustments Trend assumption: Linear deterministic trend Series: IR LNRGDP LNRGEXP LNRPVCT Lags interval (in first differences): 1 to 1 Unrestricted Cointegration Rank Test (Trace) Hypothesized No. of CE(s) Eigenvalue Trace Statistic 0.05 Critical Value Prob.** None * At most 1 * At most 2 At most 3 0.497158 0.362774 0.310181 0.025108 58.32488 32.20065 15.07670 0.966282 47.85613 29.79707 15.49471 3.841466 0.0039 0.0259 0.0577 0.3256 Trace test indicates 2 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values Unrestricted Cointegration Rank Test (Maximum Eigenvalue) Hypothesized No. of CE(s) Eigenvalue Max-Eigen Statistic 0.05 Critical Value Prob.** None At most 1 At most 2 At most 3 0.497158 0.362774 0.310181 0.025108 26.12422 17.12396 14.11042 0.966282 27.58434 21.13162 14.26460 3.841466 0.0759 0.1662 0.0528 0.3256 Max-eigenvalue test indicates no cointegration at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values 20 NRB ECONOMIC REVIEW 2. Co-integration Test (Two Variables) Date: 10/20/14 Time: 14:42 Sample (adjusted): 1977 2014 Included observations: 38 after adjustments Trend assumption: Linear deterministic trend Series: LNRGDP LNRPVCT Lags interval (in first differences): 1 to 1 Unrestricted Cointegration Rank Test (Trace) Hypothesized No. of CE(s) Eigenvalue Trace Statistic 0.05 Critical Value Prob.** None * At most 1 0.336304 0.001536 15.63582 0.058428 15.49471 3.841466 0.0476 0.8090 Trace test indicates 1 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values Unrestricted Cointegration Rank Test (Maximum Eigenvalue) Hypothesized No. of CE(s) Eigenvalue Max-Eigen Statistic 0.05 Critical Value Prob.** None * At most 1 0.336304 0.001536 15.57739 0.058428 14.26460 3.841466 0.0308 0.8090 Max-eigenvalue test indicates 1 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values Bank Credit and Economic Growth in Nepal: An Empirical Analysis 21 3. Pairwise Granger Causality Tests (With two lags) Sample: 1975 2014 Lags: 2 Null Hypothesis: Obs LNRGDP does not Granger Cause LNRPVCT 37 LNRPVCT does not Granger Cause LNRGDP 4. F-Statistic Prob. 7.57364 0.00338 0.0020 0.9966 Pairwise Granger Causality Tests (With five lags) Sample: 1975 2014 Lags: 5 Null Hypothesis: Obs LNRGDP does not Granger Cause LNRPVCT 34 LNRPVCT does not Granger Cause LNRGDP 5. F-Statistic Prob. 3.87323 1.60409 0.0108 0.1988 Economic Growth in Developing Countries Countries 2009 2010 2011 2012 2013 2014 China 9.2 10.4 9.3 7.8 7.7 - India 8.5 10.5 6.3 4.7 5.0 - Indonesia 4.6 6.2 6.5 6.3 5.8 - Malaysia -1.5 7.4 5.1 5.6 4.7 - Nepal 4.5 4.8 3.4 4.9 3.6 5.2 Source: World Bank 22 NRB ECONOMIC REVIEW 6. Residuals Plots move around zero. LNRPVCT Residuals .12 .08 .04 .00 -.04 -.08 -.12 1980 1985 1990 1995 2000 2005 2010 2005 2010 LNRGDP Residuals .06 .04 .02 .00 -.02 -.04 -.06 1980 7. 1985 1990 1995 2000 VEC Residual Serial Correlation LM Tests Null Hypothesis: no serial correlation at lag order h Sample: 1975 2014 Included observations: 36 Lags LM-Stat Prob 1 2 3 9.450576 5.579553 5.590300 0.0508 0.2328 0.2319 Probs from chi-square with 4 df. Bank Credit and Economic Growth in Nepal: An Empirical Analysis 23 8. Vector Error Correction Estimates Date: 10/20/14 Time: 14:37 Sample (adjusted): 1978 2014 Included observations: 37 after adjustments Standard errors in ( ) & t-statistics in [ ] Cointegrating Eq: CointEq1 LNRGDP(-1) 1.000000 LNRPVCT(-1) -0.409841 (0.00877) [-46.7222] C -8.153330 Error Correction: D(LNRGDP) D(LNRPVCT) CointEq1 0.055931 (0.07941) [ 0.70431] 0.916661 (0.21503) [ 4.26285] D(LNRGDP(-1)) -0.087868 (0.18502) [-0.47491] -1.392763 (0.50100) [-2.77997] D(LNRGDP(-2)) -0.215248 (0.19596) [-1.09841] -1.595705 (0.53063) [-3.00717] D(LNRPVCT(-1)) -0.014957 (0.04599) [-0.32525] 0.534695 (0.12452) [ 4.29405] D(LNRPVCT(-2)) 0.003950 (0.04950) [ 0.07980] -0.078660 (0.13404) [-0.58684] C 0.056417 (0.01218) [ 4.63310] 0.180577 (0.03297) [ 5.47649] 0.054060 -0.098511 0.013521 0.020884 0.354324 93.91664 -4.752251 -4.491021 0.042574 0.019926 0.593335 0.527744 0.099137 0.056551 9.045982 57.05943 -2.759969 -2.498739 0.104796 0.082290 R-squared Adj. R-squared Sum sq. resids S.E. equation F-statistic Log likelihood Akaike AIC Schwarz SC Mean dependent S.D. dependent Determinant resid covariance (dof adj.) Determinant resid covariance Log likelihood Akaike information criterion Schwarz criterion 1.36E-06 9.52E-07 151.4944 -7.432128 -6.822591 24 NRB ECONOMIC REVIEW 9. Autocorrelations with 2 Std.Err. Bounds Cor(LNRPVCT,LNRPVCT(-i)) Cor(LNRPVCT,LNRGDP(-i)) .4 .4 .2 .2 .0 .0 -.2 -.2 -.4 -.4 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 Cor(LNRGDP,LNRPVCT(-i)) 4 5 6 7 8 9 10 11 12 10 11 12 Cor(LNRGDP,LNRGDP(-i)) .4 .4 .2 .2 .0 .0 -.2 -.2 -.4 -.4 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 10. Inverse Roots of AR Characteristic Polynomial 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 6 7 8 9 Determinants of Stock Market Performance in Nepal# Prakash Kumar Shrestha, Ph.D.* Biggyan Raj Subedi** Abstract This paper empirically examines the determinants of the stock market performance in Nepal using monthly data for the period of mid-August 2000 to mid-July 2014. The impact of major changes in politics and Nepal Rastra Bank’s policy on lending against share collateral has also been assessed. Empirical results obtained from OLS estimations of behavioural equations reveal that the performance of stock market is found to respond positively to inflation and broad money growth, and negatively to interest rate. This suggests that, in Nepal, share investors seem to take equity as a hedge against inflation and consider stock as an alternative financial instrument. Further, availability of liquidity and the low interest rates stimulate the performance of the Nepalese stock market. More importantly, stock market has been found to respond significantly to changes in political environment and the policy of Nepal Rastra Bank. These findings help to design policies to stabilize or stimulate the share market in Nepal. Key Words: Stock Market, Macro Variables, Nepal JEL Classification: G10, E44 # The earlier version of this paper is available at www.nrb.org.np under NRB Working Paper series, NRB-WP-24, 2014. * Director, Nepal Rastra Bank, Research Department, Central Office, Baluwatar, Kathmandu, Nepal. Email:[email protected] ** Deputy Director, Nepal Rastra Bank, Research Department, Central Office, Baluwatar, Kathmandu, Nepal. Email:[email protected] We would like to thank Mr. Shalikram Pokharel and Mr. Nanda Dhakal, Assistant Directors, Research Department, Nepal Rastra Bank for their help in preparing this paper. In addition, we are grateful to the Editorial Board and an anonymous external reviewer for providing valuable comments to revise this paper. 26 NRB ECONOMIC REVIEW I. BACKGROUND The history of stock market is not long in Nepal. Securities Exchange Centre (SEC) was established in 1976 with an objective of facilitating and promoting the growth of capital market (Gurung, 2004). However, it opened its floor for secondary trading of shares only in 1981, which was only for government bonds (NRB, 1996). With enactment of Securities Exchange Act 1984, SEC opened its floor for corporate share trading also, but it was very limited. The organized and full fledged stock market began with the conversion of Securities Exchange Centre into Nepal Stock Exchange (NEPSE) Limited in 1993. The NEPSE opened its trading floor in the beginning of 1994. Till now, it is the only stock exchange in Nepal. Hence, the stock market in Nepal is still in evolving stage but of special interest as it has grown significantly since its establishment. It was established in order to mobilize capital alternative to traditional banking sector for promoting economic growth and development in the country. Within a short period of time since its inception, the NEPSE index witnessed significant ups and downs. Recently, after the results of the second CA election in November 2014, the NEPSE index took an upward trend until August 2014. On July 14, 2014 the benchmark index reached 1036.1, the highest in the last six years. Earlier on August 31, 2008, the NEPSE index had reached its all-time high of 1175 points before plunging to a record low of 292 on June 15, 2011. Normally, the stock market index is taken as a barometer of an economy. Growth in stock index is normally considered as a good sign since it implies the investors are confident about the future prospect of the economy. It helps promote investment in the economy. However, a rapid increase in the stock market index is always a matter of concern. If the increase in the index is not justified by the fundamentals, such a rise cannot be sustained and eventually the index will plummet endangering the economic and financial stability. Hence, it is essential that the policymakers keep eyes on the stock market development and be ready to take appropriate measures, if needs arise, to prevent the build up of bubbles and collapse in the market. For this, it is necessary to understand the relationship between the stock market index and the factors that influence it. Several factors may affect the stock market. Any factors that have an effect on cash flows of firms or discount rate will have impact on the stock market. However, which factors affect to what degree will vary from country to country, depending on the size, type and other characteristics of the economy and the market. In this context, this paper aims to analyze the relationship between the performance of NEPSE index and major macroeconomic variables in Nepal using monthly data that span from mid-August 2000 to mid-July 2014. In addition to main variables, this paper also assesses the impact of changes in politics and Nepal Rastra Bank's policy on lending against share collateral. It is expected that the findings of this study would provide some meaningful insights to understand the determinants behind the performance of Nepalese stock market, useful for both policymakers and investors. Determinants of Stock Market Performance in Nepal 27 There are a lot of research studies on the determinants of stock market in other countries such as Asprem (1989), Yosuf and Majid (2007), Rahman et al. (2009), Singh (2010), Hsing (2011, 2014), Eita (2012), Quadir (2012), Naik and Padhi (2012), Jauhari and Yadav (2014), and Khan (2014). A very few studies have been done on the Nepalese stock market such as Dangol (2008, 2010), Pradhan and KC (2010), Bhatta (2010) and Regmi (2012). These studies mainly focused on micro perspective rather than macro and policy perspectives. This study differs from them since we have examined the impact of macroeconomic variables as well as politics and NRB’s policy changes on the stock market performance. The paper is structured as follows. Section 2 presents the glimpse of the Nepalese stock market, which is followed by the review of literature in section3. Section 4 describes the data and methodology used and section 5 presents the empirical results and discussion. Finally, Section 6 concludes the study. II. GLIMPSE OF THE NEPALESE STOCK MARKET The Nepalese stock market is still in infant stage. However, there has been some progress. In the last two decades, the number of listed companies at NEPSE has increased from 79 in 1995 to 237 in 2014. During the same period, market capitalization has increased from 5.9 percent to 54.8 percent of GDP (Table 1). The growth in the listed companies mostly includes banks and financial institutions that were opened with the adoption of financial liberalization. Existing regulations require bank and financial institutions to publicly float at least 30 percent of shares and get listed in the stock exchange within a specific period of time. However, there is no such a mandatory requirement for companies in the real sector. As such, very few real sector companies have been listed in the stock market. As of mid-July 2014, there were 182 (76 %) financial institutions out of 237 listed companies at NEPSE (Table 2). Similarly, banks and financial institutions contributed to 64.3 percent of the total market capitalization followed by insurance (13.3 percent) and hydropower (8.7 percent). Market capitalization of manufacturing and processing firms remained just at 1.9 percent. Table 1: Glimpse of the Nepalese Stock Market Year No. of listed companies Market Capitalization (Rs in million) Market Capitalization/GDP (percent) 1995 79 12963 5.9 2000 110 43123 11.4 2005 125 61366 10.4 2010 176 376871 31.6 2014 237 1057166 54.8 Source: Quarterly Economic Bulletin and Current Macroeconomic Situation of Nepal (2013/14), NRB 28 NRB ECONOMIC REVIEW Table 2: Structure of the Nepalese Stock Market (Mid-July 2014) Type of Institution Number Market Capitalization (%) 182 Financial Institutions 64.3 22 Insurance Companies 13.3 18 Manufacturing & Processing 1.9 4 Hotel 2.4 4 Trading 0.1 5 Hydro Power 8.7 2 Others 9.3 237 Total 100.0 Source: Current Macroeconomic Situation of Nepal (2013/14), NRB As regards the movement of the NEPSE index, it hovered around 200 points between 1994 and 1999. This was also the period when Nepalese stock market was evolving in terms of number of listed companies and the market capitalization. From 2000 onwards, the NEPSE index observed a greater fluctuation. In Figure 1, we can see the NEPSE peaking up three times in the past such as in November 2000, December 2007 and August 2008 before taking a sharp plunge. Now again in 2014, after the election of second Constituent Assembly, the NEPSE index reached as high as 1036.1 points in mid-July 2014. What factors can explain the movement of the NEPSE is a matter of study in this paper. Figure 1: NEPSE Index (mid-month) 1200 1000 800 600 400 200 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2006 2007 2008 2009 2010 2011 2012 2013 0 Source: www.nepalstock.com.np Determinants of Stock Market Performance in Nepal 29 III. REVIEW OF LITERATURE The Arbitrage Pricing Theory, introduced by Ross (1976), establishes the theoretical framework to link stock returns with several variables which can influence the source of income volatility (Rahman, et al. 2009). Mukherjee and Naka (1995) showed that economic variables influence stock market returns through their effects on future dividends and discount rates. Macroeconomic variables selected to examine the determinants of stock market tend to differ slightly across studies, however (Rahman, et.al. 2009). Most common variables are the rate of inflation, money growth, interest rates, industrial production and exchange rates for explaining the stock market movement. Selection of these macroeconomic variables has theoretical justifications as follows. Higher interest rates or discount rates would reduce the present value of cash flows, which would reduce the attractiveness of investment, hence, shrinks the value of stock returns (Rahman, et al. 2009). Another impact could be through portfolio substitution, a rise in the rate of interest increases the opportunity cost of holding cash, which later on leads to a substitution effect between stocks and other interest bearing securities like bonds (Rahman, 2009, p.98). In the literature, the common interest rate proxies are the treasury bills rates as being employed by Mukherjee and Naka (1995), Ratanapakorn and Sharma (2007), Yusof and Majid (2007), and Eita (2012)1. In case of money supply, Mukherjee and Naka (1995) argue that if an increase in money supply leads to economic growth, stock prices would benefit from expansionary monetary policy. In another way, with increase in money supply, the availability of liquidity at a lower interest rate increases, which can flow into the stock market. In contrast, Fama (1981) argues that an increase in money supply leads to inflation (or expected inflation) in the economy, which in turn increases the discount rate and lowers the stock market returns. Moreover, inflation is also an important variable that investors consider before making any investment decisions. Theoretically, Asprem (1989) put forward that inflation should be positively related to stock return if stocks provide a hedge against inflation. This is based on Fisher (1930) who posits that stock markets are independent of inflation expectations since equities are a claim against real assets of the company. Fama (1981) however, disagrees with the generalized Fisher hypothesis on the basis that an increase in inflation causes uncertainty and reduces future economic activity, which reduces the stock price. Another variable of interest used in the literature is the exchange rate. The exchange rate influences the firm’s cash flow and the amount of dividend to be paid, especially in open economy (Eita, 2012). A depreciation of the local currency makes exporting goods less expensive and may lead to an increase in foreign demand and sales for the 1 Lending rate used by Hsing (2014). 30 NRB ECONOMIC REVIEW exporting firms (Pan et al., 2007). As a result, the value of exporting (importing) firms would increase (decrease). Rehman et al. (2009) argue that the importance of international trade in the economy determines the impact of exchange rate on stock price. However, we do not consider exchange rate in our case because of several reasons. First, the Nepalese stock exchange is overwhelmingly dominated by banks and financial institution; there are no any trading companies. Second, Nepal has not opened the capital account so that there is no foreign portfolio investment in stock market. Third, Nepal has been following the pegged exchange rate with India currency so that exchange rate may not be the important variable for stock market. Other than monetary variables mentioned above, the level of real economic activity is the crucial factor in determining the stock market returns (Rehman et al. 2009). There is a general consensus that an increase in economic activity causes stock market returns to increase (Eita, 2012, p874). The most popular measure of real economic activity is the gross domestic product (GDP). Unfortunately, data on GDP is normally on annual basis and only in some countries, it can be available on a quarterly frequency. Some use industrial production index as another measure for real economic indicator (Rashid, 2008; Rehman et al., 2009). In addition, researchers have used other additional variables as well such as debt/GDP ratio and yields of alternative financial assets by Hsing (2014), foreign reserves by Rahman et al. (2009), and variables like capital formation and gold price by Jauhari and Yadav (2014), gross capital formation relative to GDP, credit to the private Sector to GDP and net remittance relative to GDP by El-Nadar and Alraimony (2013) and federal fund rate by Yusof and Majid (2007) as factors affecting the performance of stock market. Empirical results regarding macroeconomic determinants are mixed types. Estonian and Hungarian stock market index have a positive relationship with debt/GDP ratio, real GDP and the German stock market index and a negative relationship with the exchange rate, the domestic interest rate, the expected inflation rate, and the euro area government bond yield (Hsing 2011; 2014). In case of Namibia, an increase in economic activity and the money supply increases stock market prices, while increases in inflation and interest rates decrease stock prices (Eita, 2012). The results suggest that equities are not a hedge against inflation in Namibia, and contractionary monetary policy generally depresses stock prices. In Jordon, money supply, gross capital formation, inflation, and credit to the private sector have significant positive relationship, and income and net remittance have negative relationship with stock market (El-Nadar and Alraimony, 2013). Moreover, there is a co-integrating relationship of Malaysian stock market index with changes in money supply, interest rate, exchange rate, reserves and industrial production index (Rahman et al., 2009). In case of India, the macroeconomic variables like GDP, savings, capital formation, gold price, industrial output, money supply, exchange rate, WPI, and interest rate have concurrence with the variability of the Sensex index (Jauhari and Yadav, 2014). On the other hand, Naik and Padhi (2012) also examined the Indian stock market index (BSE Sensex) and observed the positive relationship between stock price and money supply and industrial production but negative relationship with inflation. The exchange rate and the short-term interest rate were found to be insignificant in determining stock prices in India. However, Rashid (2008) showed the long run relationship between stock Determinants of Stock Market Performance in Nepal 31 prices and macroeconomic variables like exchange rate, industrial index, interest rate, inflation in Pakistan. Specially, in Pakistan, exchange rate, inflation and GDP growth rate were positively related with stock prices (KSE-100 index) while the interest rate was negatively related as found by Khan (2014). Yusof and Majid (2007) found a significant direct impact of US federal fund rate on the Malaysian stock market, reflecting the impact of capital flows on the stock market. Most studies use either monthly or quarterly data for examining the determinants of stock performance. Ratanapakorn and Sharma (2007), Eita (2012), and Kemboi and Tarus (2012) use quarterly data, while Yusof and Majid (2007), Rahman et al. (2009), Singh (2010), El-Nadar and Alraimony (2013) use monthly data. With regards to methodology, Rahman et al. (2009), Eita (2012) employ VAR framework. Kwon and Shin (2001), Rashid (2008), and El-Nadar and Alraimony (2013) use cointergration and variance decomposition, while Hsing (2011, 2014) uses GARCH method and Rashid (2008), Singh (2010), and Jauhari and Yadav (2014) apply Granger causality test. On the other hand, Yusof and Majid (2007) apply the ARDL approach. Hence, there is no unique way to investigate the determinants of stock market performance. 3.1 Politics and Stock Market The stock market index, in general, is considered as the reflection of the expectation of future profitability of the companies. This market, therefore, tends to be influenced not only by macroeconomic fundamentals, but also by the unexpected political events as well as policy changes. Several studies have found the relationship between the political event and the stock market performance. For example, Beaulieu et al. (2006) investigated the short run impact of the political uncertainty associated with the 1995 Quebec referendum on the stock returns. The study found that the uncertainty surrounding the referendum outcome had short run impact on stock returns of Quebec firm, implying that the stock market was directly influenced by the political risk and uncertainty. Similarly, Jensen and Schmith (2005) estimated the impact of the four main Brazilian presidential candidates on the mean and variance of the Brazilian stock market using a number of time-series regressions. They argue that political events, such as the election of a politician that is expected to enact “market-friendly” policies, lead to increases in stock market returns while political events that are expected to have a negative impact on the economy and specific firms lead to decreases in stock market returns. 3.2 News and Stock Market Stock markets are heavily affected by news and rumours, like a “beauty context” as described by Keynes (1936). News can affect sentiments as well as expectation of the investors and performance of the companies. Most importantly, people interpret news differently based on their own cognitive power. There are some empirical examinations on the impacts of news on the performance of stock. For example, Boudoukh et.al.(2013) investigated the relation between news and the stock prices of 795 S&P500 companies, 32 NRB ECONOMIC REVIEW covering the period of January 1, 2000 to December 31, 2009. Using advanced textual analysis method, they find that, when information can be identified and that the tone (i.e., positive versus negative) of this information can be determined, there is a closer link between stock prices and information. Similarly, Alanyali et. al. (2013) investigated daily print issues of the Financial Times from 2nd January 2007 to 31st December 2012 to quantify the relationship between decisions taken in financial markets and developments in financial news. They find a positive correlation between numbers of times the name of a company mentioned daily in the Financial Times and the daily transaction volume of a company's stock both on the day before and on the same day of the news released. Their results provide quantitative support for the suggestion that movements in financial markets and movements in financial news are closely interlinked. 3.3 Past Empirical Evidence from Nepal There are a few other studies on the explaining stock market performance, mainly from micro perspectives. For example, Joshi (2012) examined the impact of dividends on stock price in the context of Nepal and found the impact of dividends is more pronounced than that of retained earnings on stock prices in Nepal. Dangol (2008) studied the reaction of Nepalese stock market to announcements of unanticipated political events using the event analysis methodology. His analysis covered the period from 2001 to 2006. He found that good-news (bad news) political announcements generate positive (negative) abnormal returns in the post-event period. This finding suggests that there is a strong linkage between political uncertainty and common stock returns in Nepal. In another study of Dangol (2010) examined the random walk behaviour on daily market returns of the Nepal Stock Exchange for the period between July 2000 and January 2010 and found that the Nepalese stock market does not show any characteristics of random walk and thus, is not weak form efficient. Findings of Bhatt (2010) are also similar. This means news affects the movement of the stock market index. Further, Pradhan and KC (2010) assessed equity share price behaviour in Nepal and tested the hypothesis that share price changes are independent using weekly data of 26 listed companies from mid-July 2005 to mid-July 2008. They found that random walk hypothesis holds for less frequently traded stocks but do not hold for highly traded stocks at NEPSE. IV. DATA AND METHODOLOGY 4.1 Data and Sample Based on the availability of data and their relevancy as guided by the literature and considering the feature of Nepalese stock market, the following data are taken to examine the determinants of stock market index in Nepal as shown in Table 3. Determinants of Stock Market Performance in Nepal 33 Table 3: Variables and their Description Variable SI GDP CPI M2 TB91 D1 D2 Description Unit NEPSE Index Annual real GDP Rs in million CPI index, monthly average (base year = 2005/06) Broad Money Supply Rs in million 91 day Treasury bill rate Percent Political Event Dummy (takes value 1 if negative scenario, 0 otherwise) Policy Change Dummy (takes value 1 if margin lending is tightened, 0 if it is relaxed) The level of real economic activities is one of the crucial determinants of the stock market performance as a scale variable. The traditional measure for such activities is the gross domestic product (GDP). However, GDP data are unavailable on a monthly basis (not even on a quarterly basis). Hence, GDP variable has been dropped in further empirical estimation. All other data are collected on a monthly basis. Given the data availability and relevancy, the sample period of August 2000 to July 2014 has been chosen. Though the formal trading in Nepalese stock market started in 1994, the stock market was in evolving stage and highly immature until 2000. This fact is also reflected in Figure 1, which shows that NEPSE remained relatively flat until 2000. 4.2 Methodology Based on the literature and the availability of data, the study has used the following general behaviour model. SIt = f (CPIt,M2t, TB91t, SIt-1, D1, D2) …… (1) where the meanings of symbols are same as described in Table 3. The two dummies d1 and d2 are introduced to capture the impact of political changes and the NRB's policy changes. All other variables are standard in the literature. The first lag of stock market index is also included in our model as the literature suggests that stock prices tend to be highly persistent. A large section of investors are “chartist” who just follows the trend of movement of stock market index. Moreover, information on fundamental comes late so a majority of stock investors apply their own gut feeling. Though stock returns are theoretically assumed to follow random walk as argued by the efficient market hypothesis, many studies have found that the stock returns are autocorrelated. Boudoukh et al. (1994) points out that time series patterns occur in stock returns because investors either overreact or partially adjust to information arriving to the market. Prior to deciding on the appropriate method, a preliminary examination of the nature of the data is necessary. We follow the standard procedure of unit root testing by employing the Augmented Dickey Fuller (ADF) test. Since the ADF test is often 34 NRB ECONOMIC REVIEW criticized for low power, we complement this test with the Phillips Perron (PP) test. Based on the unit root test results in Table 4, all variables, except TB91 are I(1). Table 4: ADF and Philip Perron test Variables ADF Test H0: Variable non-stationary is Philips Perron Test H0: Variable is non-stationary Order of Integration log(SI) -1.552 -1.304 I(1) dlog(SI) -10.297*** -10.297*** log(CPI) -2.419 -2.612 I(1) dlog(CPI) -0.582** -10.435*** log(M2) -1.516 -1.516 I(1) dlog(M2) -12.408*** -13.768*** TB91 -2.821* -2.918** I(0) *** implies significant at 1% level, ** implies significant at 5% level and * implies significant at 10% level. Source: Authors’ calculation Since not all selected variables are in same order, we cannot follow VAR or cointegration approach. More importantly, application of VAR method may not be appropriate when the real sector variable is missing. Rather, the following models are estimated by OLS using first difference of I(1) variable2. It seems that following models are able to capture the performance of stock market. Considering the possibility of multicollinearity among explanatory variables, we do the estimation on step by step basis, and finally all explanatory variables are included in equation (5). dlog(SI)t = + 1 dlog(CPI)t + 1D1 + 2D2 + dlog(SI)(t–1) + t …… (2) dlog(SI)t = + 2 dlog(M2)t + 1D1 + 2D2 + dlog(SI)(t–1) + t …… (3) dlog(SI)t = + 3 TB91t + 1D1 + 2D2 + dlog(SI)(t–1) + t …… (4) dlog(SI)t = + 1 dlog(CPI)t + 2 dlog(M2)t + 3 TB91t + 1D1 + 2D2 + dlog(SI)(t–1) + t …… (5) V. EMPIRICAL RESULTS 5.1 Correlation Analysis Based on annual data from 2000/01 to 2013/14, the correlation between the real GDP and NEPSE index is found to be 0.57 (P-value=0.03), both positive and significant. Lack of GDP (or industrial production index) at a monthly frequency prevent us to use it in econometric estimation. The correlation of other macro variables (except interest rate, TB91) such as Consumer Price Index (CPI), Broad Money (M2) with NEPSE index (SI) 2 However, this study can be extended using ARDL approach with the expansion of stock market in Nepal. Determinants of Stock Market Performance in Nepal 35 are found to be statistically significant at 5 percent level of significance on monthly data covering mid-August 2000 to mid-July 2014 (Table 5). This suggests that there must be some relationship between stock market index and these macro variables. Table 5: Correlation between Log (SI) and macroeconomic variables LOG(CPI) Correlation 0.552 P-Value 0.000 Source: Authors’ calculation 5.2 LOG(M2) 0.563 0.000 TB91 0.033 0.671 Empirical Estimation and Discussion Table 6 presents the empirical results of the above model, estimated by using Eviews software. Each of the macro variables CPI, M2 and TB91 rate are found to be statistically significant, separately as well as taking all together. The dummy variables for political changes as well as for NRB's policy on lending against share collateral are also found to be significant. The signs of the coefficient are also as expected. Table 6: Regression Results Dependent Variable: dlog(SI) Number of observations: 166 Eq 5 0.038*** (0.005) dlog(CPI) 0.713** (0.039) dlog(M2) 0.800*** 0.754*** (0.005) (0.007) tb91 -0.005* -0.005* (0.042) (0.064) d1 -0.042*** -0.041*** -0.037*** -0.035*** (0.001) (0.001) (0.005) (0.005) d2 -0.039** -0.047*** -0.026 -0.039** (0.019) (0.005) (0.130) (0.022) dlog(SI(-1)) 0.149* 0.175** 0.115 0.130* (0.049) (0.020) (0.140) (0.089) Adj. R-squared 0.119 0.138 0.117 0.169 D-W stat 2.000 2.048 2.034 2.036 Note: *** implies significant at 1% level, ** implies significant at 5% level and * implies significant at 10% level. Figures in parenthesis are the respective P-values. Source: Authors' calculation const Eq 2 0.035*** (0.002) 0.755** (0.034) Eq 3 0.029** (0.013) Eq 4 0.054*** (0.000) 36 NRB ECONOMIC REVIEW The performance of the stock market is found to be positively related to inflation and growth of M2, and inversely related to TB91. This implies that higher inflation induces investors to invest in equity as a hedge against inflation, thereby pushing up stock prices. Likewise, growth in money supply leads to greater demand for stocks as result of portfolio substitution with ample liquidity. Given the limited supply of stocks, this exerts upward pressure on stock prices. Negative relationship between interest rate and stock market index implies that low interest rate make stocks more attractive because of low cost of credit as well as low opportunity cost foregone by holding bank deposits. Hence, in case of low interest rates, depositors may use their deposits to buy stock on the one hand and on the other hand, people can borrow at the low interest rates from banks and financial institution to make investment in share market. Our findings are similar to Khan (2014). The negative signs for the coefficients of both dummies indicate that political uncertainty and tightening of loans against share collateral by the NRB have negative impact on the NEPSE index. The positive coefficient for lagged stock market index term indicates the past month's stock price has a significant impact on the current month stock index. It shows the persistence behaviour, in other words, chartist behaviour in stock market. In all four equations, R2 is not so high which indicates that news, rumours and speculations must have played the important role in fluctuating stock market index. Moreover, stock market changes daily while other macroeconomic data are not available on a daily basis. For the results of equation (5) to be robust, it is necessary that it should not suffer from the problem of multicollinearity. Though CPI, M2 and TB91 have significant correlation in level form, no such correlation was found in log difference form (Table 7)3. This gives the indication that the possibility of multicollinearity in equation (5) is very low. Table 7: Cross -correlation between Explanatory Variables CPI M2 TB91 DLOG(CPI) DLOG(M2) CPI 1 DLOG(CPI) 1 --------M2 0.994 1 DLOG(M2) 0.036 1 (0.000) ----(0.643) ----TB91 -0.257 -0.271 1 TB91 -0.034 0.018 (0.000) (0.000) ----(0.663) (0.814) Note: Figures in parenthesis are the P-value for the null hypothesis of no correlation. Source: Authors’ calculation 3 TB91 1 ----- Most researchers appear to consider the value of 0.9 as the threshold beyond which problem of multicollinearity can occur (Asteriou and Hall, 2007). Determinants of Stock Market Performance in Nepal 37 VI. CONCLUSIONS This paper examines the determinants of stock market performance in Nepal, which has been passing through up and down in recent years. Since stock market tends to be highly sensitive and volatile, we examine the determinants of stock market index on monthly data. We have found the Nepalese stock market has been behaving as we expected theoretically. It has strong positive relationship with inflation and growth of money supply, and negative response to interest rate. It shows that people have been gradually taking stock market as a hedge against inflation and invest in this market when there is ample liquidity available at a low interest rate. More importantly, the stock market performance has been found to be influenced by political changes similar to finding of Dangol (2008) and the NRB’s policy. The positive outlook for political stability has positive impact on stock market index. Similarly change in NRB’s policy on lending against share collateral has significant impact on the movement of stock market index. A number of policy implications can be drawn from this study. First, Nepalese stock market has been quite responsive to macroeconomic development, especially monetary sector development. Second, a loose monetary policy could trigger an asset price bubble in share market, which is mainly dominated by banks and financial institutions. Third, share investors seem to watch the political development closely. Hence, a positive political development with stability can promote share market further which can play a vital role for financial intermediation and resource mobilization through capital market. Fourth, NRB’s policy on lending against share collateral has been effective in influencing the share market. This indicates the significant role of NRB’s policy in the share market. As our results reveal that share market is also influenced by rumours, news and speculations, transparency should be increased in this market by making information related to listed companies easily accessible. Transparency and communication should, in fact, be enhanced by the concerned authorities in order to clear gossips and rumours in the market. ***** REFERENCES Alanyali, M., H. S. Moat, and T. Preis. 2013. "Quantifying the Relationship Between Financial News and the Stock Market." Scientific Reports, 3, Article number 3578, http://www.nature.com/srep/2013/131220/srep03578/full/srep03578.html. Asprem, M. 1989. "Stock Prices, Asset Portfolios and Macroeconomic Variables in Ten European Countries." Journal of Banking and Finance. 13: 589-612. Asteriou, D. and S. G. Hall. 2007. Applied Econometrics, Palgrave Macmillan. Beaulieu, M. C., J. C. Cosset and N. Essaddam. 2006. “Political Uncertainty and Stock Market Returns: Evidence from the 1995 Quebec Referendum.” Canadian Journal of Economic, 39(2) : 621-641. 38 NRB ECONOMIC REVIEW Bhatta, G. P. 2010. "Does Nepalese Stock Market Follow Random Walk ?" SEBON Journal, 4 : 18-58. Boudoukh, J., R. Feldman, S. Kogan, and M. Richardson. 2013. "Which News Moves Stock Prices? A Textual Analysis." NBER Working Paper, No. 18725. Boudoukh, J., M. P. Richardson, and R. F. Whitelaw. 1994. "A Tale of Three Schools: Insights on Autocorrelations of Short-Horizon Stock Returns." Review of Financial Studies, 7(3) : 539-573. Dangol, J. 2008. "Unanticipated Political Events and Stock Returns: An Event Study." Economic Review, 20 : 86-110. Dangol, J. 2010. "Testing Random-Walk Behavior in Nepalese Stock Market." PYC Nepal Journal of Management, 3(1) : 28-36. Eita, J. H. 2012. "Modelling Macroeconomic Determinants of Stock Market Prices: Evidence from Namibia." The Journal of Applied Business Research, 28(5) : 871884. El-Nadar, H. M. and A. D. Alraimony. 2013. "The Macroeconomic Determinants of Stock Market Development in Jordon." International Journal of Economics and Finance, 5(6) : 91-103. Fama, E. F. 1981. "Stock returns, Real Activity, Inflation and Money." American Economic Review, 71(4) : 545-565. Fisher, I. 1930. The Theory of Interest. New York: Macmillan. Gurung, J. B. 2004. "Growth and Performance of Securities Market in Nepal." The Journal of Nepalese Business Studies, 1(1) : 85-92. Hsing, Y. 2014. "Impacts of Macroeconomic Factors on the Stock Market in Estonia." Journal of Economics and Development Studies, 2(2) : 23-31. Hsing, Y. 2011. "Macroeconomic Determinants of the Stock Market Index and Policy Implications: The Case of a Central European Country." Eurasian Journal of Business and Economics, 4(7) : 1-11. Jauhari, S. and H. S. Yadav. 2014. "Relationship between Stock Index and Macroeconomic Determinants: A Study of Post Globalization Era." International Journal of Core Engineering and Management, 1(3) : 79-100. Joshi. R. 2012. "Effects of Dividends on Stock Prices in Nepal." NRB Economic Review, 24(2) : 61-75 Jensen, N. M. and S. Schmith. 2005. "Market Responses to Politics: The Rise of Lula and the Decline of the Brazilian Stock Market." Comparative Political Studies, 38(10) : 1245-1270. Kemboi, J. K. and D. K. Tarus. 2012. "Macroeconomic Determinants of Stock Market Development in Emerging Markets: Evidence from Kenya." Research Journal of Finance and Accounting, 3(5) : 57-68. Khan, M. S. 2014. "Macroeconomic Variables and Its Impact on KSE-100 Index." Universal Journal of Accounting and Finance, 2(2) : 33-39. Determinants of Stock Market Performance in Nepal 39 Keynes, J. M. 1936. The General Theory of Employment, Interest and Money, New York: Harcourt Brace and Co. Kwon, C. S. and T. S. Shin. 1999. "Cointegration and Causality between Macroeconomic Variables and Stock Market Returns." Global Finance Journal, 10(1) : 71-81. Mukherjee, T. K. and A. Naka. 1995. "Dynamic Relations between Macroeconomic Variables and the Japanese Stock Market: An Application of a Vector Error Correction Model." Journal of Financial Research, XVIII(2 ): 223-237. Naik, P. K. and P. Padhi. 2012. "The Impact of Macroeconomic Fundamentals on Stock Prices Revisited: Evidence from Indian Data." Eurasian Journal of Business and Economics, 5(10) : 25-44. NRB.1996. 40 Years of Nepal Rastra Bank. Nepal Rastra Bank, Kathmandu. Pan, M. S., R. Fok and Y. A. Liu. 2007. "Dynamic Linkages between Exchange Rates and Stock Prices: Evidence from East Asian Markets." International Review of Economics and Finance, 16 : 503-520. Pradhan, R. S. and K. C. Saraswari. 2010. "Efficient Market Hypothesis and Behaviour of Share Prices: the Nepalese Evidence." SEBON Journal, 4 : 104-117. Quadir, M. M. 2012. "The Effect of Macroeconomic Variables on Stock Returns on Dhaka Stock Exchange." International Journal of Economics and Financial Issues 2(4) : 480-487. Rahman, A. A., N. Z. M. Sidek and H. T. Fauziah. 2009. "Macroeconomic Determinants of Malaysian Stock Market", African Journal of Business Management, 3(3) : 95106. Rashid, A. 2008. “Macroeconomic Variables and Stock Market Performance: Testing for Dynamic Linkage with a Known Structural Break.” Saving and Investment, 32(1) : 77-102. Ratanapakorn, O. and S. C. Sharma. 2007. "Dynamics analysis between the US Stock Return and the Macroeconomics Variables." Applied Financial Economics 17(4) : 369-377. Ross, S. A. 1976. "The Arbitrage Theory of Capital Asset Pricing." Economics Theory, 13 : 341-360. Journal of Singh, D. 2010. "Causal Relationship Between Macro-Economic Variables and Stock Market: A Case Study for India.", Pakistan Journal of Social Sciences, 30(2) : 263274. Yusof, R. M. and M. S. A. Majid. 2007. “Macroeconomic Variables and Stock Return in Malaysia: An Application of ARDL Bound Testing Approach." Saving and Investment, 31(4) : 449-469. 40 NRB ECONOMIC REVIEW Appendix 1 List of Major Political Events and Likely Impact on Share Market S. N. 1 June 2001 2 Feb. 2005 3 4 5 Oct. 2005 Jan. 2006 Apr. 2006 6 Nov. 2006 7 Apr. 2007 8 Jan. 2008 9 Apr. 2008 10 Aug. 2008 11 May 2009 12 13 Jun. 2010 May, 2011 14 Aug. 2011 15 May 2012 16 Nov. 2013 Date Event The Royal massacre. King Gyanendra dismissed Prime Minister Sher Bahadur Deuba and took up executive power. Cease fire by the Maoists. Cease fire withdrawn by the Maoists Restoration of Parliament and start of peace process Peace agreement between the government and Maoists; Maoists agreed to lay down arms. Maoists joined interim government, a move that took them into the political mainstream. A series of bomb blasts killed and injured dozens in the southern Terai plains, where activists were demanding regional autonomy. Former Maoist rebels became the largest party in elections of the new Constituent Assembly (CA), but failed to get an outright majority. Maoist leader Puspa Kamal Dahal (Prachanda) formed coalition government, with Nepali Congress in opposition. Prime Minister Prachanda resigned in a row with President Yadav. Maoists left the government after other parties opposed integration of former rebel fighters into national army. PM Madhav Kumar Nepal quit under Maoist pressure. Constituent Assembly failed to meet deadline for drawing up new constitution. PM Jhalnath Khanal resigned after government failed to reach compromise with opposition on new constitution. Prime Minister Baburam Bhattarai dissolved CA, called elections for November 2012, after politicians missed a final deadline to agree on a new constitution. Election for CA second time. Nepali Congress party, Nepal Communist Party (UML) became the first and second largest party with two-third majority together. These two parties have some common political agenda. Possible Impact Bad Bad Good Bad Good Good Good Bad Bad Good Bad Bad Bad Bad Bad Good Sources: Dangol (2008) and BBC News, South Asia: http://www.bbc.com/news/world-south-asia-12499391 Appendix 2 List of Major Policy Changes by NRB on Loans against Share Collateral and Likely Impact on Share Market S. N. Date 1 Oct7, 2007 Event Margin lending limit reduced to 50 % of last 90 days average price of shares; restriction on restructuring of margin loan; regulation requiring maximum period of margin loan not to exceed 1 year. 2 Jan 22, 2008 Margin lending limit not to exceed 50 % of the last 180 days average price of shares or 50 % of market price, whichever is minimum. 3. Jan 15, 2009 Regulation requiring to make a margin call if the collateral is seen not sufficient to secure the loan. 4. Oct 30, 2009 Restructuring of the margin loan was allowed provided that the 50 % of principal and interest has been repaid. 5. Feb 22, 2010 No need to make margin call if the price fall of the share is not more than 10%; About 75 % of margin loan amount was allowed to restructure 6. Aug 10, 2010 Margin lending limit increased to 60% of the last 180 days average price of shares or 50 % of market price, whichever is minimum. 7. Jul 14, 2011 BFIs were allowed to make self decision on the limit of margin lending based on the last 180 days average price of shares or 50 % of market price, whichever is minimum; Revaluating the shares and extending loan limit was restricted. 8. Jun 10, 2012 Loan could be extended with the guarantee from the broker instead of pledging original share certificates. Source: Various NRB Circulars Possible Impact Bad Bad Bad Good Good Good Good Good Structural Change and Per Capita Income in Nepal: Empirical Evidences# Guna Raj Bhatta Abstract This paper empirically examines Nepalese economic structure by applying OLS technique on the annual series of sectoral growth, population and capital related variables ranging from 1975 – 2012. The estimates obtained with due consideration of stationarity of the series including HP filter revealed that industrial sector is significant to increase per capita income compared to the agriculture and service sectors in Nepal. Moreover, health as indicated by life expectancy and population at working age are found to be substantial to increase the income but, education and capital formation are found insignificant. It is inferred that employment matters for raising per capita income, requiring employment-led growth rather mere growth of economic sub-sectors. Hence, it is needed to have balanced contribution of economic sub-sectors and their employment share to national economy along with healthy workforce to raise the per capita income. Key Words: Structural Change, Employment, Per Capita Income, Nepalese Economy JEL Classification: O10, O49, L16, N10 # The earlier version of this paper is available at www.nrb.org.np under NRB Working Paper series, NRB-WP-23, 2014. Assistant Director, Nepal Rastra Bank, Research Department, Central Office, Baluwatar, Kathmandu, Nepal. Email: [email protected], [email protected] Acknowledgement: I would like to express my sincere thanks to the Editorial Board of NRB Economic Review for the comments and suggestions. 42 NRB ECONOMIC REVIEW I. INTRODUCTION Economic growth of the country is always a major concern worldwide since rise in GDP is one of the major human welfare indicators. Direct correlation is found between increased real output and income, with improvements in development factors in the history (Welker, 2012). Higher GDP growth not only provides better opportunities to improve access over basic requirements for the livelihood, but also provides more saving and revenue to the government. Nevertheless, economic transformation from rural agricultural to modern industrial or service sectors is the fundamental requirement to achieve high and sustainable growth. This can be said as the rapid and sustainable economic development in most of the developed as well as emerging economies has been achieved with the permanent shifts in their economic structure over the long-run. They have experienced a gradual transformation of the economy from rural subsistent agriculture to the modern industrial and then ultimately to the service dominant. Although there are ample resources such as sufficient arable land, natural resources and labour force, Nepal is still among the poorest countries in the world as the latest human development index ranked the country 157th out of 187 and the rank for per capita income is 207th out of 229 countries (based on purchasing power parity). Nevertheless, the rank is 35th in labor force availability and 46th in percentage of arable land (CIA Fact Book, 2013). Likewise, Nepal is ranked fifth in employees per hectare, requiring 3.6 people to cultivate one hectare of land. Economic growth is predominantly determined by the performance of agricultural sector in Nepal. This sector contributes more than one third to the country's gross domestic product (GDP) and employs about two-thirds of the total labour force inferring a low productivity. Moreover, the country experiences a monsoon-based growth as it witnesses an improved agricultural GDP at the time of favorable rainfall (Acharya & Bhatta, 2013). With these scenarios, Nepal witnessed a 4 percent growth of the economy on an average in recent ten years, in which agriculture and industry sectors had grown by 3.3 percent and 2.7 percent respectively whereas services sector had witnessed a growth of 5.3 percent. The share of agriculture was gradually declining over the study period whilst the share of services steadily increasing, being more than 50 percent in 2013 and 2014. However, the industrial share to GDP was found to be increasing until late 1990s and started declining. The aforesaid facts and figures clearly depicts Nepalese economy's gradual structural shift from agro to services sector lead economy. However, problem can be witnessed in the employment pattern. The agriculture sector contributes only one-third to the economy but more than 64 percent of the total employment is on this sector. Similarly, the contribution of service sector to the economy has been growing rapidly but the total employment share of it is around 15 percent. In this milieu, this paper attempts to examine the Nepalese economic structure more closely by comparing and contrasting with the prominent literatures and prescribing some perceived policies for high and sustainable growth of the economy. Structural Change and Per Capita Income in Nepal: Empirical Evidences 43 The rest of the paper flows as follows. The next section reviews the prominent literature of structural change. Section three portrays the structural change of the Nepalese economy. Data and methodology are discussed in section four. Section five explains the results and findings and finally section six concludes the paper with some policy prescriptions for high and sustainable growth. II. STRUCTURAL CHANGE MODELS AND LITERATURE The economic structural change is often considered as a permanent shift in the fundamental structure of an economy, basically an agrarian economy shifts to either industry or service based. In many countries, it primarily involves a decline in share of agriculture to the GDP and a rise in share of services (Maddison, 1991; Buera and Kaboski, 2012). It is believed that without the structural change, modern economic development is impossible (Kuznets, 1971) which is mostly associated with promising growth and continuous transformation (Pasinetti, 1981) in the globalized and dynamic economic system. Although employment shares in manufacturing were previously thought to be increasing monotonically as countries develop (Uy et. Al., 2013), the rise of new world economic powers has been primarily determined by the rapid structural change of their economies, that is, the shift from mining and agriculture to manufacturing and then to skill and technology-intensive sectors (Olga and Lelio, 2010). Lewis (1954) emphasizes the need to transform the structure of an economy from low labour productive agriculture sector to the high labour productive modern industrial sector. In the least developed countries (LDCs), a large population depends upon traditional rural subsistence sector with surplus labour and hence, such surplus labour can be transferred to a highly productive modern sector in the process of development. Observing the happenings in the United States, Fuchs (1980) emphasized the importance of services sector in the economy, particularly, the changing patterns of employment, which grew across western economies as time passed. Likewise, Fuchs (1980) argues that to augment the contribution of services sector, it is required to increase participation of females in labor force as working-wives are likely to spend more out of their earnings to the services compared to males. Besides the development of primary and secondary sectors, Fisher (1939) advocated about the emergence of large services sector for the economic progress, also known as tertiary sector development. Later on, Clark (1940) established the Fisher's theme as a tertiary sector development model. Fisher-Clark approach of structural transformation explains that large amount of labour force working in the services sector will lead the country to the development and high-growth. The model proposes two significant factors in the emergence of service sector, i.e., high income elasticity of demand and low productivity of labor in services. Fisher-Clerk analogy is further supported by Cost Disease Hypothesis of Baumol (1967). This hypothesis argues that there will be shift to service from manufacturing due to low productivity, less progressiveness, higher costs and higher relative prices of service compared to manufacturing. 44 NRB ECONOMIC REVIEW In the stage of economic development, innovation led by dissemination and imitation seems to be most dominant factor for structural change of the economy (Schumpeter, 1939) and structural changes especially in specific industry are significant determinants of aggregate income and growth (Pender, 2002). Todays' advanced economies had followed two most prominent growth strategies, short-run strategy for stimulating growth, and a medium to long-run strategy to sustain that growth (Ocampo, 2003; Haggard and Kaufman, 1983). The emergence of international trade has also shifted the pattern of employment as we observe the decline in U.S. manufacturing employment as an effect of its trade with China (Autor, et.al., 2011). In addition, the gain received today by China and India from the external sector has been realized by the transformation of their economies. If they had not have emphasized on innovation and change towards industry and services, traditional garments and agricultural products would not have been sufficient to get advantage of international trade and investment to their economies (Rodrik, 2007). Nevertheless, the pattern of structural transformation varies with region, for instance, the path followed by developed economies and SAARC countries is different being heterogeneity in the transformation processes (Sawhney, 2010). III. CHANGES IN ECONOMIC STRUCTURE: GLOBAL AND NEPALESE SCENARIO 3.1 Global Change in Economic Structure As discussed earlier in section II, the structure of the advanced economies has a very low contribution of agriculture sector and predominance of service sector. Depending upon the individual economy, the contribution of industrial sector to GDP is found less than 50 percent from the beginning of study period, at the middle of agriculture and service. Likewise, the pattern of employment from agriculture, industry and services are similar in the contribution to GDP. A significant dominance of service sector in job opportunities has been observed as compared to the agricultural sector in most of the advanced economies (Figure 1 and Figure 2). In emerging economies, share of each sector to the GDP has been oriented to catch the path of advanced economies, though some countries are still far behind. It can be identified as a declining share of agriculture and increasing share of services to GDP over time. But sectoral contribution to employment has yet to be balanced with the contribution to GDP in these economies. Thus, the structure of developed and emerging economies shows a similar trend in contribution to GDP and employment, however, the perfect balance on employment and sectoral share can only be observed in the advanced economies (Figure 1, Figure 2 Figure 3 and Figure 4). This issue is also discussed on the trend of Nepal's abut neighbors, India and China in the following section 3.2. Structural Change and Per Capita Income in Nepal: Empirical Evidences 45 Figure 1: Economic Structure and Employment of United States 100 Figure 2: Economic Structure and Employment of Japan 80 USA: Value Added, % of GDP Japan: Value Added, % of GDP 60 50 40 20 0 0 1970 1975 1980 1985 1990 1995 2000 2005 2010 Agriculture 100 Industry 1970 1975 1980 1985 1990 1995 2000 2005 2010 Services Agriculture 100 USA: Sectoral Employment Share, % 50 50 0 0 Industry Services Japan: Sectoral Employment Share, % 1980 1985 1990 1995 2000 2005 2010 1980 1985 1990 1995 2000 2005 2010 Agriculture Industry Agriculture Services Industry Services Data Source: World Development Indicators, World Bank. Figure 3: Economic Structure and Employment of Philippines Figure 3: Sectorwise Contribution of GDP to the Economy of PCI Figure 4:Economic Structure and Employment of Thailand Figure 3: Sectorwise Contribution of GDP to the Economy of PCI Data Source: World Development Indicators, World Bank. 46 NRB ECONOMIC REVIEW 3.2 The Indian and Chinese Economic Structure Economic structures of two giant Nepalese neighbors namely China and India are substantially different than the structure of advanced economies. Although Chinese economy gives a different picture, Indian economy still possess some fundamental structural problem. In India, service sector has 57 percent share to GDP whereas agriculture sector accounts for 17 percent in 2012.The problem for India is observed in employment pattern as in Nepal. In 2010, for instance, the contribution of agriculture and service to GDP are 18 percent and 54 percent respectively but contribution to the employment of those sectors for the same year accounted for 51 percent and 27 percent respectively, indicating a low productivity in agriculture and little contribution of service sector towards the employment generation. Nonetheless in China, the contribution of industrial sector to GDP has been 47 percent in 2010 being continually the largest subsector of the economy but for providing employment, the sector is at the lowest level with 29 percent. Similarly, service sector in China has 43 percent shares in the GDP with 35 percent contribution to employment generation in 2010. Compared to Nepal and India, Chinese economic structure has been better in productivity and in employment generation. But the problem is in industrial employment in China; though the contribution to GDP is the highest, it is the lowest in employing population (Figure 5 and Figure 6). Figure 5: Economic Structure and Employment of China Figure 6: Economic Structure and Employment of India Data Source: World Development Indicators, World Bank. 3.3 Structure of Nepalese Economy Even though Nepal has sufficient natural resources and labour force among others, the historical average GDP growth rate is just about 3.7 percent since 1960 onwards Structural Change and Per Capita Income in Nepal: Empirical Evidences 47 (Figure 7). The industrial growth remained more unstable throughout the study period relative to the growth of agriculture and service (Table 1and Figure 8). Table 1: Standard Deviations of Nepal's Growth Sector Agriculture Industry Services Aggregate Full Sample 3.3 7.2 3.0 2.5 1991 Onwards 2.2 4.0 2.3 1.6 Source: Author's Calculation Figure 7: Real GDP Growth Rate in Nepal Figure 8: Sectoral GDP Growth Rate in Nepal Data Source: World Development Indicators, World Bank Data Source: World Development Indicators, World Bank. According to Nepal Labor Force Survey (NLFS) 1998/99, of the total labour force, 76 percent were engaged in agriculture, 10 percent in industry and 14 percent in service. After about one decade, as NLFS-2008 presented, 74 percent were in agriculture, 11 percent in industry and 15 percent in services – not much different from 1998/99. Nevertheless, the share of these three sectors to the GDP had changed over that period. In 1998/99, the contribution of agriculture sector was 38 percent, industry 23 percent and services 39 percent. The share of service sector in GDP jumped up to 48 percent in 2008/09, while the share of industry declined to 16 percent. The share agriculture declined marginally to 36 percent in 2008/09. In 2012/13, agriculture sector contributed 34 percent; industry 15 percent and service sector 51 percent. In this way, the contribution of service sector has been increasing while that of agriculture and industrial sector has been declining (Figure 9). In short, a gradual change is observed in economic structure since the share of services sector to GDP exceeded the sum total of agriculture and industry sectors so far. However, the major bottleneck in Nepalese economic transformation is employment pattern. It is believed that increased employment opportunities are the prerequisites for continued and sustained economic growth. In Nepal, nonetheless, we can observe a massive underemployment with very low productivity in agriculture. The opposite is the case of services as the contribution to economy is more than half but it provides employment for only 15 percent of work force. From the economic sense, however, the industrial sector is still playing vital role with closer similarities in contribution to both GDP and employment opportunities (Figure 10). 48 NRB ECONOMIC REVIEW Figure 10: Share of Sectoral Employment in Nepal Figure 9: Sectoral Contribution in the GDP in Nepal 100 80 14.1 9.8 15.3 10.8 76.1 73.9 60 40 20 0 1998/99 Agriculture Data Source: World Development Indicators, World Bank IV. 2008 Industry Services Data Source: Nepal Labor Force Survey, 1998/99 and 2008. DATA AND METHODOLOGY The paper follows the methodology of Pender (2002) to identify the determinants of structural change variables with slight modification. As Pender (2002) uses the concept with dynamic panel data analysis of OECD countries, the same technique has been adopted here only for Nepalese data to model ordinary least squares (OLS). Since per capita income of an economy is total production of the country within a year divided by the total population, factors that may cause to change income can be hypothesized as: PCI = f(agri_growth, ind_growth, ser_growth, edu, health, pop, popw, capital, others)…(1) Here, PCI refers to per capita income, agri_growth, ind_growth and ser_growth is the growth of three major economic sectors namely, agriculture, industry and service. The level of education (edu), health condition, total population and working population (pop, popw), capital injection and other variables are presumed to be the major determinants of per capita income. More precisely, based on this income hypothesis, the income model can be estimated as: 𝑃𝐶𝐼𝑡 = 𝛼 + 𝛽1 𝐴𝐺𝑅𝐼_𝐶𝐺𝑡 + 𝛽2 𝐼𝑁𝐷_𝐶𝐺𝑡 + 𝛽3 𝑆𝐸𝑅_𝐶𝐺𝑡 + 𝛽4 𝐸𝐷𝑈𝑡 + 𝛽5 𝐺𝐹𝐶𝐹𝑡 + 𝛽6 𝐿𝐸𝑡 + 𝛽7 𝑃𝑂𝑃𝑡 + 𝛽8 𝑃𝑂𝑃𝑊𝑡 + 𝜀𝑡 … (2) Where, per capita income (PCI) is the nominal annual US dollar per capita income in purchasing power parity. Growth rate of share of agriculture, industry and services are termed as AGRI_CG, IND_CG and SER_CG in the model, which is the percentage growth of sectoral contribution into the total Nepalese GDP, calculated as follows. Agricultural GVA 100} Total GVA t AGRI_CG = [{ Agricultural GVA 100} ] ×100-100 Total GVA t-1 { … (3) Structural Change and Per Capita Income in Nepal: Empirical Evidences 49 Industrial GVA 100} Total GVA t { Industrial GVA 100} ] ×100-100 Total GVA t-1 … (4) Services GVA 100} Total GVA t Services GVA 100} ] ×100-100 Total GVA t-1 … (5) IND_CG = [{ and SER_CG =[{ { Life expectancy (LE) is the expected years of life at birth, total population (POP) is the total number of population in million residing in the country and population at working age (POPW) is the population in million, with 15 to 64 years. The above data are obtained from World Bank Database. Gross fixed capital formation (GFCF) is the annual fixed capital formation in million rupees, obtained from national accounts statistics published by central bureau of statistics (CBS). Years of schooling (EDU) is the average year of schooling of working age population, calculated by multiplying currently available information of enrollment of the students ranging from primary school to advanced university degree that is obtained from Economic Survey (Various Editions). The augmented Dicky-Fuller (ADF) test for unit root has been presented below (Table 2). Table 2: Augmented Dickey Fuller (ADF) Test for Unit Root Variable AGRI_CG IND_CG SER_CG LOG(GFCF) EDU LE LOG (LE) Log (PCI) Log (PCI) @ Trend (AIC) POP Log(POPW) t-Stat -5.93 -4.59 -7.57 -0.78 -1.51 -2.90 -3.24 0.61 -3.303 0.72 -0.16 Level First Diff P Value t-Stat P Value 0.000 0.001 0.000 0.813 -6.786 0.000 0.518 -8.183 0.000 0.058 0.028 0.988 -6.475 0.000 0.086 0.991 2.822 0.066 0.934 3.134 0.034 Source: Author’s calculation The Augmented Dickey Fuller (ADF) test for unit root shows the variables AGRI_CG, IND_CG, SER_CG and LE are stationary at level whilst rests are found to be nonstationary. After first difference, log (GFCF), EDU, POP and log(POPW) become stationary. Nonetheless, per capita income (PCI) variable shows a trend stationary nature. When time trend is included in the test equation, PCI is found to be stationary at 10 50 NRB ECONOMIC REVIEW percent significance level and log(PCI) at 5 percent significance in even in level data (Table 2). Hence, to address the trend stationarity of PCI, the Hodric-Prescott (HP) filter is applied to extract trend and cycle from PCI. The HP filter generates new cycle and trend from the trend-stationary series that minimizes the variance of the old series around the new one, subject to a penalty constant . Once trend and cycle is extracted, we can use cycle in the regression equation. Hence, in case of PCI, the filter chooses PCI_Cyclet to minimize: ∑𝑇𝑡=1(𝑃𝐶𝐼𝑡 − 𝑃𝐶𝐼_𝐶𝑦𝑐𝑙𝑒𝑡 )2 + 𝜆 ∑𝑇−1 𝑡=2 ((𝑃𝐶𝐼_𝐶𝑦𝑐𝑙𝑒𝑡+1 − 𝑃𝐶𝐼_𝐶𝑦𝑐𝑙𝑒𝑡 ) − (𝑃𝐶𝐼_𝐶𝑦𝑐𝑙𝑒𝑡 − 2 𝑃𝐶𝐼_𝐶𝑦𝑐𝑙𝑒𝑡−1 )) … (6) By applying this method, the new series of per capita income, pci_cyclet, which is trendstationary free and contains all the information of PCI too. Based on Pender (2002)’s modeling framework and considering the nature of data and properties, the best fit model can be presented as follows. 𝑃𝐶_𝐶𝑦𝑐𝑙𝑒𝑡 = 𝛼 + 𝛽1 𝐴𝐺𝑅_𝐶𝐺𝑡 + 𝛽2 𝐼𝑁𝐷_𝐶𝐺𝑡 + 𝛽3 𝑆𝐸𝑅_𝐶𝐺𝑡 + 𝛽4 ∆𝐸𝐷𝑈𝑡 ++𝛽5 ∆ (log(GFCF))𝑡 + 𝛽6 log(𝐿𝐸)𝑡 + 𝛽7 ∆(𝑃𝑂𝑃)𝑡 + 𝛽8 ∆(log(𝑃𝑂𝑃𝑊))𝑡 + 𝛽9 𝑑𝑢𝑚01 + 𝜀𝑡 … (7) Equation (7) illustrates the prime factors in influencing per capita income of the citizens in an economy. In addition to the equation (2), one additional dummy variable is introduced. Dummy variable (dum01) is the variable with value one if the year of analysis is 2001 and zero otherwise, which is used to capture compilation break from 2001 as Nepal switched in accounting GDP with new system of National Accounts (SNA), 1993 with the broad categorization of the sectors especially that of services. The impact of sectoral growth variables is assumed to be positive for per capita income. Level of education, as explained by EDUt is also expected to increase the income since education is a human capital. Gross fixed capital formation is assumed to impact positively to income as capital is most significant factor for productivity increment and high growth. Effect of Life expectancy (LE) and population at working age (POPW) are also hypothesized to have positive impact on per capita income. Nevertheless, total population (POP) is presumed to reduce the income, as the population rises, income is to be distributed among citizens. There may be the possibility of multi collinearity among the regressors. To identify whether there exists serious collinearity problem, variance inflation factor (VIF) has been estimated. VIF helps quantifying the inflation of the variance due Structural Change and Per Capita Income in Nepal: Empirical Evidences 51 to the collinearity with other regressors in the estimated equation. The VIF factor for β̂i have been calculated as follows: 𝑉𝐼𝐹 = 1 1−𝑅𝑖2 …… (8) V. MODEL ESTIMATION AND RESULT ANALYSIS Equation (7) is estimated by applying ordinary least squares (OLS) method of estimation in EViews 8. The estimated coefficients of equation (7) have been presented in Table 3. Table 3: Empirical Results S.N. Variable Name Coefficient t-Stat 1. Constant - 166.24 -2.196** 2. AGRI_CGt 0.695 0.597 3. IND_CGt 0.936 2.251** 4. SER_CGt - 0.092 -0.139 -0.452 5. -5.017 EDUt 0.314 6. 8.32 log(GFCF)t 7. Log(LE)t 32.62 1.742* -2.081** 8. -93.12 POPt 3.431** 9. 3032.91 log(POPW)t 10. Dum01 50.71 3.088** *=significant at 10 percent level ** = significant at 5 percent or less level Adj. R2= 0.41, DW = 1.6, F-Stat = 3.60** Source: Author’s calculation. In contradiction to the hypothesis, coefficient of AGRI_CGt, represented by the growth rate of the agricultural share to the GDP, is found with positive sign and SER_CGt with a negative, both the coefficient are insignificant though. The IND_CGt, which represents the growth rate of industrial share to GDP, has expected sign and is significant at 5 percent level indicating that increased share to industrial GDP has a vital role in increasing per capita income. The reason behind the insignificance of agriculture sector could be justifiable. Agricultural productivity matters for other sectors development too, as very low agricultural productivity can severely damage modernization of economy (Kim & Whang, 2012). Moreover, Nepal's agriculture is largely at sustenance level, being high level of underemployment and only 40 percent farmer produce sufficient foods for one year's consumption (CBS, 2013)4. As discussed before in section three and proved statistically, the industrial sector is still playing vital role in raising per capita income only 4 Nepal Living Standards Survey-III reveals 32 percent underemployment in all sectors and share of wage employment in agriculture is just 2.8 percent compared to 12.6 percent in nonagriculture. Moreover, share of self-employment in agriculture is still 61 percent, only 10 percent down from 1996 level. 52 NRB ECONOMIC REVIEW due to the closer similarities in contribution to both GDP and employment opportunities. As the sector contributes about 15 percent to the economy by employing 11 percent of total employment, it is the closest combination in the share to employment and to the GDP so far. Nonetheless, there is also strong evidence on the insignificance of the service sector. The case of service sector is like the opposite the case of agriculture in terms of employment. Service sector's contribution to economy is more than half but it provides employment only for 15 percent of the total employed. With this, from income perspective, service sector's growth is still playing no role till date. Unless it absorbs the workforce at a speed of its growth and then its sectoral contribution to the national economy, it would not raise the living standards of societies. As hypothesized earlier, both life expectancy and population at working age have significant positive impact to per capita income; the coefficient of log(LE)t is significant at 10 percent level and log(POPW)t at 5 percent or lower level. These statistical results can be inferred as the improved health condition and young working groups foster the overall per capita income. As presumed before, increase in country's population reduces per capita income, POPt significant at 5 percent or lower level. The dummy variable, dum01 is significant at 5 percent or lower level. Hence, it has captured the compilation break of services sector in 2001. Nevertheless, education and gross fixed capital formation have been found insignificant to raise income. Although the sign of log(GFCF)t is positive as expected, the sign of EDUt is even negative. This contradictory finding, that is, the growth of industrial share to GDP is significant but capital formation is not increasing the income can be argued as follows. These two phenomena have been regressed with different scenarios, as the former indicates the sectoral growth in the share to the total production of the economy, and the later, with one of the factors of production that usually input for all three sectors in aggregate (agriculture, industry and services). Although capital injection may increase the productivity, the productive use of capital matters which may be suffering in Nepal (Bhatta, 2014) 5. Most importantly, capital injection should directly hit the income of the people, especially in employment creation; this might have missing in Nepalese context. On the other hand, the insignificance of the education variable as measured by years of schooling of working age population could be due to the couple of reasons. The increased number of outgoing migrants in the recent years (and impact of remittance on the education is still to witness), lack of labor movement from agriculture, being high share of underemployment, lower level of vocational trainings etc. may be impeding the role of education to the national economy. Besides, it is also witnessed a large chunk of educated unemployment in Nepal, educated youths being unable to get job due to the 'lack of access to relevant education and training, and lack of information' among others (United 5 This issue has been highlighted in an article at The Himalayan Times, March 11, 2014. Structural Change and Per Capita Income in Nepal: Empirical Evidences 53 Nations, n.d., and Sharma, 2013). The unemployed youths tend to get higher education, which is easily accessed without any qualifying exam restrictions in Nepal. Besides these all, there may the minor adjustment possibility of education data used in analysis since the education data is computed self. Nevertheless, change in sign with large variation in the coefficient couldn't be expected even after the revision of the series.6 So that, the insignificance of education and capital variable indicates that both the current level of education and capital injection have not contributed significantly to increase per capita income. Thus, it is essential to enhance the level of education and capital formation drastically in the days to come if Nepal intends to increase income of the people through education and investment as in advanced and emerging economies. This can be inferred on the basis of literature supports in the importance of capital, both human and physical, in OECD and other emerging economies. The goodness of fit, diagnostic and stability tests satisfy the minimum criteria required for the statistical inference. The Lagrange multiplier (LM) test for autocorrelation shows no serial correlation in residual as p-value of the test is 0.64. The residual plot of the model shows a random move around mean (Annex I - Figure 12). The stability test of the model is also significant since the recursive estimates represented by CUSUM and CUSUM squares test for stability lie within 5 percent range (Annex I - Figure 13).The adjusted R2, Durbin-Watson statistics and F-Stat for overall model significance show the satisfactory results. The VIF estimates for identifying the multi-collinearity among the regressors has been presented in Table 4. Table 4: Variance Inflation Factor (VIF) Estimates Variable AGRI_CG IND_CG SER_CG D(POP) D(LOG(POPW)) D(EDU) LOG(LE) DUM01 D(LOG(GFCF)) 6 Centered VIF 7.82 3.22 6.39 4.48 3.82 1.07 1.81 2.43 1.51 Source: Author’s calculation The negative sign of coefficient of EDUt is something weird in our estimation. Perhaps, a significant variation may not result even if the education data is revised. The estimated data are near to the official published series for the specific years. See annex II-A for the details. 54 NRB ECONOMIC REVIEW Generally, a very low value of VIF is the indication of no multi-collinearity problem, in which some researchers say only below 5 is the tolerable, for instance, Rogerson (2001). However, many researchers such as Neter et al., 1989: 409; Hair et al., 1995; Marquardt, 1970; Mason et al., 1989 have set the centered VIF below 10 as a tolerable limit for collinearity. In our VIF estimates, all the values of the centered VIFs are below 10. The VIFs of AGRI_CGt and SER_CGt have been found relatively higher but within the tolerable limit. The empirical findings, hence, suggest the requirement of an employment-generating economic growth. Even though we may achieve a higher sector-specific growth, the concern would be whether there is new employment generation. The message is that the balance of contribution to the GDP and to the total employment is a must for increasing income of people. VI. CONCLUSION Although the contribution of industrial sector does not change much in Nepal, historical data shows a gradual shift in the share of economy from agriculture to services. But the employment pattern has not changed in line with the change in sectoral composition of GDP. Unbalanced contribution of agriculture, industry and service sectors is found in the share of GDP and total employment. In Nepal, empirical estimates show that industry is the most significant sector to increase income compared to agriculture and service sectors. Improvement in health is also found significant to increase per capita income. Besides, working age population contributes to enhance per capita income of total population. Nevertheless, as against the theory and international empirics, capitals both human and physical have been found not contributing to raise per capita income, being investment and education variables insignificant in the empirical analysis. This could be because increased educated unemployment and lack of productive investment. The unbalanced contribution of employment, that is, high subsistence on agriculture and very low employment by the service sector could be blamed as the insignificance of these sectors in increasing the income. Hence, it is the major structural problem in Nepaldeviation in economic and the employment structure especially higher level of underemployment and eroded productivity in agriculture and employment unfriendly service sector. Industrial sector relatively observed better in increasing per capita income as the sector is much closer in employment generation and the share of the economic growth. Thus, employment generation is the utmost importance in an economy to raise the income of the people, so is for Nepal. In addition, improved health and larger share of working age people are also needed. The focus should be on increasing the productivity of the agriculture sector and move agriculture-based labors to other sectors of economy. Nonetheless, massive employment can only be generated with increased productive investment in the aforesaid sectors. Structural Change and Per Capita Income in Nepal: Empirical Evidences 55 The paper can be further improved by analyzing the panel data of similar economies that helps in identifying random and fixed effect estimations much comprehensively. ***** REFERENCES Acharya, S. P. and G. R. Bhatta. 2013. "Climate Change and Agricultural Growth in Nepal." NRB Economic Review, 25(2) : 1-16. Autor, D. H., D. Dorn and G. H. Hansen. 2013. "The China Syndrome: Local Labor Market Effects of Import Competition in the United States." American Economic Review, 103(6) : 21-68. Baumol, W. J. 1967. "Macroeconomics of Unbalanced Growth: the Anatomy of Urban Crisis." American Economic Review, 57(3) : 415-426. Bhatta, G. R. 2014. "Nepal's economic growth scenario: Boosting savings and investment." The Himalayan Times op-ed article, March 11. Available at http://thehimalayantimes.com/fullTodays.php?headline=Nepal%27s+economic+gro wth+scenario%3A+ Boosting+savings+and+investment+&NewsID=408304 Buera, F. J. and J. P. Kaboski. 2009. "Can Traditional Theories of Structural Change Fit the Data?" Journal of the European Economic Association, 7 : 469-477. CBS. 1999. Nepal Labour Force Survey 1998-99, Statistical Tables, Central Bureau of Statistics, Kathmandu. ____ 2009. Nepal Labour Force Survey 2008, Statistical Tables, Central Bureau of Statistics, Kathmandu. ____ 2013. "National Sample Census of Agriculture Nepal-2010/11." Central Bureau of Statistics, Kathmandu. CIA. 2013. The World Fact Book, 2013, Central Intelligence Agency, US. Accessed 13th November 13, 2013onhttps://www.cia.gov/library/publications/the-world-factbook/ geos/ np.html Clark, C .1940. The Conditions of Economic Progress, London: Macmillan. Economics Online. Accessed 9th November, 2013 on http://www.economicsonline.co.uk/ Global_economics /Structural_change_theory.html Fagerberg, J. 1994. "Technology and International Differences in Growth Rates." Journal of Economic Literature, 32 : 1147-117. Fisher, A. 1939. "Production: Primary, Secondary and Tertiary." Economic Record. Fuchs, V. R. 1980. "Economic Growth and the Rise of Service Employment." National Bureau of Economic Research. NBER Working Paper, No. 486. 56 NRB ECONOMIC REVIEW Haggard, S. and R. Kaufman, eds. 1983.The Politics of Economic Adjustment, Princeton, NJ: Princeton University Press. Hair, J. F. Jr., et. al . 1995. Multivariate Data Analysis, 3rd eds., New York: Macmillan. Harberger, A. C. 1998. "A Vision of the Growth Process." The American Economic Review, 88(1) : 1-32. Kim, H. J. 2006. "The Shift to the Service Economy: Causes and Effects." Institute for Monetary and Economic Research,. The Bank of Korea. Kim, J. H. and U. Whang, 2012. "Structural Transformation and Comparative Advantage: The Implications for Small Open Economies." Editorial Express. Accessed 15th October on https://editorialexpress.com/cgi-bin/conference/download.cgi?db_name= MWM2012& paper _id=95 Kuznets, S. 1971. "Economic Growth of Nations: Total Output and Production Structure." Political Science Quarterly, 86(4) : 654-657. Lee, D. and K. I. Wolpinm. 2006. "Intersectoral Labor Mobility and the Growth of the Service Sector." Econometrica, 74(1) : 1-46. Lewis, W. A. 1954. “Economic Development with Unlimited Supplies of Labor.” Manchester School of Economic and Social Studies, 22 : 139-91. Maddison, A.. 1991. "Dynamic Forces in Capitalist Development: A Long-Run Comparative View." Oxford: Oxford University Press. Marquardt, D. W. 1970. Generalized inverses, ridge regression, biased linear estimation, and nonlinear estimation, Technometrics, 12 : 591–256. Mason, R. L., R. F. Gunst and J. L. Hess. 1989. Statistical Design and Analysis of Experiments: Applications to Engineering and Science, New York: Wiley. Metcalfe, J. S. 2001. "Consumption, Preferences, and the Evolutionary Agenda." Journal of Evolutionary Economics, 11(1) : 37-58. Montobbio, F. 2000. "An Evolutionary Model of Industrial Growth and Structural Change". CRIC Discussion Paper, 34, University of Manchester. Nelson, R. 1995. "Recent Evolutionary Theorizing about Economic Change." Journal of Economic Literature, 33 : 48-90. Neter, J., W. Wasserman, and M. H. Kutner. 1989. Applied Linear Regression Models. Homewood, IL: Irwin. Noland, M., D. Park and G. B. Estrada. 2012. "Developing the Service Sector as Engine of Growth for Asia: An Overview." Working Paper No. 320, Asian Development Bank. Retrived from http://www.adb.org/sites/default/files/publication/30080/ economics-wp320.pdf on 10 May, 2014. Ocampo, J. A. 2003. "Structural Dynamics and Economic Growth in Developing Countries.” United Nations Economic Commission for Latin America and the Caribbean (ECLAC), Santiago: Chile. Structural Change and Per Capita Income in Nepal: Empirical Evidences 57 OECD. 2010. "The Service Economy." Organization for Economic Cooperation and Development. Olga, M. and A. Lelio. 2010 . "Structural Change in the World Economy: Main Features and Trends." WP 24/2009, UNIDO, Vienna. Pasinetti, L. L. 1981. Structural Change and Economic Growth, Cambridge University Press, Cambridge. ISBN 9780521274104. Pender, M. 2002. "Structural Change and Aggregate Growth." WIFO Working Papers, No. 182, Vienna accessed 3rd October 2013 onhttp://www.wiwi.unijena.de/Mikro/pdf/ peneder-281101.pdf accessed 12/1/2013. Rodrik, D. 2007. One Economics, Many Recipes: Globalization, Institutions and Economic Growth. Princeton University Press. __________. 2013. IMF Survey Magazine. June 28 2013.Accessed 10thNovember 2013 on http://www.imf.org/external/pubs/ft/survey/so/2013/INT062813A.htm Rogerson, P. A. 2001. Statistical methods for geography, London: Sage. Sawhney, U. 2010. "Growth and Structural Change in SAARC Economies." International Journal of Economics and Finance Studies, 2(2), ISSN: 1309-8055 (Online) Schumpeter, J. A. 1939. Business Cycles: A Theoretical, Historical, and Statistical Analysis of the Capitalist Process, New York and London: McGraw-Hill. Sharma, S. 2013. "Educated, unemployed and mobile." The Kathmandu Post. September 29. Available at http://www.ekantipur.com/the-kathmandu-post/2013/09/28/free-thewords/educated-unemployed-and-mobile/254166.html Silverberg, G. 1998. "Modeling Economic Dynamics and Technical Change: Mathematical Approaches to Self-Organization and Evolution." in Dosi, G., Freeman, C., Nelson, R., Silverberg, G., Soete, L. (eds.), Technical Change and Economic Theory, 531-559. Silverberg, G., and B. Verspagen. 1998. "Economic Growth: An Evolutionary Perspective", in Reijnders, J. (ed.), Economics and Evolution, Edward Elgar, Cheltenham, 137-170. United Nations (n.d.). "Under and Unemployed Youth (15-29 years)." United Nations Nepal Information Platform, Electronic version. Available at http://un.org.np/oneun/ undaf/ unemployed Uy T., K. Yi, and J. Zhang. 2013. Structural Change in an Open Economy, University of Michigan. March 12. Verspagen, B. 2001. "Economic Growth and Technological Change: An Evolutionary Interpretation." STI Working Papers, OECD, Paris, 1. Welker, J. 2012. Models of Economic Growth and Development. Online version accessed on 2nd November, 2013 fromhttp://welkerswikinomics.com/blog/2012/01/30/ models-for-economic-growth-ibeconomics/World Development Indicators. 2013. World Bank Database. 58 NRB ECONOMIC REVIEW Appendix I Figure 11: Hodrick Prescott Decomposition Figure 12: Residual Plot of the Model 40 20 0 20 -20 10 -40 0 -10 -20 -30 1980 1985 1990 Residual 1995 2000 Actual 2005 2010 Fitted Figure 13: Stability Tests of the Model 10.0 1.6 7.5 1.2 5.0 2.5 0.8 0.0 0.4 -2.5 -5.0 0.0 -7.5 -10.0 -0.4 2002 2003 2004 2005 2006 CUSUM 2007 2008 2009 5% Significance 2010 2011 2002 2003 2004 2005 2006 CUSUM of Squares 2007 2008 2009 5% Significance 2010 2011 Structural Change and Per Capita Income in Nepal: Empirical Evidences 59 Appendix II A. Data on Average Years of Schooling for Population and Working age, 15-64 years Year 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Total Population, Million Population at Working Age, Million Average Years of Schooling 12.87 13.16 13.45 13.75 14.06 14.38 14.72 15.06 15.42 15.78 16.14 16.51 16.89 17.27 17.68 18.11 18.57 19.05 19.55 20.07 20.59 21.12 21.65 22.18 22.69 23.18 23.66 24.10 24.53 24.92 25.29 25.63 25.95 26.25 26.54 26.85 27.16 27.47 7.16 7.31 7.46 7.62 7.78 7.95 8.12 8.30 8.48 8.67 8.85 9.03 9.21 9.39 9.60 9.83 10.10 10.39 10.71 11.03 11.36 11.69 12.02 12.34 12.65 12.95 13.22 13.47 13.72 13.96 14.20 14.45 14.69 14.96 15.24 15.56 15.92 16.31 0.86 0.93 1.05 1.15 1.34 1.42 1.47 1.54 1.60 1.55 1.70 1.67 1.73 1.83 2.07 2.31 2.40 2.47 2.41 2.40 2.37 2.47 1.89 2.50 2.58 2.50 2.63 2.66 2.74 2.87 3.01 2.98 3.01 3.13 3.28 3.22 3.21 3.09 UNDP Data7 0.6 2.0 2.4 2.7 2.8 2.9 3.0 3.1 3.2 3.2 3.2 Data Source: Total Population and Population at Working Age Data is downloaded from World Bank Database and Average Years of Schooling Data is self-calculated by using school enrollment data available fr om Economic Survey,. Various Issues. 7 Average number of years of education received by people ages 25 and older, converted from education attainment levels using official durations of each level. The data is put here for the reference, obtained from United Nations Human Development Indicators Accessed: 2/25/2013, at http://hdr.undp.org. 60 NRB ECONOMIC REVIEW B. Other data series used in analysis Year 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 GFCF Million, NPR 2223.0 2443.0 2580.0 3294.0 3263.0 3681.0 4299.0 5465.0 6576.0 6907.0 9386.0 9431.0 11825.0 13414.0 16392.0 17002.0 22780.0 29277.0 37278.0 42032.0 48370.0 56081.0 60794.0 65375.0 65269.0 73324.0 84750.5 89889.3 98072.8 109181.3 117538.9 135532.0 153336.9 178445.5 211039.0 264888.0 292730.0 307384.0 IND_G NA 8.301994 26.50899 6.529822 0.370499 -0.518811 3.736784 3.999741 -0.467740 -1.562553 19.92926 4.953289 -0.174051 2.159217 2.274153 -1.812131 6.940153 17.47837 1.290158 5.101934 4.783464 0.710507 -0.228906 -1.624526 -3.036574 1.480300 -19.60559 1.672383 0.315668 -1.603388 -0.900514 -2.815206 -0.604771 1.403537 -5.551835 -4.502185 -1.939240 -1.552970 SER_G NA 8.890165 13.87903 0.561690 -5.285280 11.00017 1.614016 -2.237120 2.925061 -1.819503 25.63819 -1.540583 2.383194 -1.540314 0.471022 -2.897028 10.24382 -1.934237 6.833271 -5.102471 0.767073 0.258643 0.351942 5.315892 -1.869622 0.404737 20.28716 -2.793550 2.293923 1.482795 2.190292 4.798017 2.461346 1.198472 -0.686357 -3.374023 -2.733647 4.311903 AGRI_G NA -3.432291 -7.764521 -1.363254 2.027867 -3.956940 -1.408213 0.170157 -1.154514 1.143192 -15.21721 -0.458424 -1.458986 0.342023 -1.027216 2.497816 -8.557058 -4.976672 -5.877285 1.987817 -3.039191 -0.606958 -0.175306 -3.684482 3.473210 -1.143537 -7.782908 2.516736 -2.723048 -0.975243 -2.217146 -4.695526 -3.121590 -2.477033 3.987479 7.327898 4.408893 -4.645298 PCI, PPP USD 281.6 287.6 289.8 295.9 296.3 282.9 327.6 352.5 347.3 386.2 412.9 431.6 441.7 481.3 509.1 539.7 578.9 599.8 620.3 667.4 688.7 719.7 751.7 766.6 793.7 842.8 885.3 884.0 922.1 976.6 1028.9 1083.1 1138.5 1220.6 1272.8 1336.8 1402.1 1484.3 PCICYCLE (HP Filtered) 19.05878 15.29110 7.541765 3.168187 -7.985212 -34.18853 -4.086187 4.321822 -19.03958 -0.192662 4.774010 0.195859 -14.24678 -0.560529 0.312850 3.265048 14.31836 6.774899 -1.438216 16.79517 9.083011 10.83688 12.95151 -2.929753 -7.870098 7.363763 13.84589 -26.02243 -29.71056 -20.39875 -16.86215 -14.64082 -14.19630 10.54934 3.586678 7.142556 11.21697 31.97413 Note: The data of 1975 represents the Nepal's fiscal year 1974/75, 1976 as fiscal year 1975/76 and so on in the whole data sets. Data Source: World Development Indicators, the World Bank and Central Bureau of Statistics, Kathmandu, Nepal. Determinants of Inflation in Nepal: An Empirical Assessment Shoora B. Paudyal Ph.D. Abstract This paper examines short term and long term effects of the macroeconomic variables on the inflation in Nepal during 1975-2011. The variables considered are budget deficits, Indian prices, broad money supply, exchange rate and real GDP. The regression results from Wickens-Breusch Single Equation Error Correction model suggest that all variables considered are significant in long run implying that these variables are the determinants of inflation in Nepal. However, only budget deficit, money supply and Indian prices cause inflation in the short run. The results are consistent with monetarists’ hypothesis of money matters and inflationary gap theory of Keynesian as well as supply constraints approach to inflation. Key Words: Inflation, Budget Deficits, Single Equation ECM, Imported Prices, Consumer Price Index, M2, GDP JEL Classification: E31; C22 Dr. Paudyal is a faculty member of economics, Tribhuvan University, Kathmandu. Email: [email protected] The author acknowledges the valuable comments from Editorial Board and external reviewers on the original version of this paper. 62 NRB ECONOMIC REVIEW I. INTRODUCTION A continuous rise in price level is termed as inflation (Parkin, et al, 1986: 367). Inflation is an ongoing process whereby prices are rising persistently year after year. Shapiro (1996:468) defines inflation as a rising price level. If such a rise in price level persist for long it is known as inflation. Consumer price index, gross domestic product deflator and other several indices measure the changes in price level. The use of these measures is purposely applied wherever appropriate. However, the rate of percentage change in consumer price index as a measure of inflation is widely used. We also here adopt this definition of inflation for our purpose. Inflation is everywhere and is interestingly touchy issue in macroeconomics. All daily newspapers cover the news about inflation. There is no dearth of literature on inflation. It is the mostly discussed issue all over the world among policy makers and academia. It is because of the fact that its effects are widespread and severe and the impacts are far reaching. Inflation has been the major concern for the government since it has serious implication for the living of common peoples. Moreover, it affects several macroeconomic variables such as saving, investment, real interest, real wage, real income and level of employment. Inflation depreciates domestic currency and the imports become more expensive which further push up the domestic prices. In short, inflation is a burning issue in the macroeconomics and main objective and function of central bank is to control inflation. Table 1: Consumer price index Year CPI Year CPI Year CPI Year 1974/75 9.8 1984/85 20.8 1994/95 59.7 2004/05 1975/76 9.6 1985/86 24.7 1995/96 65.2 2005/06 1976/77 10.5 1986/87 27.4 1996/97 67.7 2006/07 1977/78 11.3 1987/88 29.9 1997/98 75.4 2007/08 1978/79 11.7 1988/89 32.5 1998/99 81.0 2008/09 1979/80 13.4 1989/90 35.2 1999/00 83.0 2009/10 1980/81 14.9 1990/91 40.7 2000/01 85.2 2010/11 1981/82 16.6 1991/92 47.6 2001/02 87.8 1982/83 18.7 1992/93 51.2 2002/03 92.8 1983/84 19.2 1993/94 55.5 2003/04 93.6 Source: IMF, International Financial Statistics, various issues base year 2005=100 CPI 100.0 107.6 114.1 126.6 141.3 155.4 170.2 Table 1 shows that price index (base year 2005) has increased persistently over the years. It has increased by little over seventeen times (from 9.8 to 170.2) during 19752011. This means the purchasing power of the rupee has eroded by the same speed. Aforementioned, the impact of rising prices on the real sector is stylised fact. It constrains the rise of per capita real GDP and thereby reduces the standard of livings of the common people in the country. The stationary price level has thus been one of success parameters of the elected government. However, it has been a Herculean task to achieve in Determinants of Inflation in Nepal: An Empirical Assessment 63 developing countries. In case of Nepal, however, there appear some positive signals in slowing down the speed of price rise in the later years. For instance, CPI took ten years to double from 9.8 in 1975 to 20.8 in 1985; it doubled even faster within six years between 1985 and 1991 and within eight year period between 1991 to 1999. This has, however, turned upside down since the doubling period lengthened to twelve years between 1999 and 2011. This clearly indicates that prices have accelerated at slower motion especially after 1991’s political change. One of the reasons for this might be relatively improved supply situation of the commodities during this period. Partly because Nepal’s improved bilateral relation with India in the changing context and partly because of the sharply improved trade openness index due to trade liberalisation policy adopted by the government during early 1990s (Bowdler, et al; 2004)8. Some empirical studies substantiate that trade openness index has important bearing on the combating hyper inflation. This paper attempts to examine the relation between inflation and other related variables that influence the inflation in the country and suggest policy implications. There may be a bunch of factors that may influence the inflation. In Nepal, price level, budget deficits, money supply, real GDP are continuously rising for many years. However, this does not prove that one causes other. We examine in this paper effects of a number of variables including budget deficits on prices dividing the paper into six sections: Section I is introduction, section II presents a literature review, and section III analytical framework, while section IV displays regression results and interpretation. The last section V concludes the paper. II. REVIEWS OF LITERATURE This section is divided into two parts: a) review of theoretical foundation and b) reviews of empirical studies. The theoretical review shape the relationship between inflation and other macroeconomic variables under classical/monetary hypothesis, Keynesian, New Keynesian and New Classical theories, while the empirical review examines both the short run and long run relationship between the variables under consideration. The classical economists’ famous quantity theory of money can be summarised in MV=PT, in which velocity of money in circulation (V) and quantity of goods (T) remain constant in the short run. So an increase in stock of money (M) brings a proportionate rise in price level (P). This equation became the foundation of monetarists economic thought. The monetarist economists opine that too much money chasing too few goods is the cause of inflation. As per the modern quantity theory of money, demand for money is given by 1/v* PQ. If the economy is at full employment, real income does not change and v being equals, there is a direct and proportionate relationship between changes in quantity of money and price level. The central bank prints more paper notes that directly causes inflation. But under the less than full employment economy, real income does not remain constant. So, v being constant if real income increases by 4% together with a 10% increase in money supply implies a 6% increase in inflation- a less than proportionate rise in price level. 8 They claim that greater trade openness decreases the probability of inflation. 64 NRB ECONOMIC REVIEW Historically, it is seen that if deficits are financed by money created by central bank that increases inflation directly (Friedman,1984). Friedman opines strongly that inflation is always and everywhere a monetary phenomenon. Monetarists including Friedman believe that money supply has direct effects on the inflation i.e., too much money to chase a few goods causes inflation. This implies that inflation is the function of money supply and real output. Friedman (1968) believed that central bank can control the inflation in the long run by controlling money supply. It is now well accepted that the primary goal of central bank is achieve a stable and low rate of inflation. However, the governments in many countries seem fail to support such contractionary monetary policy because of their temptation toward achieving higher economic growth and employment. Figure 1 and Table A5 in appendices show the growth of money supply and that of prices in the country. Both move in the same direction but broad supply has alway been higher than inflation. High positive correlation between money supply and budget deficits (Appendices, Table A1) exhibits a very close association between these two variables. Keynesian views of stimulating economy through aggregate demand often lead to a situation where government spending exceeds the mobilisation of taxes. It becomes thus a customary for the government from the developing economies to present deficit budget. However, it is widely believed that higher budget deficits have serious impacts on the macroeconomic variables. Among these most striking effect of budget deficits is on inflation. For this reason, budget deficit is largely considered as one of the major determinants of the inflation. Deficit compels the central bank to print new money to finance the budget deficit which led to an increase in money stock9. Deficits can be financed by other sources such as borrowing from the market which also indirectly affects inflation. Moreover, many believe that such effects of the budget deficits pass through interest rates to the major macroeconomic variables such as investment, employment, price level, consumption, exports and imports in an economy. 9 See Malcolm Gillis, Dwight H Perkins, Michael Roemer and Donald R Snodgrass. 1983. Economics of Development, Second edition, p326. A country’s money supply may be defined as the sum of all liquid assets in the financial system. Determinants of Inflation in Nepal: An Empirical Assessment 65 Because, the additional resources that are borrowed from the financial market by the government to finance the budget deficits which may crowd out the private investment. This increases the demand for credit, other things being equal; this will lead to a higher interest rate. Its immediate effects will be seen in the reduced private sector investment and consumption that implies lower aggregate demand, which virtually results in a lower income and employment. On the other hand, reduced private investment will result in lower aggregate supply. So, central bank generally attempts to monetise deficits to control the interest rate from rising. For this, it will purchase back the government securities pumping the new money in the market. This generally pushes up the price level through increase in money stock. The monetary authority adopts thus the expansionary monetary policy that nullifies the rise in interest rate by increasing money supply but it causes inflation. So, deficits have indirect effects on the inflation through neutralised interest rate. This is consistence with a study that finds budget deficits in Nepal are interest rate neutral (Paudyal, 2013). However, the growth of budget deficits has been lower in later years but not the growth of money supply. This indicates that deficit is not only the cause of an increase in money supply. The rise in budget deficits in the annual budget of the government of Nepal has been substantial over the years. The amount of budget deficits has increased to Rs 496 billion in 2010/11, which accounts for about 4% of GDP but this is lower than 5.5% in 2000/01 and about 10% in 1988/89 (NRB, 2010; MoF, 2012). This clearly indicates that budget deficits, though still at upper ladder, has substantially declined in recent years. Figure 2 reveals that the relationship between inflation and deficits. The deficits as a percent of GDP almost coincided with inflation only for few years (1989/90, 1992/93 and others), but for the rest of years inflation departs either downwards or upwards of deficits. For instance, in 1979/80 deficit was little over 3% of GDP, but inflation rate was about 15%. This means inflation exceeds deficits by 12%. In 1982/83, deficits rose to 9% of GDP but inflation rate slipped to around 12% from the level of 1979/80 (see Table A2 in Appendices). The gap between two rates became only 3%. But interestingly they moved to opposite direction, that is, the increment in deficits appeared with reduction in inflation. Both declined in the next year (1983/84), but deficits decreased marginally by 66 NRB ECONOMIC REVIEW one percent point (from 9% to 8%) while prices fell dramatically by over nine percentage point (from 12% to 2.8%). See Table A2 in Appendices. This indicates that deficits have not neck to neck relationship with inflation. However, two rates follow a similar positive trend for most of the years during the study period. To note, inflation rate exceeds deficits for the most of the years. It may be indicative that deficits are not only factors influencing inflation. However, the positive association between budget deficit and prices is high enough (Table A1). Some empirical studies such as Vuyyuri (2004) for developing economies suggest a strong influence of budget deficits on inflation rate. However, others such as Blanchard (1989) claim that budget deficits and inflation rate rarely show a positive association. Figure 6 in Appendices also indicates that the growth rates of money supplies and domestic borrowing through securities to finance budget deficits are associated to each other. However, the government borrowing is highly fluctuating against the relatively stable money supply. The reason for the fluctuation might be explained in association to the disbursed amount of external loan and absorptive capacity of bureaucracy in the country. The growth curves of deficits and domestic borrowings reveal that they move concurrently. The qualitative discussion clearly reveals that some variables seem to have very close link between them and others contain some distant relationship. But almost all of the variables mostly move in the same direction and exceptionally they move in opposite direction. Figures 7, 8, 9 and 10 in appendices show that the positive association between growth rates of broad money and budget deficits, that of borrowing and budget deficits, that of debt and deficits, and lastly relationship between three variables namely deficits, narrow money supply and inflation rate respectively. Once Keynes says, “in the long run we all are dead.” So, Keynesian economics focuses on the analysis of the short run behavior of the economic variables. For Keynesians, only aggregate demand matters. To them, in the short run, price level is determined by the aggregate demand and fixed aggregate supply. Any rise in aggregate demand in the short run thus causes inflation. For this reason, any pressure of the excess demand over fixed quantity of aggregate supply of goods and services causes inflationary gap. This leads to demand pull inflation (Shapiro, 1982). Keynesian Philips curve comfortably explains tradeoff between inflation rate and unemployment rate, that is lower the unemployment rate higher the inflation. Keynesian aggregate demand (AD) policy to increase employment leads to a wider inflationary gap in the short run as aforementioned output cannot be increased immediately. Up to this point inflation was thus the resultant of demand shocks. The oil shock of 1974 created a new problem of stagflation i.e. higher inflation appears with lower employment/real output (Shapiro, 1982). In other words, the world faced a new type of economic problem of hyper inflation and recession simultaneously that could not be explained by the old Keynesian Philips curve. Many countries faced double digit inflation rate in 1974 and 1979/80 due to supply shocks originated outside the country Determinants of Inflation in Nepal: An Empirical Assessment 67 (Shapiro, 1982). This suggested that supply shocks may have a dominant role in determination of inflation. The supply shocks generated outside the country cannot be controlled by the domestic policy. In addition, it is seen that past experiences on inflation may influence the future expected rate of inflation. This pushed the Keynesian economists to formulate new theory which is popularly known as New Keynesian economics that claims that inflation rate is determined by demand pull inflation, cost push and built-in-inflation (Gordon, 1988)10. The supply shocks from within and outside the domestic economy are accepted as one of the major causes of inflation. The economy wide shortages of goods and services cause the cost push inflation, that is, supply bottleneck causes this inflation. The cost push inflation may have many sources. One of these is the rising cost of production which causes inflation. The reasons behind such rise in cost may be either wage push, profit push or supply shock due to crop failure or lockouts or drought or foreign blockade, or war or supply shocks originated from outside country such as OPEC oil price rise in 1974 and 1979/80 (Shapiro, 1982). Supply shocks typically become rule, where domestic production shares small proportion of the total supply of the commodities in the market and agriculture largely depends on rain water in a small landlocked country like Nepal. Supply constraints, to some extent, might demonstrate higher domestic prices as compared to the international imported prices. The built-in-inflation is another component of current inflation is the carryover from the past events and persists in the present time. The past events in the built in inflation may be persistence of either demand pull inflation or large cost push inflation or both in the past. Besides, inflationary expectations theory and conflict theory of inflation also contribute to the built in inflation. The past experiences of high inflation rate lead to the higher current inflation. The subjective judgments of workers and employers that current inflation will persist in future primarily push them to make an agreement to increase the prices of goods and money wages that causes built-in-inflation. The conflict theory of prices caused by wage-price spiral states that demand for higher money wages to protect real wages leads to built-in-inflation. This is also based on the subjective judgment of the workers. An increase in money wages pushes the prices upward and the purchasing power of the money again fall down. The built-in-inflation in New Keynesian seems to have largely influenced by the New Classical theory of rational expectation. The good thing is that this notion brings two major schools of thought-Keynesian and Monetaristthat differ for the century closer to each other. 10 Gordon built triangle model of New Keynesian Theory of Inflation. 68 NRB ECONOMIC REVIEW Figure 3 shows that growth in prices and real GDP move in the same direction. The growth of prices is higher for almost years and that of real output seems to be lower for almost all years. However, the high positive correlation coefficient between prices and real output (Table A1) demonstrates strong association between two variables. The Keynesian view that a modern capitalist system is not self-regulating and thus requires government intervention has given the rebirth of classical theory of economics (Cornwall, 1990:47). Among the advocates of New Classical theory of rational expectation, Lucas (1976) is in forefront. Lucas theory is very close to rational expectation in financial (stock) markets. This theory states that the predictions of economic agents about the future value of the variables are not systematically wrong in that all errors are random. This theory of inflation argues that rational expectation is the key cause of inflation. The actual price will only deviate from the expectation due to information shock caused by unforeseeable information at the time of expectation. So, actual price is the sum of expected price and error term. The control of inflation largely depends on the credibility, integrity and autonomy of the central bank. The economic actors look rationally into the future and try to maximize their well-being. They carefully watch the activities of the central bank and make their own perception about the decision of monetary authority and future expectations about the rate of inflation. Accordingly they frame their own strategies for maximization of benefits. They hardly believe the central bank which remained soft in the enforcement of its own decision in the past. In such backdrop, central bank policy to curb the inflation fails rather than succeed because market actors will hardly believe in the enforcement of the decision and so they expect higher inflation in future. In short, Keynesian aggregate demand policy is considered as appropriate to reduce unemployment but inappropriate for regulating inflation (Cornwall, 1990:5). Rational expectation theory has some virtues and getting popularity explaining inflation. Modern macroeconomics theory posits that the output supplied depends on the unexpected movement in price level as well as on the actual and unexpected technological shocks (Blanchard, et. al, 1989: 519). So, inflation is not only the effect of demand and supply shocks but also is the cause of supply. Determinants of Inflation in Nepal: An Empirical Assessment 69 Some empirical studies such as Pahlavani (2009) find that even the international inflation and expected inflation have influential bearings on domestic inflation. Others such as Khan (2007) constructed econometric model to study inflation incorporating fiscal and monetary policies of the government. In reality, evident from empirical studies that several factors including money supply play roles in macro-economy (Kennedy, 2012: 189). McMillan (1986) finds that budget deficits cause inflation (Vuyyuri, et. al, 2004). In a study for Pakistan’s inflation, Khan (2007) finds that the most important determinants of inflation in 2005-06 were adaptive expectations, private sector credit and rising import prices whereas, the fiscal policy’s contribution to inflation was minimal. Bayo on the study for Nigeria reveals that fiscal deficits, money supply, interest and exchange rates are cause of inflation in Nigeria during the period under review. Pahlavani (2009) states that inflation in Iran is largely determined by money supply, exchange rate, GDP, expected inflation rate and imported inflation along with dummy variable. Kumar (2013) finds that money supply and imports index is the most important variables in explaining inflation in India while Laryea (2001) states that inflation in Tanzania is largely influenced by monetary factors both in the short run or the long run. A study for NRB notes that Indian prices have a significant bearing on variation of domestic prices in the country (NRB, 1994:100). Besides, they find that an increase in money stock causes price rise and the gradual depreciation of the exchange rate of domestic currency has been partly responsible for the price rise in Nepal. Thapa (2010) attempts to study the determinants of inflation in Nepal. A study by Neupane (1992) finds that one year lagged money supply, cost of holding real balances, budget deficits, low output growth rates and import prices are the important determinants of price inflation in Nepal. NRB (2001) reveals that there is no structural shift in money price relationship in Nepal. This study finds that broad money has stronger relationship with inflation compared to narrow money. Mathema (1998) finds that a rise in wages in industrial sector causes national inflation while Koirala (2008) discloses a significant relationship between inflation and inflation expectations in Nepal. Koirala (2013) again finds non-constant time varying parameters of both the constant and autoregressive of order one AR(1) coefficient of inflation over the long run. He opines that the changes in the expectations of rational economic agents on macroeconomic policies due to the lack of policy commitment, credibility and dynamic consistency might have contributed for this. Empirical studies in several other countries have shown that a set of explanatory variables such as real gross domestic product (RGDP), budget deficits (BD) or government expenditure, exchange rate (EXC), imported price (MP), broad money (M2) and expected inflation (Pe) explain the variation in price level (P). We examine these variables to explain the inflation in Nepal. In our case, CPI series is a measure of price level and among the independent variables the budget deficits after grants (BD), real GDP, the liquidity or money supply (M2) and the imported price is consumer price index of India (CPII) since Nepal’s imports from India accounts about 66 per cent for 2013, which includes goods of daily consumption such as vegetables, clothes, medicines, transport equipments and petroleum products (MoF, 2014). 70 NRB ECONOMIC REVIEW Two curves in Figure 4a shows that the movement of the price series in Nepal and India. The coincided curves to each other imply that they are very close mates and move together. They not only move to the same direction but also almost coincides each other. This implies that these two variables are perfectly co-related. Figure 4b presents the inflation rates in Indian and Nepali economies. This clearly shows that inflation, with an exception for a few years, is always higher in Nepal compared to India. Table A1 in appendices shows a strong positive association between imported price and inflation. Besides, it also reflects the higher dominance of imported Indian goods in the domestic market. Determinants of Inflation in Nepal: An Empirical Assessment 71 Two curves in Figure 5 demonstrate that how closely last year price and current year price move together. The inflationary expectation in future is based on past inflation in built-in-inflation of new Keynesian economics that affects the current inflation. This may be examined with the help of the one year lagged cpi variable in the Nepali context. The lagged cpi will be interpreted as adoptive expectation. III. ANALYTICAL FRAMEWORK We discuss about the methodology in this section. This study covers time period 19752011. The data series on CPI for India and Nepal are retrieved from the international financial statistics of IMF. The rest data series are extracted from the quarterly economic bulletin of Nepal Rastra Bank and from various issues of Economic Survey published by Ministry of Finance of Nepal government. We use Wickens-Breusch Single Equation Error Correction model for the analysis of the data discussed somewhere in this section. The regression of a non-stationary time series on a set of non-stationary time series can likely produce spurious results. But the difference of the non-stationary time series is individually stationary. However, the regression results from difference variables may results in loss of information about long run relationship between the variables. So it necessitates the regression of variables at level. Sometimes, the linear combination of the two or more non-stationary time series gives stationary results (Pindyck,et al, 1991, Enders, 2014, Gujarati, 2007). So, it is customary these days to test whether time series of economic variables are individually stationary followed by co-integration test of these variables at level. Dickey-Fuller or Augmented Dickey-Fuller test are generally apply to see whether such time series are individually stationary. The time series of economic variables are shown non-stationary or follow the random walk at level I(0) and stationary at first difference, I (1). The results from the augmented DF test at level and first difference are given by Table 2. 72 NRB ECONOMIC REVIEW Table 2: Augmented Dickey-Fuller tests at the level and at the first difference Level CPI rGDP BD EXC -1.37555 -3.012338 -2.53712 0.217641 First Difference -5.175140** -7.498887** -5.869865** -4.87361** CPII -1.405367 -6.564345** M2 -1.98365 -3.872057* **, * significance at 1%, 5% level 1% 5% 10% -4.243644 -4.243644 -4.243644 4.243644 -3.54033 -3.54033 -3.54033 -3.544284 -4.243644 -4.243644 -3.54033 -3.54033 -3.20244 -3.20244 -3.20244 3.204699 -3.20244 -3.20244 Order of Integration I(1) I(1) I(1) I(1) I(1) I(1) Augmented Dickey-Fuller tests (1979, 1981), shows that variables such as logarithm of CPI, RGDP, BD, EXC, CPII and M2 series are non-stationary at level but stationary at first difference as expected. So, we can regress difference of logarithm of prices (CPI) on the difference of logarithm series of budget deficits (BD), real gross domestic product (RGDP), exchange rate (EXC), Indian prices (CPII), and broad money supply (M2). Besides, Johansen co-integration test results presented in Table 2 suggest that these variables are co-integrated with dependent variable. It implies that there exist long run relation between inflation and other variables. This means that these variables can be regressed at level form also although individual series are stationary only at first difference. The coefficients from the regression of the variables at level measure the effects of independent variables on the dependent variable in the long run. Table 3 reveals that there are multiple cointegrating equations but we apply only one for the analysis of the data. Table 3: Johansen co-integration test at level [series CPI , GDP, BD, EXC, M2 and CPII] Trace test Critical Max-Eigenvalue Critical value statistics value 0.5% statistics 0.5% None 129.4204* 95.75366 45.56497* 40.07757 At most 1 83.85546* 69.81889 33.09859 33.87687 At most 2 50.75687* 47.85613 23.18014 27.58434 At most 3 27.57673 29.79707 16.28424 21.13162 At most 4 11.29249 15.49471 8.732777 14.26460 At most 5 2.559717 3.841466 2.559717 3.841466 Note:*Trace test indicates 3 cointegrating equations while max-eigenvalue test indicates 1 cointegrating equation at the 0.05 level Variables at level Economic theories are based on the long run equilibrium relation between a set of variables. However, their relation can be disturbed by short term shocks and thus, disequilibrium occurs in short run. But the economists believe that if the variables are integrated, it implies that they have long run relation. So sooner or later, the disequilibrium in the short run returns to the long run equilibrium path. Such relation is stable and produces optimal results. Determinants of Inflation in Nepal: An Empirical Assessment 73 In this paper we examine the long run and short term effects of the related variables on inflation with the help of the W-B single error correction mechanism. So, this model is useful to examine the short run disturbances as well as long run equilibrium, which was first used by Sargon (1984; Cited in Gujarati, 2007) and later popularized by Engle and Granger (Cited in Gujarati, 2007) for the correction of disequilibrium between economic variables in the short run. The Engle and Granger method is two steps EC method. However, we estimate here error correction mechanism in a different way. The WickensBreusch approach (1988) is a single equation error correction model. This model evaluates the state of equilibrium between dependent and independent variables as well as estimates the speed at which a dependent variable returns to long run equilibrium path. So, this embodies a number of desirable properties which are as follows: a) it estimates the short and long term effects in one step; b) it provides easy interpretation of short and long term effects; c) applications to both integrated and stationary time series data; d) can be estimated with OLS; and e) model is built as per the theoretical relation between variables; f) useful for small sample data analysis; g) it assumes only one co-integration relationship among the variables in the equation and only one error correction model is formulated. So, this model is consistent with our economic theories. We can estimate long term effects of Xs on Y, short run effects of Xs on Y and measure the speed of Y variable moving toward the long run equilibrium level with the help of single equation error correction model. As its name suggests, this is a multi-variables single equation model with first order difference and lag variables of both dependent and independent variables. However, we first begin our discussion with the specific error correction model which takes the following form: Y 0 X t 1 (Yt 1 2 X t 1 ) t ---- (i) Where ß1 is error correction component, which measures the rate of return per time unit; or speed of rerun to the equilibrium path; -1< ß1<0, i.e. the value of ß1 lies between -1 and 0; ß0 is short term effects while ß2 estimates long term effects of X on Y. Where Y is a dependent variable and X is explanatory or independent variable, t-l is one year lag length, where ß1 is the error correction component estimated from the equation, which is expected significant and to appear with negative sign. This model is based on the behavioural assumption that two or more time series exhibit an equilibrium relationship in the long run but short run disturbances push the dependent variable away from the equilibrium path. One of the important virtues of this model is that it estimates both short run effects and long run effects of right hand variables in the model. We extend model (i) by including aforementioned variables. Our single equation error correction model which makes the model in its full extent is as follows: ∆Pt = α + ß01 ∆RGDPt + ß02∆BDt + ß03 ∆EXCt + ß04 ∆MP+ ß05∆M2t +Pe- ß1(Pt-1 – ß21RGDPt-1- ß22 BDt-1- + ß23 EXCt-1 + ß240MPIt-1 + ß25 M2t-1+ ß25Pet-1+ εt ---- (ii) Where ∆ denotes first difference of log of variable and t-1 is one year lag of the variable. In the next section we present the regression results from this model. 74 NRB ECONOMIC REVIEW IV. REGRESSION RESULTS AND INTERPRETATION We already discussed about the movement and relationship between the time series of macroeconomic variables. We now are analysing the time series annual data of budget deficits and other macroeconomic variables to see their effects on price level of Nepal. Besides we tried to include the dummy variables for external supply shocks such as 1989 trade embargo imposed by India, 1979/80’s oil price rise and policy change variable in 1991 onward. However, all these dummies appeared with insignificant t-statistics and for the reason we dropped them from the model. Aforementioned we use log form of all variables for the regression purpose. Table 4: Single equation ECM Model Dependent D(CPI) Coefficient Std. Error t-Statistic Prob. Intercept D(rGDP) D(BD) D(EXC) D(dCPII) D(M2) CPIt-1 rGDPt-1 BDt-1 EXCt-1 dCPIIt-1 M2 -5.648542** 0.159783 0.062471* 0.036576 0.055639** 0.227992* -0.760707** 0.433375* 0.072413** 0.169250** 0.109898** 0.142841+ 1.644366 0.164301 0.027578 0.083414 0.011400 0.109064 0.180061 0.160304 0.025298 0.052021 0.018445 0.082091 -3.435089 0.972499 2.265231 0.438494 4.880555 2.090450 -4.224709 2.703459 2.862440 3.253465 5.958097 1.740025 0.0024 0.3414 0.0337 0.6653 0.0001 0.0483 0.0003 0.0130 0.0091 0.0036 0.0000 0.0958 Long run multiplier Adj R2=0.71; F=8.3**; SE=0.02; DW=1.93; AIC=-4.62; SIC=-4.08; N(0)=0.68; SC(1) = 2.33; HC(1)= 19.2; N=34 0.56* 0.10** 0.22** 0.14** 0.19+ **, * & + significant at 1%, 5% and 10% respectively, ns=not significant Table 4 displays the regression results from the single equation ECM model which pass all diagnostic tests shown in the far right column of the table. As priori expectation, the error correction component in this model is highly significant at 1% and appears with negative sign. This shows that the system distracted temporarily out of long run equilibrium path but it will return soon. The speed of return to the equilibrium path is 0.76% next year and the remaining will be corrected following years. All variables considered including real income are influencing inflation in long run while only variables such as budget deficits, change in imported prices and broad money are significant in both short run and long run. The exchange rate, budget deficits and change in imported prices are significant at 1% level whereas real income at 5% level and broad money supply at 10% in the long run. Similarly, in short run change in imported price is significant at 1% while budget deficits and broad money at 5% level. This clear shows that the change in import prices is the most important variable both in short run and long run. Obviously, in the short run, the change in import prices, budget deficits and broad money supply have critical role in determining the rise in prices in Nepal while in the long run almost all variables have influenced inflation. Among these budget deficits, Determinants of Inflation in Nepal: An Empirical Assessment 75 exchange rate and imported prices have played dominant role in determining the level of price in the country. The estimated long run income elasticity of inflation is inelastic or less than one. In other words, long run multiplier is only 0.56, which implies that a 1% rise in real income leads to 0.56% increase in price level followed by exchange rate multiplier (0.22), broad money supply (0.19) and change in imported prices (0.14). However, broad money supply is significant at only 10% in long run as compared to at 5% in short run implies that the effect of this variable on price level slips low in long run. Short run elasticity of the variables is lower than long run elasticity except in case of broad money. The elasticity for budget deficits and change in imported prices is estimated at 0.06 each while that of the broad money supply at 0.23. This implies that prices in Nepal are more sensitive to the change in broad money supply than the changes in imported prices which are the most influential variables in the short run. So, this can have important policy implication. The regression result that broad money supply is significant and real GDP is insignificant in the short run seems to obey the quantity theory of money that says only money supply matters to the inflation as real output and velocity of money remaining the equal. As regression results show that both money supply and real GDP are important variables in the long run, this further satiates the modern monetarists’ approach to inflation theory under less than full equilibrium where real output changes and affects the price level. Furthermore, this result is consistent with the previous study which suggests interest neutrality in Nepal (Paudyal, 2013). Nepal Rastra Bank seems to have monetised deficits to contain the interest rate. It might have taken expansionary monetary policy to offset the effect of deficits on interest rate. This led to increase in money supply causing inflation. On the other hand, budget deficit from the perspective of the Keynesian aggregate demand approach is the resultant of increase in aggregate demand. Such increase in aggregate demand due to excess government expenditure leading to budget deficits shows an excellent influence on inflation (inflationary gap theory) in both periods since this variable is significant at 5% and 1% in short run and long run. Moreover, highly significant effects of imported prices (cpii) variable at 1% both in short run and long run is consistent with the skyrocketing increase in imports which may be also the reflection of domestic supply constraints in the country. This is also comparable with the previous findings of NRB that finds Indian prices are the most important source of Nepali inflation (1994). However, the highly significant imported price variable implies that inflation is largely the function of external price discloses the difficulty of Nepal Rastra Bank which by law has to make attempts to keep the inflation rate low and stable. Still, policy variables such as money supply and budget deficits have important implications. Moreover, the empirical findings reveal that the growth of real GDP has an excellent influence on the inflation in the long run, through growth of real sector and increased supply. In the sum, this study finds that the variations in inflation can be explained largely by imported price, exchange rate, budget deficits, real GDP and broad money in long run. However, in the short run, only the variables like budget deficit, broad money supply and imported prices can be considered as the major determinants of inflation in this country. 76 NRB ECONOMIC REVIEW V. CONCLUSIONS Controlling inflation is not easy task in a country like ours which shares open borders with neighbours and heavily dependent on the imported goods for the daily consumption and materials for other development purposes. In this backdrop, the dominant cause of domestic inflation becomes supply shock generated outside the country. In this context, the control of inflation typically becomes much more complicated and challenging for monetary authorities since the monetary and fiscal policies framed to tame the inflation seem to have lesser implications. However, it does not imply that there is no room for such policy implications at all. This empirical study suggests that prices in Nepal became highly dependent on Indian prices especially after 1991’s political change. It is because of a weaker supply of domestic production supplemented by the increased imported goods from India. The movement of Nepali prices are very close to Indian prices after 1991/92. Because of liberalisation and privatisation policy, the existing limited domestic profit making import substituting enterprises like Basbari shoe factories were closed down partly because they were sold to the private sector. Besides, some of the domestic products could not compete with Indian goods in domestic market in the changing context of reduced import duties under preferential trade agreement with India. This led further rise in the imports from Indian goods and thereby the influence of Indian prices in Nepali prices. So, this has obviously increased the dominance of Indian prices in the domestic prices in the later years. In short run following Keynesians, demand management is a major concern of the government to control the inflation. Well synchronized fiscal and monetary policies targeting the reduction in budget deficits through cutting down recurring expenditures and augmenting revenues of the government together with restrained money supply can be the effective measures to control inflation. Still, there remain some doubts about the effectiveness of the monetary and fiscal policies in the short run since this empirical study suggests the high domination of Indian prices. So far long run is concerned, an increase in domestic supply is the major step to control the inflation. The long run strategy thus to combat inflation is to increase the production of goods and services through the use of productive resources from money and capital markets. The government efforts to create conducive environment for foreign direct investment and ODA in the areas of energy and other infrastructure development along with investment in the productive sector thus can be the effective measures to improve the supply situation/real out and thereby control the inflation. Even so, as suggested by this empirical study, given the open border with India and liberalised trade regime, there exists higher prospect of Indian domination on the domestic prices in this country. Noting that there is a higher prospect of lower prices in India as revealed by its expected higher growth and prosperity, however, suggests that better connectivity with the Indian and Chinese economies will pave the road to deeper integration of Nepali economy with these Asian power houses help Nepal to control inflation. This along with other monetary and fiscal policy measures may be appropriate strategy to combat the inflation in Nepal. ***** Determinants of Inflation in Nepal: An Empirical Assessment 77 REFERENCES Arestis, Philip. No date. “Keynesian Economics, New Keynesian/NCM schools of thought,” Retrieved from http://finance5.net/Keynesian-Economics-and-the-NewKeynesian-NCM-Schools-of-Thought-download-w3462.htm. Bashir, F., S. Nawaz, K. Ashim, U. Khursid, J. Khan and M. J. Qureshi. 2011. “Determinants of Inflation in Pakistan: An Econometric Model using Co-integration approach.” Retrieved from http://www.ajbmr.com/articlepdf/ajbmrv01n0509.pdf Bayo, F. No date. “Determinants of Inflation in Nigeria: An Empirical Analysis”, International Journal of Humanities and Social Sciences, 1(18), Special Issue. Retrieved from http://ijhssnet.com/journals/Vol_1_No_18_Special_Issue/29.pdf Best, R. 2008. “An Introduction to Error Correction Models." Oxford Spring School for Quantitative Methods in Social Research. London. Blanchard, O. J. and G. Jordy. 2006. “A New Keynesian Model with Unemployment.” CFS Working Paper, Retrieved from https:// www.ifk-cfs.de/ fileadmin/downloads/ publications/ wp/07_08.pdf Blanchard, O. J. and F. Stanley. 1989. Lecturers on Macroeconomics, Massachusetts Institute of Technology, USA. Blinder, A. No date. "Keeping Keynesian Faith." Retrieved http://finance5.net/Keeping-the-Keynesian-Faith-download-w3459.html from Bowdler, C. and L. Nunziata. 2004. A Note on the Determinants of Inflation Starts in OECD, University of Oxford. Carre, E. No date. “The New Keynesian Philips Curve: A Meta Analysis.” Retrieved from http://finance5.net/THE-NEW-KEYNESIAN-PHILLIPS-CURVE-A-METAANALYSIS-download-w3491.html Chugh, S. K. 2014. New Keynesian Economics, Retrieved from http://finance5.net/ Chapter-12-New-Keynesian-Economics-download-w3456.html Cornwall, J. 1990. The Theory of Economic Breakdown, Basil Backwell Inc. Cambridge, MA. Cullis, J. and J. Philip. 2009. Public Finance and Public Choice, Oxford University Press, New York. Cunningham_New Keynesian Theory, Retrieved Keynesian-Theory-I-download-w3454.html from http://finance5.net/New- Deist, C. 2011. “ The Role of Contractionary Monetary Policy in the Great Depression.” retrieved from http://econ.berkeley.edu/sites/default/files/deist_charlie.pdf Dennis, R. 2004. “New Keynesian Optimal Policy Model: An empirical Assessment.” Retrieved from http://finance5.net/New-Keynesian-Optimal-Policy-Models-AnEmpirical-Assessment-download-w3472.html. 78 NRB ECONOMIC REVIEW Dixon, H. “Reflection on New Keynesian Economics and Role of Imperfect Competition.” Retrieved fromhttp://finance5.net/Huw-Dixon-Reflections-on-NewKeynesian-Economics;-the-role-of-download- w3468.html. Enders, W. 2014. Applied Econometric Time Series, Wiley India Pvt. Ltd., New Delhi. Friedman, M. 1984. “Inflation is not Beaten.” New York Times. April 3. Retrieved from http://0055d26.netsolhost.com/friedman/pdfs/nyt/NYT.04.03.1984.pdf Friedman, M. 1983.”Deficits and Inflation." Bright Promises, Dismal Performance. Harcourt Brace Jovanovich, New York. Friedman, B. M. 1968. “The Role of Monetary Policy.” The American Economic Review. LVIII, No 1. Retrieved from https://www.aeaweb.org/aer/top20/58.1.1-17.pdf Greenwald, B. and J. E. Stizler. 1987. “Keynesian, New Keynesian and New Classical Economics.” Retrieved from http://www.nber.org/papers/w2160.pdf. Gordon, R. J. 1988, Macroeconomics: Theory and Policy, McGraw-Hill Inc. Gujarati, D. and Sangeeta. 2007. Basic Econometrics, Tata McGraw-Hill Publishing Company Limited, New Delhi. Hiroshi, Y. 2012. A New Micro Foundation of Keynesian Economics, Oxford. Kandi, et al. 2009. “Determinants of Inflation in GCC.” Retrieved from http://www.imf.org/ external/pubs/ft/wp/2009/wp0982.pdf. Keele, L. and Suzanna De Boef. 2004. “Not Just for Cointegration: Error Correction Models with Stationary Data.” Retrieved from http://www.nuffield.ox.ac.uk/Politics/ papers/2005/ Keele%20DeBoef%20ECM%20041213.pdf Khan, A. A., K. H. B. Syed and M. A. Qazi. 2007. “Determinants of Recent Inflation in Pakistan.” Retrieved from http://mpra.ub.unimuenchen.de/16254/1/determinants_of_recent_inflation_in_pakistanRR.pdf Koirala, T. P. 2013. “Time Varying Parameters of Inflation in Nepal: A State Space Modelling.” NRB Economic Review, 25(2) : 66-77. Kumar, R. 2013. “A Study of Inflation Dynamics in India: A Cointegrated Autoregressive Approach.” Retrieved from http://www.iosrjournals.org/iosr-jhss/papers/Vol8issue1/J0816572.pdf. Laryea, S. A. and R. S. Ussif. 2001. “Determinants of Inflation in Tanzania.” Working Paper, Chr. Michelsen Institute, Norway. Mathema, S. R. 1998. “Determinants of Inflation with special reference to wages in Nepal.” NRB Economic Review, 10 : 1-18. Menz, J. O. 2008. “Behavioural Macroeconomics and New Keynesian Mode.” Retrieved from https://docs.google.com/gview?url=http://www.boeckler.de/pdf/ v200810_31 _menz.pdf & chrome = rue Determinants of Inflation in Nepal: An Empirical Assessment 79 Mossayeb, P. and R. Mohammad. 2009. “Sources of Inflation in Iran: An Application of the ARDL Approach.” International Journal of Applied Econometrics and Quantitative Studies, 6(1) : 61-76. MoF. 2014. Economic Survey FY 2013/14, Ministry of Finance. Kathmandu. MoF. Economic Survey, various issues, Ministry of Finance, Kathmandu. Neupane, G. 1992. “Causes of Inflation: A Quantitative Analysis.” NRB Economic Review, 6 : 33-49. Retrieved from http://www.nrb.org.np/ecorev /pdffiles/ vol6_ art3. pdf NRB. 2014. Quarterly Economic Bulletin, Nepal Rastra Bank, Kathmandu NRB. 2010. A Handbook of Government Statistics, Nepal Rastra Bank. Kathmandu. NRB. 2007. Inflation in Nepal, NRB Special Publication, Nepal Rastra Bank. Kathmandu. NRB.1994. “Inflation in Nepal.” Economic Review. Occasional Paper, No. 7, Nepal Rastra Bank, Kathmandu. NRB Research Department. 2001. “Money and Price Relationship in Nepal: A Revisit.” NRB Economic Review, 13 : 50-65. Pahlavani, M. and R. Mohammad. 2009. “Sources of Inflation in Iran: Application of the ARDL Approach.” International Journal of Applied Econometrics and Quantitative Studies, Vol. 6. Parkin, M. and R. Bade. 1986. Modern Macroeconomics, Prentice-Hall Canada Inc. Ontario. Paudyal, S. 2013."Do Budget Deficits Raise Interest Rates in Nepal?” NRB Economic Review, 25(1) : 51-66. Pindyck, R. C., and L. R. Daniel. 1991. Econometric Models and Economic Forecasts, McGraw-Hill Inc, New York. Shapiro, E. 1982. Macroeconomic Analysis, Harcourt Brace Jovanovich, Inc. New York Song, H. A., F. W. Stephen and L. Gang. 2003. "Modelling and forecasting the demand for Thai tourism.” in Journal of Tourism Economics, 9(4) : 363-387. Retrieved from http://epubs.surrey.ac.uk/1125/1/fulltext.pdf. Thapa, R. 2010. “Determinants of Inflation in Nepal.” Nepal Journal of Management, Public Youth Campus, Kathmandu. Vuyyuri, S. and S. V. Sethaiah. 2004. “Budget deficits and other macroeconomic variables in India.” Applied Econometrics and International Journal, 4(1). 80 NRB ECONOMIC REVIEW APPENDICES Table A1: Correlation Among Variables LOG(CPI) LOG(rGDP) LOG(BD) LOG(EXC) LOG(CPII) LOG(M2) LOG(CPI) 1 0.992 0.972 0.975 0.999 0.997 LOG(GDP) 0.998 1 0.962 0.971 0.999 0.999 LOG(BD) 0.972 0.943 1 0.945 0.965 0.959 LOG(EXC) 0.975 0.967 0.945 1 0.971 0.965 LOG(CPII) 0.999 0.995 0.965 0.971 1 0.999 LOG(M2) 0.997 0.996 0.959 0.965 0.999 1 Source: Author’s Calculation Table A2: Budget Deficits and Inflation Year BD/GDP Inflation rate Year BD/GDP Inflation rate 1975/76 2.5 -2.2 1993/94 5.8 8.4 1976/77 3.6 9.8 1994/95 4.8 7.6 1977/78 3.2 7.4 1995/96 5.6 9.2 1978/79 2.7 3.6 1996/97 5.1 3.9 1979/80 3.4 14.7 1997/98 5.9 11.3 1980/81 2.9 11.1 1998/99 5.3 7.5 1981/82 5.4 11.7 1999/00 4.7 2.5 1982/83 9.0 12.4 2000/01 5.5 2.7 1983/84 8.0 2.8 2001/02 5.0 3.0 1984/85 7.6 8.1 2002/03 3.3 5.7 1985/86 7.1 19.0 2003/04 2.9 0.9 1986/87 6.7 10.8 2004/05 3.1 6.8 1987/88 6.1 9.0 2005/06 3.8 7.6 1988/89 9.6 8.8 2006/07 4.1 6.0 1989/90 8.1 8.2 2007/08 4.1 11.0 1990/91 8.9 15.6 2008/09 5.0 11.6 1991/92 7.5 17.1 2009/10 3.5 9.9 1992/93 7.0 7.5 2010/11 3.6 9.6 Source: MoF/GoN, Economic Survey Determinants of Inflation in Nepal: An Empirical Assessment 81 82 NRB ECONOMIC REVIEW Relevance of Keynesianism in Nepal: An Empirical Analysis Hom Nath Gaire Abstract In this paper, the relevance of Keynesian postulates has been examined in the Nepalese context for the period 1975-2012 using annual time series data. The empirical results from the Johansen cointegration tests clearly show that there is long run equilibrium relationship between government expenditure and real GDP, private consumption and gross fixed capital formation. Likewise, Granger Causality test confirms that there is bilateral causal relationship between government expenditure and gross fixed capital formation in Nepal. However, no causal relationship is observed between government expenditure and real GDP and private consumption. Thus, it is confirmed by this study that the Keynesian postulates are relevant for capital formation rather than for increasing real GDP growth and private consumption in Nepal. Key Words: Keynesianism, Effective Demand, Casual Relationship, Government Expenditure, Gross Fixed Capital Formation and Real GDP JEL Classification: E12 Lecturer of Economics, Greenfield National College, Kathmandu. Email: [email protected] Remarks: The author is grateful to the Editorial Board and anonymous referees for their valuable comments that helped greatly in the improvement of this paper. 84 NRB ECONOMIC REVIEW I. INTRODUCTION Keynesianism is a macroeconomic school of thought based on the ideas of 20th century British Economist John Maynard Keynes. The concepts forming the basis of Keynesianism were first published in “The General Theory of Employment, Interest and Money” in 1936. This book is a repudiation of the foundations of laissez-faire and advocacy of active government because unemployment is primarily a matter of the volume of effective demand. Keynes argues that some individually-rational microeconomic actions, if taken collectively by a large proportion of individuals and firms, can lead to ineffective aggregate macroeconomic outcomes, where in the economy operates below its potential output level (Keynes, 1936). It is further argued that such low level economic situation can be corrected by the Government through active monetary and fiscal policies. One of the tenets of Keynesian theory is that government spending on consumption and investment, tax cuts and lower interest rates can stimulate demand and induce investment which would have otherwise remained idle to produce wealth (Keynes, 1936). Similarly, redistribution of wealth from wealthy to poor, who are perceived to have higher marginal propensity to spend would generate higher economic growth. Therefore, for four decades from mid-1945 to mid-1970 Keynesianism dominated the thinking of professional economists and public policy makers not only in the United States, and Europe but also in a number of developing countries. However, the Keynesian principles have also been subjected to considerable criticisms during the same period. The critics argue that macroeconomic policies based on Keynesianism are counter-productive to stabilize the economy and these will lead to inflation, income inequality, and incite consumers to spend even more in anticipation of future tax increase (Michael, 2006). At the same time, Keynesians advocate an active stabilization policy for reducing the magnitude of the business cycle, which they rank as the most serious economic problem by raising aggregate demand thereby stimulating economic activities, reducing unemployment and avoiding deflation. Governments in Nepal have used expansionary fiscal policy since long back to stimulate demand as a countercyclical measure as well as for political reasons. It is believed that large budgets can play influential role in generating higher growth and increasing employment. However, the reality does not confirm this as government expenditure and growth do not seem to move together. Hence, testing causality between government expenditure and economic growth or examining the relevance of Keynesianism would be a worthwhile exercise. The main objective of this paper is to gauge the relevance and implication of Keynesian notions in the Nepalese context. For this, the study aims to test the causality between the government expenditure and real GDP, private consumption and gross fixed capital formation for the period between 1975 and 2012. Conclusions drawn from the study would provide useful insights to fiscal policy makers of Nepal. Relevance of Keynesianism in Nepal: An Empirical Analysis 85 The rest of the paper is organized as follows. The second section provides a precise review of evolution of Keynesianism which covers origin of Keynesian thoughts, dominance of Keynesian policy and monetarist revolution followed by the counter revolution of Keynesianism. Section three covers the review of empirical studies on relevance as well as effectiveness of Keynesian thoughts available so far both in global and Nepalese context. Section four briefly describes the data and methodology used in this study. Section five presents the results and discussion of empirical analysis. The last section concludes the discussion. II. EVOLUTION OF KEYNESIANISM Origin John Maynard Keynes (1883–1946) had acquired an international reputation shortly after World War-I by “The Economic Consequences of Peace”. In his 1924 book, “A Tract on Monetary Reform”, Keynes declared that gold was a “barbarous relic” and that governments should control money supply to maintain a stable domestic price level as well as a stable foreign exchange rate (Anderson, 1925). In 1930 Keynes published “A Treatise on Money”, a two-volume work which established him as the reputed leading monetary theorist for the next five years. Keynes' “The General Theory of Employment, Interest, and Money” published in February 1936 is widely regarded as the cornerstone of Keynesian thought. By the end of World War-II, The General Theory became the foundation of the new “Macroeconomics”, which in turn was popularized as Keynesianism (Hutt, 1963). Keynesian Dominance: 1941–1979 From the end of the Great Depression, Keynesian ideas quickly established in America and Europe also was a leading inspiration for the English speaking common wealth countries of Asia and Africa from 1941 to the mid-1960s. In late 1965, Time Magazine in a cover story entitled "We are all Keynesians now" scaled Keynes's central theme by stating that Keynes was one of the three most important economists ever, and that his General Theory was more influential than the ‘magna opera’ of his rivals i.e. Adam Smith’s ‘The Wealth of Nations’ and Karl Marx's ‘Das Capital’.11 Hence, from early 1940s to the mid-1970s, which is also known as the Golden Age of capitalism, Keynesianism provides the main inspiration for economic policy makers and for prominent economists including the academia. Monetarist Revolution: 1979-1999 The stagflation of 1970s including the oil crisis of 1973 followed by the recession questioned the logic behind Keynesianism and lead to the development of new classical macroeconomics. Thus, Austrian School of thoughts and Monetarism charged 11 "We are all Keynesians now". Time Magazine, 1965-12-3, Retrieved 2008-11-13. 86 NRB ECONOMIC REVIEW Keynesianism and demand management as tools for 'fools' because wealth, in a better society and cleaner world along with a higher level of development, cannot be directed by the government. Meanwhile, the “Washington Consensus” which propagates that markets work best if they are unregulated came to be used as a notable anti-Keynesian view. That created the space to proliferate the Monetarism and new classical economics, which in turn displaced the Keynesianism for 1979-1999 (Hoover, 2003). Keynesian Counter Revolution: 1999–2007 The Asian Financial Crisis of 1997 in the developing world and market failure as well as Dotcom crash of the 2000 in advanced economies caused a turn back from free market policies to Keynesianism. In the meantime, Britain and Japan had shown keenness to Keynesianism saying "the real challenge was to interpret Keynes's insights for the modern world" (Carabelli, 2010). By 2007 there had been high promotion of Keynesianism in the English speaking countries including China, India and south East Asia. In the academic world, the advent of the global financial crisis in 2007 had caused the resurgence of Keynesian thought (Anthers, 2010). Keynesianism After 2008 During the global financial crisis (2007–2009), the Keynesianism was receiving most attention as fiscal stimulus was widely launched across the world. It was mid-2010 that the earlier global consensus for ongoing Keynesian stimulus had broken, especially in Europe, as there was an increasing demand for immediate fiscal tightening. By mid-2012, with the on-going Euro crisis and persistent unemployment problem in the US, there has been renewed consideration of stimulus policies by European and American policy makers, although there is no return to the pro stimulus consensus that existed in 2009 (Farrell and Quiggin, 2012). III. LITERATURE REVIEW Numerous studies have been conducted to investigate the relationship between government spending and economic growth with mixed results. Landau (1983) found that the share of government consumption to GDP reduced economic growth which was consistent with the pro-market view that the growth in government constrains overall economic growth. Ram's (1986) study made a rigorous attempt to incorporate a theoretical basis for tracing the impact of government expenditure to growth through the use of production functions specified for both public and private sectors. The author found government capital expenditure to have significant positive externalities on growth particularly in the developing countries. Lin (1994) used a sample of 62 countries (196085) and found that non productive spending had no effect in growth in the advanced countries but a positive impact in LDCs. Josaphat et al. (2000) investigate the impact of government spending on economic growth in Tanzania (1965-1996) using time series data for 32 years. The results revealed that expenditure on human capital investment was Relevance of Keynesianism in Nepal: An Empirical Analysis 87 insignificant in their regression and confirm the view that public investment in Tanzania was not productive. Junko and Vitali (2008) in an investigation of the impact of government expenditure on economic growth in Azerbaijan suggested that the initial growth performance largely depends on the efficiency of scale-up expenditure. Komain and Brahmasrene (2007) gauged the relationships between government expenditure and economic growth for Thailand during the period 1970-2005. The results suggest that there was a long-run relationship between government expenditure and economic growth, thus supporting the Keynesian hypothesis. Jamshaid et al. (2010) found a wide range of evidences on the impacts of government expenditure on economic development and concludes that government expenditure contributes to economic growth, both through supply and demand channels in the USA, Japan, Germany, France, United Kingdom, Italy and Canada. The study suggested government expenditure contributes in raising the quality of life by creating amenities, providing consumption goods and contributing to macroeconomic stability. Amid inconclusive evidences, Keynesian policies have been able to exert some positive impact in the global economy, especially during crises since 1930s great depression to the latest financial crisis of 2007-2009. Skidelsky (2011) made a comparison between the performance of the world economy during the Golden Age period (1951–1973) where Keynesian policies were dominant and the Washington Consensus period (1981–2008) where free market policies were adopted. The study reveals that the 'golden age' period was substantially more stable with higher growth, employment and low inequality. However, during the 'Washington Consensus' period the world economy was quite unstable with increasing inequality. In Nepalese context, Shrestha (2009) investigated the role of composition of public expenditure, particularly the expenditure on physical infrastructure, on economic growth in Nepal based on the endogenous growth model using time series data. The results suggest that the impact of public expenditure on economic growth was positive. However, Chaudhary (2010) found no causality between real GDP and government expenditure in Nepal. The findings suggest that the increase in the size of government expenditure has no influence on economic growth of Nepal. Recently, Sharma (2012) tested the impact of government expenditure on economic growth of Nepal. The results reveal that although there is a weak influence on economic growth, growth depends on the size, spending capacity, and effective use of capital expenditure in the development process. Similarly, Kharel (2012) develops a macroeconomic forecasting model focusing on fiscal policy and economic growth in Nepal using annual data from 1992/93 to 2009/10. The evidence suggests that fiscal policy, particularly government' capital expenditure affects economic growth positively and also crowds-in private investment. However, there exists a trade-off between fiscal stability and high level of economic growth as the policy goal of achieving both objectives seems to be unattainable. Within 88 NRB ECONOMIC REVIEW the above theoretical and empirical evidences this study analyzes causality between government expenditure and economic growth in Nepal. IV. DATA AND METHODOLOGY Many empirical studies of macro impact on government spending were based on the Vector Autoregressive (VAR) model of major macroeconomic variables. A number of the studies were focused to estimate the effect of government spending and fiscal deficit on growth variables. Blanchard and Perotti (1999) used data pertinent to the United States during the postwar period for VAR specification of taxes, government spending and GDP in real per capita terms. Similarly, Heppke-Falk, Tenhofen and Wolff (2006) used Structural Vector Autoregressive (SVAR) approach to investigate short-run effects of fiscal policy shocks on the German economy. As this study is primarily based on the time series secondary data of Government Expenditure (GE) and Economic Growth, Johansen Co-integration method based on VAR approach has been used. In order to test the causality between natural log values of GE vis-à-vis real GDP, Private Consumption (PC) and Gross Fixed Capital Formation (GFCF), time series annual data for the period 1975 to 2012 have been used. Model Specification In order to find out the causality between GE and Economic Growth Variables, natural log value of GE is taken as independent variables while natural log values of real GDP, PC and GFCF are taken as dependent variables. For this purpose the following models have been developed. GDPt = 0 + 1 GEt + t1 …… (1) PCt = 0 + 1 GEt + t2 …… (2) GFCFt = 0 + 1 GEt + t3 …… (3) Where, i, i,i are parameters to be estimated and ti are white noise error terms Unit Root Tests Many economic and financial time series data exhibit trending behavior or non-stationary in the mean. A series is said to be stationary if the mean and auto covariance of the series do not depend on the time. A series whose mean and auto covariance depend on time is said to be non-stationary. An important econometric task is determining the most appropriate form of trend in the data. If the data are trending then some trend removal measures are required to transform the data into stationary form prior to analysis. Two common trend removal or de-trending procedures are first differencing and time trend Relevance of Keynesianism in Nepal: An Empirical Analysis 89 regression. First differencing is appropriate for time series and time trend regression is appropriate for trend stationary time series. As the present study is based on the time series data, it is important to check whether a series is stationary or not before analysis. For this purpose, first differencing procedure i.e. Augmented Dickey-Fuller (ADF) test has been performed in this study. Since the ADF test of unit root does not follow the conventional Student's t-distribution, Mackinnon (1991, 1996) t-values have been used. Co-integration Test Economically speaking, two variables will be co-integration if they have a long term or equilibrium relationship. Although there are a number of methods for testing the cointegration, the following Vector Auto Regression (VAR) method of order p developed by Johansen has been utilized. yt = t + A1yt-1 + … + Apyt-p + Bxt + t …… (4) Where, yt is an n×1vector of variables that are integrated of order one - commonly denoted I (1) - t is an n×1 vector of innovations. In this test, the null hypothesis of r co-integrating vectors is tested against the alternative of r +1 co-integrating vectors. Thus, the null hypothesis r=0 is tested against the alternative r=1 against r=2, and so forth. Johansen proposes two different likelihood ratio tests of the significance of these canonical correlations and thereby the reduced rank of the Π matrix: the trace test and maximum Eigen value test as follows: 𝑗𝑡𝑟𝑎𝑐𝑒 (r/p) = -T∑𝑛𝑖=𝑟+1) 𝐼𝑛(1 − 𝑖 ) …… (5) 𝑗𝑚𝑎𝑥 (r/r + 1) = -T 𝐼𝑛(1 − 𝑖+1 ) …… (6) Here T is the sample size and is the ith largest canonical correlation. As the co-integration tests are very sensitive to the choice of lag length, following Akaike Information Criteria (AIC) and Schwarz Information Criteria (SIC) after existence of cointegration between the variables in the equations, the Granger Causality test has been performed. Granger Causality Test The common practice in testing the direction of causation between two variables is the Granger Causality test. According to Granger (1969), series X causes Y if the past values of X can more accurately predict Y than simply the past values of Y. In simple words, if past value X improves the prediction of Y with statistical significance, then we can 90 NRB ECONOMIC REVIEW conclude that X “Granger Causes” Y. The Granger Causality test for the above equations (1), (2) and (3) has been performed on the basis of the following fundamental model. 𝑌𝑡 = 0 + 1 𝑌𝑡−1 + 2 𝑌𝑡−2 + ⋯ 2 𝑌𝑡−𝑛 + 1 𝑋𝑡−1 + ⋯ + 𝑚 𝑋𝑡−𝑚 + 𝑈𝑡 …… (7) Where, 𝑈𝑡 white noise error is term series. V. RESULTS AND DISCUSSIONS Findings In order to gauge the relevance of Keynesianism in Nepal, in this study, first of all ADF tests have been performed to examine the unit root in all the set of 4 series comprising log values of GE, GDP, PC and GFCF for the period of 1975-2012. The results of ADF tests presented in the table-1 support that the log value series under consideration are not stationary at both level and first difference. This is confirmed as the calculated values of t-statistics, in absolute sense, are smaller than the tabulated values at both 1% and 5% level of significance accepting the null hypotheses that the series are non-stationary. This indicates that there is trending behavior in mean of all the series under consideration. Table 1: Unit Root Tests Variables NLGDP NLGE NLPC NLGFCF At Level 0.535 -1.411 -2.192 -2.532 t-statistics First Difference 0.300 -0.787 -2.179 -2.779 MacKinnon p-value At Level First Difference 0.9859 0.9774 0.5770 0.8229 0.2092 0.2141 0.1079 0.0614 Critical values for level at 1% and 5% are respectively -3.668 and -2.966 Similarly critical values for first difference at 1% and 5% respectively are -3.675 and 2.969 Figure-1: Log Value of Variables under Consideration Relevance of Keynesianism in Nepal: An Empirical Analysis 91 Johansen Co-integration Tests After confirming the non-stationary nature of series under consideration, it is required to test whether the variables are co-integrated or not i.e. whether they exhibit the tendency of co-movement over the long run and converge towards equilibrium. Table 2 depicts the results of the Johansen Co-integration tests. Both the trace test and maximum Eigen value test reject the null hypotheses of all models that there is no cointegration between the variables under consideration at 99 percent confidence level. Table 2: Results of Johansen Co-integration Tests GDP and GE (Sample-1976 – 2012), Trend- Linear, Lags-1 Null Hypothesis (H0) r=0 r≤1 Eigen Value Trace Statistics Critical Value 5%/1% Max-Eigen Statistics 0.59018 0.37549 47.6995 16.4778 15.41/20.04 3.76/6.65 31.2217 16.4778 Critical Value 5%/1% 14.07/18.63 3.76/6.65 PC and GE (Sample-1976 – 2012), Trend- Linear, Lags-1 Null Hypothesis (H0) r=0 r≤1 Eigen Value Trace Statistics Critical Value 5%/1% Max-Eigen Statistics 0.56200 0.31429 42.0988 13.2053 15.41/20.04 3.76/6.65 28.8935 13.2053 Critical Value 5%/1% 14.07/18.63 3.76/6.65 GFCF and GE (Sample-1976 – 2012), Trend- Linear, Lags-1 Null Hypothesis (H0) r=0 r≤1 Eigen Value Trace Statistics Critical Value 5%/1% Max-Eigen Statistics 0.61558 0.37636 49.9874 16.5266 15.41/20.04 3.76/6.65 33.4608 16.5266 Critical Value 5%/1% 14.07/18.63 3.76/6.65 The above result of Johansen co-integration tests confirms that there is co-integration of the Government Expenditure vis-à-vis real GDP, Private Consumption and Gross Fixed Capital Formation of Nepal. The existence of co-integration implies that there is long-run relationship between the Government Expenditure variables and Economic Growth Variables in Nepal partially supporting the Keynesian notion. Granger Causality Tests The results of Granger Causality Test are reported in the following Table 3. The Wald F-statistics and the corresponding critical values indicate there is no any causality between the Government Expenditure vis-à-vis real GDP and Private Consumption, since the null hypotheses of equations (1) and (2) that GE does not Granger Cause real GDP and PC accepted with high probability values. However, there is a bilateral causality 92 NRB ECONOMIC REVIEW between the Government Expenditure and Gross Fixed Capital Formation. This is confirmed since the null hypothesis of equation (3) that GE does not Granger Causes GFCF is rejected at 5 % level of significance to very low probability values. Table 3: Pairs wise Granger causality (Wald) tests (Sample-1976 – 2012), Lags-1 Null Hypothesis (H0) F-Statistics Probability Decision GE does not granger cause GDP .40944 0.6675 (H0) Accepted GE does not granger cause PC .39971 0.6738 (H0) Accepted GE does not granger cause GFCF 4.9003 (3.32)* 0.0139 (H0) Rejected GFCF does not granger cause GE 5.3341 (3.32)* 0.0100 (H0) Rejected * indicates the rejection of H0 at 5% level of significance respectively. Figures in parenthesis are the tabulated values of F-distribution for corresponding degree of freedoms. VI. CONCLUSION This paper examined co-integration and causality between the Government Expenditure (GE) vis-à-vis real Gross Domestic Product (GDP), Private Consumption (PC) and Gross Fixed Capital Formation (GFCF) with an aim of testing the relevancy of Keynesianism in the context of Nepal using time series data of 1975 to 2012. Using the methods of the unit root tests and co-integration tests, the study confirmed that there is long-run equilibrium relationship between the Government Expenditure variables and Economic Growth variables in Nepal. However, Granger Causality test revealed that there is no causality between the Government Expenditure and real GDP as well as private consumption for the review period. However, there is bilateral causality between Government Expenditure and Gross Fixed Capital Formation (GFCF) in Nepal. The evidence from this study reveals that Keynesian notion, which claims positive impact of Government Expenditure on real GDP and private consumption, is not valid for Nepal. This may be because that the GDP of Nepal is mainly dependent on agriculture production which is subject to the favorable weather conditions and the private consumption is highly depends on remittance received from foreign employment. Similarly, because of high propensity to consume and supply side constraints in the economy a given increment in government expenditure is leaked out of the country in the form of imports. But the Keynesian notion that the Government can play pivotal role in capital formation through its expenditure, which in turn stimulate the private investment and growth of the economy is proved. Thus the Government can contribute in creating favorable environment for private sector and business community through infrastructure development and capital formation by raising capital expenditure. The results of this study, in line of some literatures, confirm that the notion of Keynesianism to promote economic activities and growth through government intervention is partially relevant in Nepal. This means the Keynesian notion which is based on industrialized economies could not fully perform in the agriculture dominated least developed economies like Nepal. However, there is a role of Government in such economies where there are market imperfections and the private sector is not capable enough for huge investment in infrastructure development and capital formation. Relevance of Keynesianism in Nepal: An Empirical Analysis 93 The findings of this study suggest that the Government should not be involved in general kind of business activities such as production and distribution of goods and services rather should focus on effective governance and mobilization of resources in order to increase the capital expenditure for capital formation and infrastructure development. ***** REFERENCES Anderson, B. 1925. “The Gold Standard vs. a Managed Currency.” Chase Economic Bulletin, 1925, pp. 39 Anthers, J. 2010. “The Fearful Rise of Markets: Short View of Global Bubbles and Synchronized Meltdowns.” Prentice Hall, ISBN 978-0-273-73168-9 Blanchard, O. and R. Perotti. 1999. “An Empirical Characterization of the Dynamic Effects of Changes in Government Spending and Taxes on Output.” NBER Working Paper, No. 2685 Carabelli, A. M. 2010. “Current Global Imbalances: Might Keynes be of help.” Chp 14, p. 257 –274, University of Toronto Press Chaudhary, S. K. 2010. “Public Expenditure and Economic Development in Nepal.” Economic Literature, IX : 96-104 Engle, R. and C. Granger, 1987. “Co-integration and Error Correction: Representation, Estimation and Testing.” Econometrica, 35 : 251-276. Farrell, H. and J. Quiggin. 2012. “Consensus, Dissensus and Economic Ideas: The Rise and fall of Keynesianism during the Economic Crisis.” The Center for the Study of Development Strategies, Gujrati, D. N., D.C. Porter and S. Gunasekar. 2009. Basic Econometrics, McGraw Hill Education Pvt. Ltd, New Delhi, India Fifth Edition Heppke-Falk, K. H., J. Tenhofen and G. B. Wolff. 2006. “The Macroeconomic Effects of Exogenous Fiscal Policy Shocks in Germany: A Disaggregate SVAR Analysis.” Discussion Paper Series 1, Economic Studies, No. 41, Deutsche Bundes bank Hoover, K. R. 2003. “Economics as ideology: Keynes, Laski, Hayek, and the creation of contemporary politics.” Rowman & Littlefield, Manchester University Press. pp. 16 Hutt, W. H. 1963. “Keynesianism: Retrospect and Prospect: A Critical Restatement of Basic Economic Principles.” Chicago: Henry Regnery 94 NRB ECONOMIC REVIEW Jiranyakul, K. and T. Brahmasrene. 2007. "The relationship between government expenditures and economic growth in Thailand." Journal of Economics and Economic Education Research, 8(1) : 93-102. Josaphat, P. K. and M. Oliver. 2000. “Government Spending and Economic Growth in Tanzania.” 1965-996: CREDIT Research Paper. Junko K. and K. Vitali. 2008. “Impact of Government Expenditure on Growth.” IMF Working Paper Retrieved on 2012-05-29. Jamshaid, R., A. Iqbal and M. Siddiqi, 2010. “Cointegration-Causality Analysis between Public Expenditures and Economic Growth in Pakistan.” European Journal Social Sciences, 13(4) : 556. Keynes, J. M. 1930. “A Treatise on Money.” 2 Volumes, New York: Harcourt Brace. Keynes, J. M. 1936. “The General Theory of Employment, Interest and Money.” London: Macmillan (Reprinted 2007). Kharel, R. S. 2012. “Modeling and Forecasting Fiscal Policy and Economic Growth in Nepal.” Nepal Rastra Bank working paper series, NRB-WP-10-2012. Landau, D. 1983. “Government Expenditure and Economic Growth: a Cross- Country Study.” Southern Economic Journal, 49(3) : 783-792. Michael, L. 2006. “The economics of Keynes in historical context.” London: Palgrave Macmillan. Ram, R. 1986. “Government Size and Economic Growth: A new Framework and some Empirical Evidence from Cross-sectional and Time Series Data.” American Economic Review, 76 : 191-203. Sharma, B. 2012. “Government expenditure and economic growth in Nepal a minute analysis,” Basic Research Journal of Business Management and Accounts ISSN 2315-6899, 1(4) : 37-40 Shrestha, P. K. 2009. “The Composition of Public Expenditure, Physical Infrastructure and Economic Growth in Nepal.” NRB Economic Review, 21 : 79-98. Skidelsky, R. 2011, “John Maynard Keynes: 1883–1946: Economist, Philosopher, Statesman.” Macmillan, ISBN 0-330-48867-8. Steven, A. Y. Lin. 1994. “Government Spending and Economic Growth.” Applied Economics, 26(1) : 83-94. Relevance of Keynesianism in Nepal: An Empirical Analysis 95 Appendix 1 Data Used in the Study (Figure in Rs 10 Million) Year 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Government Expenditure 151.4 191.3 233.0 267.5 302.5 347.1 409.2 536.1 697.9 743.7 839.5 979.7 1151.3 1410.5 1800.5 1966.9 2355.0 2641.8 3089.8 3359.7 3906.0 4654.2 5072.4 5611.8 5957.9 6627.3 7983.5 8007.2 8400.6 8944.3 10256.0 11088.9 13360.5 16135.0 21966.2 25968.9 29536.3 34514.6 Real GDP 13106.2 13609.4 13838.9 14288.6 14524.0 14573.4 15874.7 16644.1 16820.4 18299.2 19552.9 20483.8 20915.2 22390.3 23597.9 24749.1 26395.5 27687.5 28644.9 30911.5 31840.7 33668.1 35358.6 36559.2 38234.8 40574.6 44151.8 44204.9 45948.8 48100.8 49773.9 51448.6 53203.8 56451.7 58941.9 61625.7 64255.3 67232.6 Source: Economic Survey Various Issues (CBS) Private Consumption 1365.2 1406.0 1368.9 1572.9 1774.1 1919.5 2241.1 2527.2 2745.8 3186.0 3597.7 4478.2 5074.6 6240.7 7017.3 8631.4 9777.1 12137.2 13340.2 15406.5 16644.3 19146.9 21636.4 23139.2 26494.4 28794.7 34898.9 36094.7 37142.1 37405.7 39221.9 41321.7 42541.9 43076.3 45546.8 48298.4 48524.9 51025.7 Gross Fixed Capital Formation 222.3 244.3 258.0 329.4 326.3 368.1 429.9 546.5 657.6 690.7 938.6 943.1 1182.5 1341.4 1639.2 1700.2 2278.0 2927.7 3727.8 4203.2 4837.0 5608.1 6079.4 6537.5 6526.9 7332.4 8475.1 8486.3 8806.9 9094.9 9142.7 10157.0 10694.0 10892.2 10945.9 12764.7 12672.3 12197.9 GUIDELINES FOR ARTICLE SUBMISSION NRB Economic Review, previously published as the "Economic Review Occasional Paper", is a bi-annual peer-reviewed economic journal being published in April and October. Submission of a paper for the NRB Economic Review will be taken to imply that it represents original work not previously published, and it is not being considered elsewhere for publication, and that if accepted for publication it will not be published anywhere without the consent of the Editorial Board. The papers so received have to undergo a double blind review process and are then subject to approval by the Editorial Board. However, the ideas and opinions expressed in the papers published in the Review are solely those of authors and in no way represent views and policies of Nepal Rastra Bank or that of the Editorial Board. Submitted manuscripts should be written in English, typed in double spacing with wide margins (3 cm) on one side of standard paper. The title page should contain the title, the name, institutional affiliation(s), JEL classification, key words, full postal address, telephone/fax number and E-mail of each author, and, in the case of co-authorship indicate the corresponding author. In case the author(s) is provided grant or any type of financial support from any organization or institution, this should be spelled out clearly below the key words. Footnotes, if any, should be numbered consecutively with superscript arithmetic numerals at the foot of each page. Figures and tables should be on separate sheets and have descriptive titles. References in the text should follow the author-date format. References should be listed alphabetically in the following style: Anderson, T. W., and C. Hsiao. 1982. "Formulation and Estimation of Dynamic Models Using Panel Data." Journal of Econometrics 18: 47–82. Goldstrein, M. and M. Khan. 1985. "Income and Price Effects in Foreign Trade." In R. W. Joners and P. B. Kenen, eds., Handbook of International Economics, vol. II, Elsevier, New York. Hemphill, W. 1974. "The Effect of Foreign Exchange Receipts on Imports of Less Developed Countries." IMF Staff Papers 21: 637–77. The manuscript should be accompanied by an abstract not exceeding 300 words, and the preferred maximum length of a submission is 10,000 words. The preferred word processing software for the Review is Microsoft Word. Authors should e-mail their manuscript to: The Editorial Board NRB Economic Review Nepal Rastra Bank Research Department Baluwatar, Kathmandu Email: [email protected] Telephone: 977-1-4419804, Ext. 357 Past Issues of NRB Economic Review are available at www.nrb.org.np under Publication