This paper focuses on the impact of the concurrent liberalization of current and capital accounts and quality institutions on stock market development. Using annual data from 1996-2013 for a panel of fifty three (53) developed and developing countries and utilizing dynamic GMM estimators, the results show that banking sector development, economic growth, and the interaction term affect stock market development positively. The paper finds that capital account liberalization affects market development negatively, but the effect of capital account liberalization on market development is contingent on the level of economic growth and development. Further, the results revealed that the impact of trade openness on stock market development is mixed. The research finds negative impact of institutional factors on market development. Finally, the paper does not find support in favour of simultaneous openness hypothesis.
One important issue at the centre of financial development and economic growth studies is that of the determinants of financial sector development. Factors such as financial and trade liberalization, quality institutional factors etc are argued to affect financial sector development [
While there is strong support for the positive impact of institutional factors on financial sector development, yet there are some contrary empirical findings [
For instance, while trade between developed countries may be seen as trade between or among equal partners, trade between developing and developed countries may be seen as trade between unequal partners. Therefore, the gains of trade liberalization in these two cases may differ between and among countries [
Finally, one major problem confronting studies on financial sector development is that of the choice of appropriate measures of financial sector development. There are several measures of financial sector development, with each having its merits and demerits. This study hopes to address this challenge by constructing principal component, from three measures each, of banking sector and stock market development.
This paper examines overall, the impact of liberalization and institutional quality on stock market development.
The rest of the paper is organized as follows: Section 2 is the literature review, Section 3 discusses the methodology of our work, Section 4 presents and analyses our results, finally, Section 5 concludes.
While there is strong support for the positive impact of institutional factors on financial sector development [
Following Law and Shah Habibullah [
where PCS, PCB, Y, TO, CAO, INST, and μ denote principal component of stock market development, principal component of banking sector development, economic growth, trade openness, capital account openness, institutional factor, and the error term respectively. Economic growth is measured by real GDP per capita while institutional factor is represented by an index of institutional quality, constructed from four measures namely, corruption control, bureaucracy, voice and accountability, and government effectiveness, and assumes a value of less or equal to 40. Stock market development is measured by the principal component of stock market development, constructed from the ratios of stock market capitalization to GDP, total value traded to GDP, and the turnover ratio. Banking sector development is measured by the principal component of banking sector development, constructed from the ratios of liquid financial liability to GDP, credit to the private sector by the deposit money bank to GDP, and deposit money bank assets to the deposit plus central bank assets. Trade openness is proxied by the ratio of the sum of export plus import to GDP. Capital account openness is measured by the Capital Account Openness Index, adopted from Chinn and Ito [
Following the argument that financial liberalization may boost financial sector development and economic growth if accompanied by other economic and institutional reforms such as stable macroeconomic conditions, political and policy stability, trade liberalization, etc Rajan and Zingales [
where INT is the interaction term, all other variables are as defined above.
To measure the effect of financial openness on financial sector development after controlling for country’s stage/level of development, we introduce dummies (d1 and d2) for developed and developing respectively and interacts the dummies with measure of capital account openness. Therefore, our new models after controlling for stage of development is specified as thus below:
Principal component analysis is a mathematical method which recognises patterns in datasets in a way that highlight similarities and differences in the original datasets while retaining most of the information contained in the original datasets in the principal component (Smith, 2002). It is defined as a statistical method that simplifies complex datasets by reducing the dimension of their matrix, where the reduced data matrix accounts for most of the variations in the original datasets/matrix while at the same time mutually uncorrelated and orthogonal (Raychaudhuri et al., 2000).
Principal component analysis models the structure of the variations of a number of variables using their linear combinations known as the components, which can be used for further analysis. This approach is appropriate if we wish to develop a smaller number of artificial measures from a large number of observed measures that will account for most of the changes in the original observed variables. In addition, it addresses the problem of multi-collinearity in our study.
This paper uses Dynamic Panel Generalized Method of Moment estimators (hereafter GMM) as proposed by Holtz-Eakin et al. (1988) and later extended by Blundell and Bond [
GMM is defined as the moment conditions formed under the assumptions that particular lagged levels of the dependent variables are uncorrelated to the difference disturbances [
If we take the difference of the level equations so as to remove the unobserved individual effects, the resulting equation is known as the Difference GMM estimator, presented as below;
where X* are exogenous variables. Based on the condition above, the lagged levels of the explanatory variables are used as instruments for the difference equations [
Arellano and Bover [
where μ and γ are the individual specific effect which is invariant to time and the error term respectively. Following Arellano and Bover [
Annual data from 1996-2013 averaged over three (3) years (thus giving us time series data of 6 years) in a sample of fifty three countries was used. Data on measures of institutional quality is collected from World Bank Governance Indicator database. Data on measures of banking sector and stock market developments are collected from financial structure database. Data on economic growth and trade openness are collected from World Bank Development Indicator database. Data on capital account openness is obtained from Chinn and Ito. Data on principal component of banking and stock market development are constructed from three measures of banking and stock market development each respectively.
TWO STEP DIFFERENCE GMM | TWO STEP SYSTEM GMM | |
---|---|---|
PCSt-1 | 0.7714*** (0.115) [0.000] | 0.5459*** (0.0687) [0.000] |
PCB | 0.0023*** (0.0005) [0.004] | 0.0025*** (0.0003) [0.000] |
LY | 1.9301*** (0.203) [0.000] | 1.621*** (0.1507) [0.000] |
LTO | 5.4322*** (1.102) [0.000] | 4.2203*** (0.5491) [0.000] |
LCAO | −8.0283 (5.8971) [0.201] | −8.9749** (4.0329) [0.002] |
LINST | −7.3512*** (1.0032) [0.000] | −8.0331*** (1.0081) [0.000] |
INT | 2.2521 (1.1057) [0.046] | 2.0414*** (0.6530) [0.004] |
AR 2 | −2.5461*** [0.0501] | −1.9861 [0.1079] |
SARGAN TEST | 9.4548 [0.1023] | 16.2450 [0.0884] |
NOTE; LY = log of real GDP per capita, PCB = principal component of banking sector development, PCS = principal component of stock market development, ltrade = log of trade openness, LCAO = log of capital account openness, LINST = log of institutional quality, LINT = simultaneous opening of trade and capital accounts. N = 53. Values in ( ) and [ ] denotes standard errors and p-values respectively. The sign *** and ** denotes significance at 1% and 5% respectively.
TWO STEP DIFF GMM | TWO STEP SYSTEM GMM | |
---|---|---|
PCSt−1 | 0.298 (0.105) [0.0024] | 0.2699 (0.1006) [0.008] |
LY | 2.254 (0.3412) [0.000] | 1.7567 (0.3301) [0.000] |
PCB | 0.0197 (0.0004) [0.000] | 0.0215 (0.000) [0.000] |
LTO | −0.5746 (1.0074) [0.446] | −1.3397 (0.7116) [0.069] |
LCAO | −15.3796 (3.789) [0.000] | −10.8811 (1.9028) [0.000] |
INT | 2.9098 (0.7417) [0.000] | 2.0021 (0.5236) [0.000] |
LINST | −1.6143 (0.4910) [0.000] | −1.6552 (0.3392) [0.000] |
D1CAO | 7.4897 (1.1245) [0.000] | 2.0679 (0.4996) [0.000] |
SARGAN TEST | 19.9897 [0.0295] | 23.5695 [0.5717] |
AR 2 | −0.8907 [0.4825] | −0.9904 [0.4433] |
NOTE; LY = log of real GDP per capita, PCB = principal component of banking sector development, PCS = principal component of stock market development, LTO = log of trade openness, LCAO = log of capital account openness, LINST = log of institution, Int = simultaneous opening of trade and capital accounts, D1 = dummy for developing countries. N = 53. Values in ( ) and [ ] denotes standard errors and p-values respectively. ***, ** denotes significance at 1% and 5% respectively.
Robust tests were run to check the consistency of the findings. The measure of institutional quality was substituted with two other measures (government effectiveness and voice and accountability). The results of the robust tests are identical to those in Tables 3-5. The results of the robust tests are presented in
The finding of a positive impact of economic growth on stock market development is expected. It is argued that as household income increases to certain level, there will be portfolio adjustment and their demand for financial and capital assets increases. This will lead to increase in the activities of the stock market and therefore, promotes stock market development. This finding is consistent with Chakraborty [
The finding of a positive effect of banking sector development on stock market is equally expected. Banking sector performs certain functions that are important to stock market development such as the provision of margin loans to stock brokers. This therefore justifies the positive impact of banking sector development on stock market development. This result is in line with Chinn and Ito [
One area of our finding that is of great empirical contention is openness and stock market development. First, proponents of trade openness argued that it will increase investment, growth, and finally leads to stock market development. Opponents of trade openness on the other hand contest that it will lead to the collapse of infant firms, which will lower investment and growth, increase unemployment, and consequently deter stock market
TWO STEP DIFF GMM | TWO STEP SYSTEM GMM | |
---|---|---|
PCSt−1 | 0.3980 (0.1142) [0.004] | 0.3671 (0.1009) [0.001] |
LY | 2.5752 (0.4304) [0.000] | 1.9871 (0.3445) [0.000] |
PCB | 0.1932 (0.0002) [0.000] | 0.1084 (0.0002) [0.000] |
LTO | −0.5976 (1.0005) [0.535] | −0.5232 (0.3652) [0.251] |
LCAO | −8.5009 (2.7632) [0.000] | −9.2253 (2.2466) [0.000] |
INT | 2.6156 (0.5522) [0.000] | 2.7318 (0.6013) [0.000] |
LINST | −1.6001 (0.3219) [0.000] | −1.5307 (0.2102) [0.000] |
D2CAO | −6.9879 (1.0319) [0.000] | −1.2146 (0.3408) [0.002] |
SARGAN TEST | 20.9101 [0.0305] | 22.8437 [0.5301] |
AR 2 | −1.2472 [0.4632] | −1.8315 [0.3071] |
NOTE; LY = log of real GDP per capita, PCB = principal component of banking sector development, PCS = principal component of stock market development, LTO = log of trade openness, LCAO = log of capital account openness, LINST = log of government effectiveness, INT = simultaneous opening of trade and capital accounts, D2 = dummy for developed countries. N = 53. Values in ( ) and [ ] denotes standard errors and p-values respectively. ***, ** denotes significance at 1% and 5% respectively.
TWO STEP DIFF GMM | TWO STEP SYSTEM GMM | |
---|---|---|
PCSt−1 | 0.3162 (0.1026) [0.006] | 0.2676 (0.1001) [0.003] |
LY | 2.3516 (0.3732) [0.000] | 1.0908 (0.1901) [0.000] |
PCB | 0.1349 (0.0003) [0.000] | 0.2016 (0.0001) [0.000] |
LTO | 0.8301 (1.1606) [0.51] | −0.1536 (0.4106) [0.97] |
LCAO | −3.9917 (2.0103) [0.051] | −5.0295 (1.2131) [0.000] |
INT | 1.9067 (0.8407) [0.003] | 2.3901 (0.3705) [0.000] |
LGE | −1.7843 (0.4123) [0.000] | −1.5588 (0.3296) [0.000] |
---|---|---|
SARGAN TEST | 21.0941 [0.033] | 20.6732 [0.197] |
AR 2 | −1.5563 [0.4392] | −1.7681 [0.4807] |
NOTE; LY = log of real GDP per capita, PCB = principal component of banking sector development, PCS = principal component of stock market development, LTO = log of trade openness, LCAO = log of capital account openness, LGE = log of government effectiveness, INT = simultaneous opening of trade and capital accounts. N = 53. Values in ( ) and [ ] denotes standard errors and p-values respectively. ***, ** denotes significance at 1% and 5% respectively.
TWO STEP DIFF GMM | TWO STEP SYSTEM GMM | |
---|---|---|
PCSt−1 | 0.6198 (0.1091) [0.004] | 0.4291 (0.0537) [0.000] |
LY | 1.9797 (0.3027) [0.000] | 1.4516 (0.1134) [0.000] |
PCB | 0.0027 (0.0003) [0.000] | 0.0019 (0.0002) [0.000] |
LTO | 0.1528 (1.0721) [0.978] | 0.1997 (0.4039) [0.625] |
LCAO | −5.0056 (2.372) [0.034] | −4.2913 (1.3078) [0.000] |
INT | 2.0636 (0.7655) [0.002] | 1.9816 (0.3610) [0.000] |
LVA | −1.4388 (0.4539) [0.000] | −1.4212 (0.3137) [0.000] |
SARGAN TEST | 20.2317 [0.045] | 22.5459 [0.269] |
AR 2 | −1.4068 [0.4733] | −1.1509 [0.6670] |
NOTE; LY = log of real GDP per capita, PCB = principal component of banking sector development, PCS = principal component of stock market development, LTO = log of trade openness, LCAO = log of capital account openness, LVA = log of voice and accountability, INT = simultaneous opening of trade and capital accounts. N = 53. Values in ( ) and [ ] denotes standard errors and p-values respectively. ***, ** denotes significance at 1% and 5% respectively.
development. The result is similar to Law and Shah Habibullah [
Like trade openness, the expected impact of capital account openness on financial sector development and hence economic growth, is contentious. Supporters of capital account liberalization argue that it will lead to financial sector efficiency and development, while critiques of financial liberalization contest that it promotes financial instability and underdevelopment. The result buttresses the result of Diaz-Alejandro [
One finding that beats popular expectation is the effect of institutions on stock market development. The finding of a negative impact of institutions on stock market development does not appeal to common sense nor theoretical expectation. This however may be explained thus: better institutions may be associated with red-tapism and bureaucratic bottle-necks while weak institutions may speed up the process. Our finding is however contrary to Baltagi et al. [
Interestingly, the finding of uneven impact of financial liberalization on stock market development is not surprising. Developed countries may stand to gain from capital account opening while developing countries may suffer from financial liberalization. The reasons for these are that the stock market is the major source of financing businesses in developed countries. Again, stock markets in developed countries are characterized by almost perfect information, which means that firms will not take undue advantage of market information asymmetry to generate undue gains and create volatility in the market. Also, there is conducive playing ground for all market participants.
On the other hand, stock markets in developing countries are highly volatile and investors are usually after short term profit. Again, the markets are highly imperfect and there are limited mechanisms to check the activities and excesses such as insider trading etc of the market participants. Further, economic and political together with policy inconsistencies lead to stock market under development.
Lastly, the results do not find support for the Simultaneous openness hypothesis. The results in Tables 1-5 reject the Rajan and Zingales [
While the role of financial development in the economic growth process has been widely acknowledged, there is contention as to the factors affecting financial sector development. One area of empirical divergence among scholars is on the area of liberalization and more specifically, financial liberalization. This paper investigates the impact of openness and institutional quality on stock market development.
Using annual data from 1996-2013 averaged over 3 years period in a sample of 53 countries and employing Dynamic GMM estimators, the results show that economic growth, banking sector development, and the interaction term affect stock market development positively while capital account openness and institutional quality negatively impact on stock market development. The effect of trade openness is mixed. The results do not find evidence in support Rajan and Zingales’ proposition of simultaneous openness hypothesis [
Z. T.Abdallah, (2016) An Evaluation of Simultaneous Openness Hypothesis in the Context of Stock Market Development: Evidence from a Panel of Fifty Three Countries Based a GMM Study. Modern Economy,07,153-163. doi: 10.4236/me.2016.72017