Stock Market Development and Economic Growth: Empirical Evidence from Nepal

Abstract

This study examines the relationship between stock market development and economic growth in Nepal by employing autoregressive distributed lag (ARDL) model with bound testing procedures. The study period covers annual time series data from 1994 to 2019. Indicators of the stock market development used are size, depth and efficiency represented by market capitalization as a percentage of gross domestic product (GDP), total value of shares traded as a percentage of GDP and total shares traded as a percentage of market capitalization, respectively. Following high correlations among these indicators, an aggregated index is constructed and used in the study. Real GDP per capita growth is taken as an economic growth indicator. The results suggest that there exists a long-run uni-directional causality relationship running from stock market development index to economic growth. Stock market size and liquidity are significant contributors, showing that stock market is able to mobilize capital and diversify risks with increased easiness in trading of stocks. The control variable market inflation shows no significant impact on either of the examined primary variables.

Keywords

ARDL model economic growth Nepal stock market development

Introduction

A well-functioning capital market is an important component of a financial system that entails economic growth (Rousseau & Sylla, 2003). A stock market’s contribution towards economic growth can be attributed to the mobilization of idle savings in the economy, thereby converting these savings into productive capital. Apart from allowing businesses to collect capital by selling securities, a stock market also provides an atmosphere for trading ownership to such businesses by investors for liquidity. Levine (1991) stood for stock market contributing economic growth by arguing that liquidity in the stock market makes investments less volatile as it allows investors to purchase or sell shares over a long investment horizon without locking in their savings, while, at the same time, businesses also get long-term capital through equity. However, according to Demirgüç-Kunt and Levine (1996), liquidity in stock market can cause an inverse impact in the economic growth in the long run through three possible channels. First, liquidity in stock market can motivate higher investments and thus higher returns, which will cause preponement of future consumption, leading to a decrease in the amount of money invested in the economy in future. Second, market liquidity will reduce the level of uncertainty, attracting high investments and discouraging precautionary savings, causing an impact on saving rate and investment in the economy. Third, a highly liquid market allows investors to make quick buy and sale of stocks, thereby allowing the stocks of ill-managed firms get through without much accountability, which, in turn, hurts the long-run economic growth.

The empirical literature shows no consensus, regarding the existence and nature of the relationship between economic growth and stock market, accompanied by a debate on the causal relationship between them (Pan & Mishra, 2018). The literature also indicates that the nature of the relationship varies among the countries that are at different levels of economic growth. According to Demirgüç-Kunt and Levine (1999), as a country reaches higher levels of income, stock markets tend to play an increasing role, indicating that in more developed countries, capital markets are larger and more liquid, whereas less developed countries are bank based.

The Nepalese economy has undergone significant structural transformations after the restoration of the multiparty system and the new constitution in 1990. The World Bank structural adjustments in the financial sector and the operation of stock exchange were significant reforms in the early 1990s. The Eighth Five Year Plan (1992–1997) was carved out to consider the private sector as instrumental in achieving sustainable development. The plan paved the way for a more organized securities market in Nepal by formulating the first comprehensive plan. As a result, the development of the securities market was strengthened by the establishment of the Securities Board of Nepal (SEBON) and Nepal Stock Exchange Ltd. (NEPSE) in 1993, where after an organized and full-fledged stock market began, and NEPSE opened its trading floor on 13 January 1994. Table 1 presents the market capitalization that has grown over the period of 25 years and so do the number of companies, indicating increasing trend of companies using the stock market for capital realization. The impact of the global financial crisis (2008) and the political uneven transition during 2010 in Nepal forced the market index to plummet sharply in 2011, which steadily rose after that. It is also observed that economic growth and inflation are not parallel to the movement of stock market development. In this regard, it is imperative to examine the relationships of these macroeconomic factors with the stock market to understand the behaviour of low-income emerging economy.

Table 1.

Nepalese stock market and economic growth

F.Y. (mid July)	No. of Listed Companies	Market capitalization (in million Rs)	Value of shares traded (in million Rs.)	NEPSE Index	Inflation (%)	Real GDP per capita (%)
1994-95	79	12963	1054	195	7.6	0.9
1998-99	107	23508	1500	217	7.5	2.4
2002-03	108	35240	576	205	5.7	2.5
2006-07	135	186301	8360	684	2.3	2.2
2010-11	209	323484	6665	363	9.2	3.3
2014-15	232	989404	65332	961	7.9	2.9
2018-19	215	1567499	110075	1259	5.6	5.0

Source: Annual publications of Nepal Rastra Bank, NEPSE and Securities Board of Nepal

Amidst a debate about a possible linkage between economic growth and financial markets, especially in developing and developed economic domains, the evolution of stock markets in low-income emerging markets and their potential impact on overall economic development represent a new scope of inquiry for researchers. The available literature is mostly on developed and developing economies, and it is still not clear if low-income emerging markets like that of Nepal have the same positive impact of stock market on economic growth. The current study, thus, attempts to fill this gap in the literature by examining the influence of stock market development of Nepal on her economic growth, so as to recommend the policymakers about how the Nepalese capital market can be improved that would in a way portray an economic performance of the country.

The remainder of this study is organized as follows: the second section briefly presents the literature review, followed by study objectives in the third section. The fourth and fifth sections present the rationale of the study and methodology discussions, respectively. The sixth section incorporates empirical results with major findings, and conclusion with policy implications are presented in the final section.

Review of Literature

Theoretical Underpinning

The classical work of Schumpeter (1911), which purports that financial sector development is essential for technological innovation and economic growth, gave rise to different studies on the linkage between financial market and economic growth. Theoretical literature based on the endogenous growth theory suggests that a well-developed stock market may affect growth in various ways. First, the stock market facilitates domestic savings mobilization and diversifies the portfolio through financial instruments. Second, the stock market allows the opportunity for share ownership and thus sharing of risks. Third, it allows efficient allocation of capital to productive investments, making domestic and foreign investments more attractive (Baier et al., 2004; Levine, 1991).

The relationships between economic growth and stock market development (as an indicator of financial development) are explained along different lines in the theoretical literature. Schumpeter (1911) followed by Goldsmith (1969) and Shaw (1973) argued that financial development leads to economic growth, thus developing a supply-leading hypothesis. Patrick (1966), however, argued that finance is only a passive factor in the economic growth process, leading to another edge of relationship, which is demand-following hypothesis that signified the role of economic growth creating a demand for financial services. Patrick (1966) also claimed that stock market development and economic growth can strengthen each other, involving mutual causality showing a feedback effect. Lucas (1988) proposed a different world view of no relationship between economic growth and stock market by arguing that economists badly exaggerate the role of the financial system in economic development.

Empirical Studies

Plethora of studies have documented the role of capital market in economic growth. Positive effect of stock market development on economic growth is observed in cross-country data (Durusu-Ciftci et al., 2017; Levine & Zervos, 1998). The analysis of time series data of individual countries reveals stock market development, promoting economic growth in Sri Lanka and Malaysia (Hoque & Yakob, 2017; Niranjala, 2015). Using panel data and employing generalized methods of moment technique, Beck and Levine (2004) observed a positive impact of stock market development on economic growth. Sharma and Bardhan (2018) applied bootstrap panel Granger causality method to examine the causality analysis for 25 advanced economies for the period from 1975 to 2011 and observed a uni-directional causality from stock market development to growth. Pradhan et al. (2020), using panel data of Group of Twenty (G-20) countries for the period from 1991 to 2016, revealed that stock market development is one of the demonstrable drivers of economic growth in the long run. Chakraborty (2008), based on quarterly data during the period from 1996 to 2005, in India, observed stable long-run causality relationship running from market capitalization to economic growth. Tripathi and Seth (2014) observed that stock market performance precedes the economic variables in the Indian equity market. Another study of the Indian stock market by Sehrawat and Giri (2017) revealed significant influence of sector-specific stock indices on sector-specific economic growth in both the short and the long run.

The growth of an economy creating demand for financial services as tools of development has also been identified in the empirical literature. Rousseau and Wachtel (2000) found the role of financial intermediation to economic growth in time series tests for five countries over the period from 1870 to 1920. Employing autoregressive distributed lag (ARDL) in yearly data from South Africa, Ho (2018) found the role of economic growth increasing the stock market development. Another study by Pan and Mishra (2018) also revealed similar causality relationship in China, using Toda Yamamoto causality test. Using a similar causality test, Tekin and Yener (2019) examined stock market growth nexus in the USA, Germany, Brazil, Russia, India, China and South Africa (BRICS) countries and Turkey for the period from 1998Q1 to 2017Q4. The results showed unilateral causalities running from indexes of stock market to economic growth in the USA, BRICS countries and Turkey. In a recent study by Bouri et al. (2020), economic output measured by industrial production is found to capture the predictability of stock market volatility for India and China of emerging BRICS nations, indicating economic influence on the financial markets. Bidirectional causality between stock market and growth is also observed in the Middle East and North Africa (MENA) region (Puryan, 2017) and in Germany (Tekin & Yener, 2019).

In the Nepalese context, the relationship between stock market development and economic growth is backed by insufficient evidence due to very little empirical studies. Uni-directional causality from stock market development to economic growth was observed in most of the studies, using time series data (Bist, 2017; Pokharel, 2020; Rana, 2014; Regmi, 2012). Surya Bahadur and Neupane (2006), however, using time series data over the period from 1988 to 2005 and employing Granger causality test, had revealed long-run bidirectional causality. Very few studies in the Nepalese perspective create the necessity to examine the association of stock market development and economic growth that can add to the existing empirical literature.

Objectives

The empirical evidence of the impact of stock market on economic growth or the impact of economic growth on stock market is very sparse in Nepal. Thus, to contribute to the empirical literature, the objective of the study is to identify causal relationship between the stock market development and economic growth in Nepal by investigating the long-run relationship, exploring short-run dynamics and identifying the direction of causality between indicators of stock market development and the economic growth.

Rationale of the Study

The study attempts to examine the relationship of size, depth and efficiency dimensions of stock market development as well as their aggregated index with economic growth using the ARDL econometric technique. As a low-income emerging economy, the findings of the study will facilitate the policymakers to design appropriate policies directed towards the development of stock market and the economy as a whole.

Methodology: Variables, Data Description and Econometric Model

Variables and Data Description

Stock market size is market capitalization (CAP) defined as total value of listed domestic securities in the stock exchange relative to gross domestic product (GDP; Hoque & Yakob, 2017). In terms of economic significance, market capitalization indicates the overall market size with the ability to mobilize capital and diversify risk on an economy-wide basis (Levine & Zervos, 1998). Stock market depth is measured as total value of shares traded as a percentage of GDP (Yartey, 2008). The ratio indicates market’s liquidity (LIQ) in terms of quickness of fund transfer between sellers and buyers on an economy-wide basis (Levine & Zervos, 1998; Yartey, 2008). Stock market efficiency is turnover ratio (TVR) measured as the ratio of value of all shares traded to total market capitalization (Yartey, 2008). The ratio captures the liquidity on market basis. According to Levine and Zervos (1998), high turnover usually implies low transaction costs, which may enhance allocative efficiency of the stock market. Annual real GDP per capita growth in percentage (RGDP) is taken as an indicator of economic growth (Guru & Yadav, 2019; Pradhan et al., 2020) expressed in base year 2001. In order to capture the effects of macroeconomic activity in Nepal, the study incorporated a control variable inflation rate (INF) as a proxy of market stability and measured as percentage change in consumer price index. It is observed that higher inflation rates are associated with smaller and lesser liquid stock markets in the theoretical literature (Boyd et al., 2001).

The study used annual time series data over the period from mid-July 1994 to mid-July 2019. Despite the first commercial bank (Nepal Bank Limited) established in 1937 and the first issue of share to the public by Biratnagar Jute Mill in 1936, the economic reforms with tangible financial sector development started only after the adoption of the liberalization policy in the late 1980s. Thus, the motivation for using this data set is to capture economic and financial reforms and also to cover the operational period of the NEPSE that was transformed from the Security Exchange Centre in 1993 to an organized and full-fledged trading floor on 31 January 1994. Real GDP (RGDP) per capita growth and inflation rates are retrieved from the World Development Indicators. Market capitalization, liquidity and turnover are calculated from the data available through annual reports of the SEBON and NEPSE.

The descriptive outcome in Table 2 shows considerable variations in the variables across time. For example, RGDP per capita growth ranges from a low of −1.41% to 6.79%. Market size also shows a very high variation from 4.53% to 83.89%, whereas market depth and market efficiency range from a low of 0.07% to a high of 7.67% and from 1.42% to 11.04%, respectively. The control variable inflation also shows high variations across the study period with a minimum of 2.27% and a maximum of 11.24%. Correlation results show that all three indicators of stock market development are positively correlated with economic growth, but the relationships are not significant. Except for market efficiency, coefficients of inflation with other two stock market indicators and economic growth are positive. All three indicators of stock market development share high correlation among themselves.

Table 2.

Descriptive Statistics and Correlations

	RGDP	CAP	LIQ	TVR	INF
Mean	3.25	26.22	1.66	5.02	6.81
Std. Dev.	1.92	22.52	2.11	2.53	2.85
Min	–1.41	4.53	0.07	1.42	2.27
Max	6.79	83.89	7.67	11.04	11.24
Obs.	26	26	26	26	26
Correlations
RGDP	1
CAP	0.324	1
LIQ	0.274	0.935***	1
TVR	0.158	0.630***	0.775***	1
INF	0.046	0.180	0.003	-0.213	1

Source: E-views 10, authors' calculations

Notes: RGDP=percentage change in real GDP per capita; CAP=stock market capitalization as a percentage of nominal GDP;

LIQ=Total shares traded as percentage of nominal GDP; TVR=total shares traded as a percentage of market capitalization;

INF=percentage change in consumer price index (Year 2010=100)

*** shows the correlation is significant at the 0.01 level (2-tailed)

Principal Component Analysis

This strong correlation observed in Table 2 shows that stock market development indicators may contain common information or may measure the same thing, which may lead to multicollinearity and also over-parameterization problems. Principal component analysis (PCA), to some extent, addresses the problem of multicollinearity because the variables will be reduced to fewer numbers with more interdependence rather than having causal relations (Coskun et al., 2017).

PCA with subsequent rotation (varimax) is conducted on three indicators of stock market development. The value of Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy obtained is 0.562 > 0.5 along with a p-value of Bartlett’s test of sphericity 0.000 < 0.05 (null hypothesis of correlation matrix being identity matrix is rejected) and thus support the application of PCA.

The results of the PCA extraction of stock market indicators are presented in Table 3. The stock market development indicator corresponds to the first principal component only, which has an eigenvalue greater than 1, and it also explains about 86% of the total variance, which is then named as the stock market development index (SMDI) that will be carried into subsequent analysis as a representative of the level of development in the Nepalese stock market.

Table 3.

Principal component analysis

Principal components	Eigen Values	Difference	Proportion of variance	Cumulative	Cumulative proportion
1	2.566	2.174	85.541	2.566	85.541
2	0.392	0.351	13.078	2.958	13.078
3	0.041		1.381	2.999	1.381
Factor loadings
Variables	PC 1
CAP	0.931
LIQ	0.981
TVR	0.859

Source: SPSS-20, authors' calculations

Notes: PC=principal component factor; CAP=stock market capitalization as a percentage of nominal GDP; LIQ=Total shares traded as percentage of nominal GDP; TVR=total shares traded as a percentage of market capitalization

Empirical Model and Methodology

To develop an empirical strategy that would enable to describe the impact of stock market development indicators on RGDP, a general model used for the estimation is represented by Equation (1).

{RGDP}_{t} = δ_{0} + δ_{smd i} SMDI t + δ_{inf} {INF}_{t} + μ_{t}

(1)

where RGDP _t is real GDP per capita growth at time t, SMDI _t is the aggregate measure of stock market development at time t, INF _t is the inflation at time t and µ_t is the usual error term. A similar equation can be written with SMDI as dependent variable which will allow for the possibility of causality from either direction. For sensitivity analysis, the same equation is used with disaggregated indicators of stock market development as target variables and regressors.

The stationary properties of the annual time series are analysed by employing Ng and Perron (2001) unit root test. The choice of Ng-Perron test is considered, which gives robust results over the other traditional tests for a small sample (Giri & Joshi, 2015; Rehman et al., 2016). After verifying that the variables attain stationarity either at level or at the first difference, the structural break in the series is also examined using Zivot and Andrew test for a single break after which long-run cointegration between the variables is examined by employing the bounds test approach proposed by Pesaran et al. (2001). ARDL model is used in order to analyse the existence of short-run and long-run static spillover relationships between stock market development and the economic growth. The error correction term in short-run dynamics will reveal the direction of Granger causality between the variables. The following ARDL model is used to examine the long-run and short-run relationships between stock market development and growth variables.

{ΔRGDP}_{t} = α_{0} + λ_{1} {RGDP}_{t - 1} + λ_{2} {SMDI}_{t - 1} + λ_{3} {INF}_{t - 1} + \sum_{i = 1}^{p} γ_{1 i} {ΔRGDP}_{t - i} + \sum_{j = 0}^{q} γ_{2 j} {ΔSMDI}_{t - j} + \sum_{k = 0}^{r} γ_{3 k} {ΔINF}_{t - l} + γ_{4} {DUM}_{t} + ε_{t}

(2)

Where α₀ is a drift component, ∇ is the first difference operator and ε_t is the white noise residual. The coefficients $λ_{1} - λ_{3}$ represent the long-run coefficients and $γ_{1} - γ_{3}$ represent short-term dynamics of the model. The values (p, q, r) are the selected number of optimum lags for cointegrating equations, based on Schwarz Bayesian Criteria (SBC). DUM is the dummy variable used to capture the structural breaks, where DUM _t = 0, if t ≤ B_t or the break year and 1 above B_t (Perron, 1989).

From Equation (2), the existence of cointegration is confirmed by testing the null hypothesis of no cointegration H₀: $λ_{1} = λ_{2} = λ_{3} = 0$ against the alternative hypothesis of presence of cointegration H₁: $λ_{1} \neq λ_{2} \neq λ_{3} \neq 0$ . The variables are cointegrated if we reject the null hypothesis, which is done if F-stastistic is greater than the upper bound critical value given by Narayan and Narayan (2005) for a small sample size. If the F-statistic falls below the lower bound values, we do not reject the null hypothesis of no cointegration, and the decision is inconclusive if it falls between the two bounds. The cointegration confirmation leads to the estimation of the long-run model as in Equation (3), and short-run dynamics, based on error correction term, as in Equation (4).

{ΔRGDP}_{t} = α_{0} + λ_{1} {RGDP}_{t - 1} + λ_{2} {SMDI}_{t - 1} + λ_{3} {INF}_{t - 1} + ε_{t}

(3)

{ΔRGDP}_{t} = α_{0} + \sum_{i = 1}^{p} γ_{1 i} {ΔRGDP}_{t - i} + \sum_{j = 0}^{q} γ_{2 j} {ΔSMDI}_{t - j}

(4)

+ \sum_{k = 0}^{r} γ_{3 k} {ΔINF}_{t - k} + μ {ECT}_{t - 1} + ε_{t}

where ECT _t ₋₁ is the lagged value of residuals obtained from the long-run regression estimation. ECT indicates both long-run causality and the speed of adjustment. The coefficient of the error correction term µ is expected to be negative, which implies that when variables drift apart from the equilibrium in the short run, they can quickly correct back to their equilibrium levels. The study also applies residual diagnostics such as the Lagrange multiplier test of residual serial correlation, Jarque–Bera test of normality and Breusch–Pagan–Godfrey test for heteroscedasticity along with Ramsey’s RESET test to check the model misspecification if any.

Empirical Analysis

Unit Roots and Structural Breaks

The stationarity properties of the variables are investigated using Ng and Perron (2001) unit root test in which MZ _a and MZ _t tests indicate unit root as the hull hypothesis, whereas MSB and MPT indicate stationarity as the null hypothesis.

The robustness of this test is also determined by the conventional augmented Dickey–Fuller and Philips–Perron unit root tests carried out with drift. The results of the tests are presented in Table 4. RGDP per capita growth and inflation rate are integrated at level, whereas SMDI is integrated at the first difference.

Table 4.

Ng-Perron, Augmented Dickey-Fuller and Philips-Perron unit root test results

Variables	MZ_a	MZ_t	MSB	MPT	ADF	PP	Stationarity
RGDP	-11.725**	-2.405**	0.205**	2.148**	-4.494***	-4.512***	I(0)
SMDI	-4.677	-1.452	0.310	5.384	-1.730	-1.730
INF	-10.947**	-2.323**	0.212**	2.298**	-3.402**	-3.402**	I(0)
∇RGDP	-11.077**	-2.343**	0.211**	2.249**	-4.822***	-16.891***
∇SMDI	-11.630**	-2.378**	0.204**	2.233**	-5.413***	-5.433***	I(1)
∇INF	-9.923**	-2.213**	0.223**	2.521**	-7.250***	-13.011***
Ng-Perron critical values
	MZa	MZt	MSB	MPT
1% [***]	-13.800	-2.580	0.174	1.780
5% [**]	-8.100	-1.980	0.233	3.170

Source: E-views 10, authors' calculations

Note: *** and ** indicate statistical significance at 1% and 5% level; T&I: Trend and Intercept; IO: Innovational Outlier

Table 5 shows the outcome of Zivot and Andrew unit roots test with one break in the series. Both trend and intercept break specification are used to identify the break in the series. The break type employed is an innovational outlier. The test reveals 2015 as break year for both RGDP and SMDI, whereas 2007 as break year for INF. The breaks in GDP and SMDI can be attributed to the catastrophic 7.8 magnitude earthquake on 25 April 2015 that affected industrial and service sectors for a long period of time. GDP growth plunged to almost 3% in 2015 from around 6% in 2014. The computed SMDI also has experienced a huge drop from 1.16 in 2014 to 0.8 in 2015. The break in inflation can be attributed to the global financial crisis of 2007–2008. The inflation sharply increased from 2.27 in 2007 to 9.91 in 2008.

Table 5.

Zivot and Andrew unit root test for one break at level

Variables	lag length	Break Specification	Break Type	Break	t-statistic	Inference
RGDP	1	T&I	IO	2015	-5.211**	Stationary
SMDI	4	T&I	IO	2015	-3.368	Nonstationary
INF	0	T&I	IO	2007	-5.100	Nonstationary

Source: E-views 10, authors' calculations

Notes: *** and ** indicate statistical significance at 1% and 5% level; T&I: Trend and Intercept; IO: Innovational Outlier

According to Perron (1989), structural breaks caused by such events, if present in the series, are found to affect the statistical power of the stationarity tests by accepting the null hypothesis of unit roots when there is clear evidence of no unit roots. Thus, the impact of the break year on the dependent variable is accounted by regressing a dummy variable of the dependent variable’s break year.

Autoregressive Distribution Lag Bounds Testing

Table 6 presents long-run relationship between SMDI and RGDP per capita growth employing ARDL bound testing procedure. Both RGDP and SMDI are treated as dependent variables in each of the two models estimated in Table 6. The equations in levels are presented vertically with coefficients, followed by bounds test statistic. The unrestricted models are estimated for each variable whose lag selection is carried out using Schwartz Bayesian Criterion (SBC). The coefficients of equations in levels are presented in Table 6 along with bounds test results.

Table 6.

Long-run and short-run relationships

Panel A: Long run relationships, ARDL and bounds test result				Panel B: Short-run and error correction form
Levels Equations	RGDP (1,1,0)	SMDI (1,0,0)		Selected model	∇RGDP (1,1,0)	∇SMDI (1,0,0)
RGDP			0.079	∇RGDP		0.061
SMDI	2.078**			∇SMDI	0.456
INF	-0.032	0.044		∇INF	-0.032	0.034
				C	4.113***	-0.663**
F-statistic	10.554***	2.592		ECT(-1)	-1.00***	-0.765***
				DUM	-3.422***	1.417**
		5%	1.0%	χ ² SC (1)	0.049(0.823)	0.214(0.644)
Finite sample: n=30	I(0)	3.710	5.333	JB NOR	4.066(0.131)	1.391(0.498)
k=3	I(1)	5.018	7.063	χ ² HET (1)	2.116(0.832)	7.830(0.098)
				F RESET	0.292(0.595)	0.605(0.446)
				R²_aj	0.5725	0.222
				F-Statistic	11.714***	4.418**

Source: E-views 10, authors' calculations

Notes: Critical values for the bounds test: case III: unrestricted intercept and no trend, Narayan (2005, p. 1988); I(0), lower bound; I(1), upper bound. *** and ** indicate statistical significance at 1 and 5% respectively; SC, Serial correlation; NOR, Normality; HET, Heteroscadasticity; RESET, Functionality. Italic figures in parenthesis are p-values.

An F-statistic of 10.544 reveals that there is statistical evidence for long-run dynamic relationship between economic growth (RGDP) and stock market development (SDMI) in the long run running from stock market development to economic growth. The long-run relationship is consistent with the findings of Chakraborty (2008) and Pradhan et al. (2020). The positive impact of stock market on economic growth is also steady with empirical studies (Beck & Levine, 2004; Durusu-Ciftci et al., 2017; Levine & Zervos, 1998). However, the alternate model with economic growth as regressor shows no cointegration. This supports the existence of uni-directional causality from stock market development to economic growth (Bist, 2017; Pokharel, 2020; Rana, 2014; Regmi, 2012; Sharma & Bardhan, 2018). The short-run dynamics in both the models reveal no significant causal impact running from either direction supporting Lucas (1988) at least in the short run. The control variable inflation shows negative coefficient with economic growth and positive coefficient with SMDI in both long-run and short-run models. The tests carried out to check the robustness show that the estimated models are free from serial correlation, residuals are normally distributed and have constant variance and have no specification bias.

Alternative Estimations for Sensitivity Analysis

Ng-Perron unit root test and Zivot Andrew test for structural break are presented in Table 6. The unit root shows that annual series of stock market size and liquidity are integrated at level, and market efficiency is integrated at the first difference. The influence of earthquake in 2015 and the global financial crisis of 2007–2008 are clearly seen as the individual indicators of stock market as observed in Zivot Andrew structural break test in Table 7.

Table 7.

Unit root and structural break tests

Ng-Perron unit root test
Variables	MZa	MZt	MSB	MPT
∇CAP	-11.984**	-2.445**	0.204**	2.051**
∇LIQ	-11.990**	-2.427**	0.202**	2.124**
TVR	-10.442**	-2.238**	0.214**	2.521**
Ng-Perron critical values
	MZa	MZt	MSB	MPT
1%	-13.800	-2.580	0.174	1.780
5%	-8.100	-1.980	0.233	3.170
Zivot Andrew structural break test with T & I break specification and IO break type
Variables	lag length	Break	t	Inference
CAP	5	2007	-8.694***	Stationary
LIQ	3	2015	-5.374**	Stationary
TVR	4	2010	-5.278**	Stationary

Source: E-views 10, authors' calculations

Notes: *** and ** indicate statistical significance at 1% and 5% level; T&I: Trend and Intercept; IO: Innovational Outlier

Equation (1) is estimated again using RGDP as dependent variables and CAP, LIQ and TVR as regressors in the respective models. To examine the reverse direction of cointegration, the same equation is used with CAP, LIQ and TVR as dependent variables and RGDP as regressor. Results of the bounds tests are presented in Table 8, which reveal that all the models with RGDP as the dependent variable show strong cointegration and weak cointegration with TVR as the dependent variable. Market size and market depth are seen to have significant long-run relationship with the economic growth but show no short-run relationship among the variables. The significant negative error correction terms in both the models indicate that market capitalization and liquidity Granger cause economic growth in the long run. They also show that any deviation of stock market development indicators from equilibrium in the short run is corrected wholly in the coming period. The results of both the models are consistent with the result of aggregated index of stock market development in ARDL bounds testing section. Turnover ratio, despite showing cointegration with economic growth, does not provide significant coefficients both in the short run and in the long run.

Table 8:

Bounds test, long-run and short-run coefficients of disaggregated stock market indicators

ARDL Models	(1,1,0)	(1,1,0)	(1,0,0)	(1,0,0)	(1,0,0)	1,0,0)
Dependent variable	RGDP			CAP	LIQ	TVR
F-statistic	12.811***	11.613***	6.426**	1.416	2.851	4.525*
Long-run coefficients
RGDP				-3.921	0.129	-0.005
CAP	0.078**
LIQ		1.057**
TVR			0.105
INF	-0.075	0.006	0.041	0.483	0.125	-0.300
Short-run coefficients
ΔRGDP				-1.673	0.091	-0.004
ΔCAP	-0.002
ΔLIQ		0.155
ΔTVR			0.104
ΔINF	-0.080	0.006	0.041	0.206	0.088	-0.223
DUMMY	-2.739***	-3.809***	0.494	19.418**	2.831**	1.658
C	2.343***	2.177***	2.230***	60.196*	-0.205	4.677***
ECT(-1)	-1.065***	-1.002***	-0.993***	-0.426***	-0.704***	-0.743***
χ² SC (1)	0.12(0.72)	0.28(0.59)	0.72(0.39)	0.09(0.75)	0.07(0.79)	0.13(0.71)
JB NOR	3.99(0.14)	1.95(0.37)	2.70(0.26)	4.99(0.08)	4.36(0.11)	1.39(0.49)
χ² HET (1)	3.39(0.64)	3.38(0.64)	3.96(0.41)	9.51(0.05)	12.65(0.02)	1.33(0.86)
F RESET	0.75(0.39)	1.20(0.29)	0.72(0.41)	2.09(0.16)	1.38(0.26)	3.54(0.08)
R²_aj	0.62	0.59	0.45	0.28	0.21	0.35
F-Statistic	14.32***	12.94***	10.63***	5.61**	4.23**	7.49***

Source: E-views 10, authors' calculations

Notes: ***, ** and * indicate the level of significance at 0.01, 0.05 and 0.10 respectively. Italic figures in parenthesis are p-values. Critical values for the bounds test (F-statistic): case III: unrestricted intercept and no trend, Narayan (2005, p. 1988); I(0), lower bound; I(1), upper bound. For k=3, at 1% and 5%, I(0) are 5.333 and 3.710 respectively and I(1) are 7.063 and 5.018 respectively. SC, Serial correlation; NOR, Normality; HET, Heteroscadasticity; RESET, Functionality

When CAP and LIQ are used as dependent variables, no model is found to show cointegration. The results are consistent with the findings from the model estimation with aggregated index in ARDL bounds testing section. Although the model with TVR as dependent variable shows mild cointegration, the coefficients of regressors are not significant.

The results in Table 8 show that it is only the stock market development indicators—mainly CAP and LIQ—that impact economic growth in the long run. Inflation is observed to have a negative coefficient in relation to RGDP when market capitalization is used as the independent variable. A similar relationship is also observed in Table 6 of ARDL bounds testing section. The robustness checks of all the models in Table 8 show no serial correlation and RESET tests show that all the models are stable.

The overall results suggest that the development of stock market encourages the economic growth. In particular, the ability of the stock market to mobilize capital and diversify risk contributes to the economic growth along with the easiness with which stocks are traded in the secondary market. With the incorporation of possible structural breaks and the estimated models being free from serial correlation with no specification bias, the findings of the study can be considered robust and thus can contribute to the empirical literature in the context of low-income economies.

Conclusion

In response to the enduring debate of stock market development causing economic growth or vice versa, the analysis of both aggregated and disaggregated stock market variables reveal that stock market development positively impacts economic growth, and there is a uni-directional long-run causality that runs from stock market development to economic growth in Nepal, supporting supply-leading hypothesis. The negative coefficient of inflation with economic growth suggests adverse impact of market instability on economy, whereas positive coefficient with stock market suggests stock investment as hedge against inflation.

The study provides an understanding to both the investors and the regulators that even in a small emerging economy, the stock market can play a significant role to contribute towards economic development. The Nepalese stock market has just started gaining attention in terms of firms’ participation, investors’ involvement and regulators’ keen monitoring and development policies. The findings of the study would thus have implications on companies and investors, motivating their more active participation for capital formation and liquidity. In addition, the results reveal the ability of the stock market to mobilize capital and diversify risk on an economy-wide basis, requiring the government to promote the stock market as a good investment platform. Furthermore, stock market regulators such as Nepal Rastra Bank (NRB), SEBON and NEPSE should incorporate different strategies to develop the market and make stock buy and sale easy for the investors.

Stock market index and market concentration may also be used in future research along with inclusion of other macroeconomic stability factors like interest rate and exchange rate as control variables. Sectoral stock market indicators may also be used to identify sector-specific relationships.

Footnotes

Acknowledgement

The authors are grateful to the anonymous referees of the journal for their extremely useful suggestions to improve the quality of the article. Usual disclaimers apply.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

ORCID ID

Keshab Khatri Chettri

References

Baier

S. L.

, Dwyer

G. P.

, & Tamura

(2004). Does opening a stock exchange increase economic growth? Journal of International Money and Finance , 23(3), 311–331.

Beck

, & Levine

(2004). Stock markets, banks, and growth: panel evidence. Journal of Banking and Finance , 28(3), 423–442.

Bist

J. P.

(2017). Stock market development and economic growth in Nepal: An ARDL representation. Journal of Finance and Economics , 5(4), 164–170.

Bouri

, Demirer

, Gupta

, & Sun

(2020). The predictability of stock market volatility in emerging economies: Relative roles of local, regional, and global business cycles. Journal of Forecasting , 2020, 1–9.

Boyd

J. H.

, Levine

, & Smith

B. D.

(2001). The impact of inflation on financial market performance. Journal of Monetary Economics , 47(2), 221–248.

Chakraborty

(2008). Does financial development cause economic growth? The case of India. South Asia Economic Journal , 9, 109–139.

Coskun

, Seven

, Ertugrul

H. M.

, & Ulussever

(2017). Capital market and economic growth nexus: Evidence from Turkey. Central Bank Review , 30(1), 1–11.

Demirgüç-Kunt

, & Levine

(1996). Stock markets, corporate finance, and economic growth: An overview. World Bank Economic Review , 10(2), 223–239.

Demirgüç-Kunt

, & Levine

(1999). Bank-based and market-based financial systems: Cross-country comparisons (Policy Research Working Paper. WP/2143/1999). The World Bank.

10.

Durusu-Ciftci

, Ispir

M. S.

, & Yetkiner

(2017). Financial development and economic growth: Some theory and more evidence. Journal of Policy Modeling , 39(2), 290–306.

11.

Giri

A. K.

, & Joshi

(2015). Macroeconomic variables and stock market development in India: An application of ARDL bounds testing approach. The Empirical Economics Letters , 14(7), 705–718.

12.

Goldsmith

R. W.

(1969). Financial structure and development . Yale University Press.

13.

Guru

B. K.

, & Yadav

I. S.

(2019). Financial development and economic growth: Panel evidence from BRICS. Journal of Economics, Finance and Administrative Science , 24(47), 113–126.

14.

S. Y.

(2018). Macroeconomic determinants of stock market development in South Africa. International Journal of Emerging Markets , 14(1), 1–29.

15.

Hoque

M. E.

, & Yakob

N. A.

(2017). Revisiting stock market development and economic growth nexus: The moderating role of foreign capital inflows and exchange rates. Cogent Economics and Finance , 5, 1–17.

16.

Levine

(1991). Stock markets, growth and tax policy. Journal of Finance , 46(4), 1445–1465.

17.

Levine

, & Zervos

(1998). Stock markets, banks, and economic growth. American Economic Review , 88(3), 537–58.

18.

Lucas

R. E.

(1988). On the mechanics of economic development. Journal of Monetary Economics , 22(1), 3–42.

19.

Narayan

P. K.

, & Narayan

(2005). Estimating income and price elasticities of imports for Fiji in a cointegration framework. Economic Modelling , 22, 423–438.

20.

, & Perron

(2001). Lag length selection and the construction of unit root tests with good size and power. Econometrica , 69(6), 1519–1554.

21.

Niranjala

S. A. U.

(2015). Stock market development and economic growth in Sri Lanka. Global Journal of Management and Business Research: B , 15(8), 39–44.

22.

Pan

, & Mishra

(2018). Stock market development and economic growth: Empirical evidence from China. Economic Modelling , 68(1), 661–673.

23.

Patrick

H. T.

(1966). Financial development and economic growth in underdeveloped countries. Economic Development and Cultural Change , 14, 174–189.

24.

Perron

(1989). The great crash, the oil price shock, and the unit root hypothesis. Econometrica , 57(6), 1361–1401.

25.

Pesaran

, Shin

, & Smith

(2001). Bounds testing approaches to the analysis of level relationship. Journal of Applied Economics , 16(3), 289–326.

26.

Pokharel

(2020). Capital market linkage to economic growth in Nepal. Westcliff International Journal of Applied Research , 1(1), 44–56.

27.

Pradhan

R. P.

, Arvin

M. B.

, Norman

N. R.

, & Bahmani

(2020). The dynamics of bond market development, stock market development and economic growth. Journal of Economics, Finance and Administrative Science , 25(49), 119–147.

28.

Puryan

(2017). The causal relationship between economic growth, banking sector development and stock market development in selected Middle-East and North African countries. International Journal of Economics and Financial Issues , 7(3), 575–580.

29.

Rana

S. B.

(2014). Stock market size, liquidity and economic growth in Nepal. The KIC Journal of Management and Economic Review , 1(2), 1–13.

30.

Regmi

U. R.

(2012). Stock market development and economic growth: empirical evidence from Nepal. Administration and Management Review , 24(1), 1–28.

31.

Rehman

I. U.

, Mahdzan

N. S.

, & Zainudin

(2016). Is the relationship between macroeconomy and stock market liquidity mutually reinforcing? Evidence from an emerging market. International Journal of Monetary Economics and Finance , 9(3), 294–316.

32.

Rousseau

P. L.

, & Sylla

(2003). Financial systems, economic growth, and globalization (NBER Working Paper No. 8323, June 2001, and forthcoming). In Bordo

, Taylor

, & Williamson

(Eds.), Globalization in historical perspective . University of Chicago Press.

33.

Rousseau

P. L.

, & Wachtel

(2000). Equity markets and growth: Cross-country evidence on timing and outcomes, 1980–1995. Journal of Banking and Finance , 24, 1933–1957.

34.

Schumpeter

J. A.

(1911). The theory of economic development . Harvard University Press.

35.

Sharma

, & Bardhan

(2018). Stock market development and economic growth: Evidence from bootstrap panel Granger causality test. Journal of Economic Development , 43(3), 57–83.

36.

Shaw

(1973). Financial deepening in economic development . Oxford University Press.

37.

Sehrawat

, & Giri

A. K.

(2017). A sectoral analysis of the role of stock market development on economic growth: Empirical evidence from Indian economy. Global Business Review , 18(4), 1–13.

38.

Surya Bahadur

G. C.

, & Neupane

(2006). Stock market and economic development: A causality test. The Journal of Nepalese Business Studies , 3(1), 36–44.

39.

Tekin

, & Yener

(2019). The causality between economic growth and stock market in developing and developed countries: Toda–Yamamoto approach. Theoretical and Applied Economics , 26(2), 79–90.

40.

Tripathi

, & Seth

(2014). Stock market performance and macroeconomic factors: The study of Indian equity market. Global Business Review , 15(2), 291–316.

41.

Yartey

C. A.

(2008). The determinants of stock market development in emerging economies: Is South Africa different? (IMF Working Papers Vol. 8 Issue 32). International Monetary Fund.