model at the same time. We take them as substitutes to avoid the collinearity problem.
Since the World Competitiveness Center editing the national competitiveness rank takes the real GDP growth as a hard data criterion, and hard criteria represent a weight of 2/3 in the overall ranking, we know that the economy performance correlates with the national competitiveness rank significantly. As what we expected, there exists significant correlation between the real GDP growth and the change of national competitiveness. As a result, we prefer not to take these two variables into the model at the same time to avoid the insignificant problem of beta caused by collinearity.
Overall, there still exist some correlation between the variables, but because the correlation value is not as high as we mention previously, and we do not think they will cause the problem to this study if we do not take them into account.
4.3 Empirical result
In order to decide which model has better explanation, we do the Hausman test first. And we find that in model (1) , the Hausman test suggests us using the random effect model. On the other hand, in the other models, through Hausman test, the evidence of fixed effect exists so we decide to use the fixed effect model for model (2) to model (8).
Table 4 illustrates the research results for all the models we set. In all the models, we set country stock return as the dependent variables. In the model (1) we put all the variables into the model, and according to table 4, we could find that the real GDP growth is significant, which represents that real GDP growth exactly has positive influence on the stock market return, and it also fits our expectation at first. And the capital formation is significantly positive to the stock return in all of the models. In
‧
this model we find other variables are not significant. We think this may result from the real GDP growth which is highly correlated with other variables. Last but not least, we do not find that the change of national competitiveness has significant effect on the stock market return in spite of the sign of the coefficient that fits our expectation.
As a result, there may exist some collinearity problems if we put them into the model at the same time. In the next models, we try to take the real GDP growth away, and try to solve the collinearity problem.
In the model (2), we take the real GDP growth variable away, and we find that this time the productivity has positive effect on the stock return significantly, which supports (CRR, 1986), that the industrial production is a source of common variation to the stock market return. And this is the most different part from the model (1).
After we take away the real GDP growth variable, the capital formation is still positive and significant to the stock market return.
In the model (3), we try to take both the capital formation and real GDP growth variables away at the same time. We find that the change of national competitiveness has a positive effect on the stock market return. When the national competitiveness rank improves, the stock market will have a better performance next year significantly.
This also fits our original expectation. On the other hand, the productivity still has positive effect on the stock market return significantly.
In the model (4), we take the real GDP growth and education variables away, and find that the capital formation and productivity are positive to the stock market return significantly. Other variables are not significant.
In the model (5), we take the real GDP growth and productivity variables away, and the only significant variable is the capital formation.
Finally, we also try to use the time effect and cross-sectional effect model for all the models we build, we use the dummies for each countries to test the time effect and
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
30
cross-sectional effect in the OLS model, but there are not obvious time effect or cross-sectional effect in our models.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
31
Table 4 Regression results for stock return and related variables
(1) (2) (3) (4) (5) (6) (7) (8)
stockreturn stockreturn stockreturn stockreturn stockreturn stockreturn stockreturn stockreturn
capital 0.677*** 1.061*** 1.031*** 1.218***
(3.15) (5.49) (5.61) (6.55)
GDPgrowth 0.0246*** 0.0722*** 0.0713*** 0.0681***
(3.84) (11.45) (12.21) (13.89)
gdp -0.00000709 -0.00000963 -0.00000916* -0.0000101 -0.0000103 1E-05 -1E-05 8.8E-06
(-0.92) (-1.24) (-2.10) (-1.28) (-1.32) (-0.27) (-0.38) (-0.23)
saving -0.0653 -0.0672 -0.0358 -0.0657 -0.0767 -0.0723 -0.0513 -0.0672
(-1.07) (-1.09) (-0.48) (-1.05) (-1.24) (-0.89) (-0.63) (-0.83)
education -0.111 -0.129 -0.287 -0.0371 -0.415*** -0.446**
(-0.79) (-0.90) (-1.91) (-0.27) (-2.80) (-3.09)
productivity 0.201 0.507*** 0.787*** 0.555*** -0.185 -0.228
(1.19) (3.36) (3.64) (3.84) (-1.11) (-1.48)
change_lead -0.0263 -0.0497 -0.0829*** -0.0575 -0.0500 0.013 0.0182 -0.0011
(-0.56) (-1.04) (-2.86) (-1.23) (-1.10) (-0.25) (-0.35) (-0.02)
_cons 0.0486 0.114*** 0.102 0.125*** 0.142*** -0.101** -0.0684* -0.0972*
(1.92) (6.04) (1.55) (7.10) (8.62) (-2.45) (-1.73) (-2.45)
R 0.2870 0.1958 0.1052 0.1688 0.1641 0.2654 0.2226 0.2716
N 528 528 529 590 539 529 592 540
t statistics in parentheses
* p < 0.1, ** p < 0.05, *** p < 0.01
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
32
In the model (6), (7), and (8), we find that the real GDP growth variables are all positive to the stock market return significantly. And the real GDP variable is extremely significant. We could proof that the economic growth is a factor, which really has significant influence on the stock market return. In the model (6) and (8) we also find the education variable is significant in model (6) and model (8), but the sign of coefficient is negative, and it implies that the education has a negative effect on the stock market return. But in all the other models, the education variables is not significant and also has a negative effect on the stock market returns.
We also try to use the lag period of education variables, but the result is not significant. Only when we use the education as a contemporaneous variable is the result significant, but the direction does not fit our expectation. It may be caused by the incomplete data we collect, that is, there exist some missing values of our data and make the result unexpected. On the other hand, the education is a long-term process.
It could not immediately response on the country. The education expenditure of a country may response on the stock market or on national competitiveness after several periods, that is, the education expenditure is a lag variable in the models. As a result, it seems not significant in our models.
We could find that the education expenditure will not reflect on the stock market return in short period, but we trust that the education expenditure is a long-term variable to reflect on the stock market return. Owing to the data period is only fourteen years, we do not have the enough evidence to prove the education expenditure will affect the stock market return. In the future, if there are other researcheres interested in this topic may collect much longer data to do further researches.
‧
GDP growth is a good variable to the stock market return and it is also a good index to predict the stock market return, and investors could use it to choose a good investment target in the future. But we still should take notice that the real GDP growth is a long term process. It is not a short period index. We should take it as a long term index to choose the stock markets, and this also fits our goal to earn the long term market return instead of only short term return caused by the event shocks or other unpredictable factors.Although only in the model (3) is the national competitiveness change significant to the stock market return, in other models the national competitiveness change still has the right direction to the stock market return, that is when a nation has improvement on its national competitiveness, it will bring a positive effect for the next few years. We could expect that raising the national competitiveness is good to the stock market. National competitiveness has a positive effect on the stock market return, and this offers a good reason for government to maintain the national competitiveness. This research tell us that the stock market could not reflect the short-term condition but the short-term condition, and this is also the reason why the national competitiveness could not reflect on the stock market return for stockholders.
Our research also provides a good index for investors to choose a good target stock market to invest. When the international investors or institution try to search good investment target, they could take the real GDP growth rate and the national competitiveness index into account to pick good stock market to invest. We also find that the intercept of our models are significant, this may imply there still exist some factors which we do not take into our model.
This research supports the classical Solow Growth Model, and is also consistent with Faug`ere, C. and J. V. Erlach (2006), Ibbotson and Chen (2003). We use nearly all stock market data of the world, and we prove that the real GDP growth will have a good impact on the stock market return. The real GDP growth is a good index to fit all the stock markets of the world, not only focusing on some developed or emerging
‧
data goes through many different economic time period, on the other hand, our data goes through the period which global economic grow rapidly. This may induce we have the different conclusions for the economic growth and stock return issue.This research also finds that the capital formation has positive effect on the stock market return, and the productivity is also an important factor, which will influence the stock market return. On the other hand, the education is not a factor which will influence the stock market return. It may be caused by the data collecting incompleteness or the fact that the education is a long term factor, and we may not see its influence in short period. The education expenditure does not reflect on the stock market return contemporaneously. It takes time to have an effect on the stock market.
All in all, we know that the economic growth is an important factor to the stock market return. The international investor and institution could have better investment return if they could find a country which has a good economic growth in the long term, such that, for more than 5 years or even 10 years.
Our data is mainly from World Competitiveness Center of IMD and DataStream.
These databases are almost having the detailed data and nearly nothing omitted. Some data is not completed, hard to collect, or the data period is too short to use. As a result, we delete data of some countries which are not long enough or have too many missing values. We try best to collect all the data we can find and make all the data keep its true value.
We should keep in mind that in the future, the world markets are opening continually, and this may induce the interact effect of each country, especially when the national competitiveness rank is a relative concept. We do not take the interact effect of each country into our study, and this may be a future research direction for researchers.
The national competitiveness statistical data is still not long enough, and there are nearly two hundred countries all around the world, the IMD statistical data may not be enough because it only includes complete data about fifty countries. The future research may try to use the other famous statistic institution, the World Economic Forum, which also does the rank about the national competitiveness. We do not take
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
35
the systematic risk factor into account in this research. The future research may try to put the systematic risk factor of each country into the model.
‧
Growth Options, and Security Returns" Journal of Finance 62, 171-192.2. Banz, Rolf W., 1981, "The relationship between return and market value of common stocks" Journal of Financial Economics 9, 3-18.
3. Barro, R., Sala-i-Martin, X., 1992, "Convergence" Journal of Political Economy 100, 223-251.
4. Barro, R., Sala-i-Martin, X., 1995, "Economic Growth" McGraw Hill, New York.
5. Begg, 1999, " Cities and Competitiveness" Urban Studies 36. 5/6 : 795-809
6. Berk J. B., R. C. Green, V. Naik, 1999, " Optimal Investment, Growth Options, and Security Returns", Journal of Finance54, Issue 5, 1553-1607
7. Bhandari, 1988, "Debt/Equity Ratio and Expected Common Stock Returns:
Empirical Evidence", Journal of Finance 43, No. 2, 507-528
8. Bodurtha, J., James N., D. C. Cho, 1989, "Economic forces and the stock market:
An international perspective" Global Finance Journal 1, 21-46.
9. Burton A. Weisbrod, 1962, "Education and Investment in Human Capital" Journal of Political Economy 70, No. 5, 106-123
10. Chen, N., R. Roll and S. Ross, 1986, "Economic Forces and the Stock Market."
Journal of Business 59, 383-403.
11. Clark J. , Guy K., 1998, " Innovation and competitiveness: a review", Technology Analysis & Strategic Management, Vol 10, issue 3, 363-395
12. Eugene F. Fama and Kenneth R. French, 1992, "The Cross-Section of Expected Stock Returns", The Journal of Finance, Vol. 47, No. 2, 427-465
13. Fama E. and French K., 1993, "Common risk factors in the returns on stocks and bonds" Journal of Financial Economics, 3-56.
14. Faug`ere, C., J. V. Erlach, 2006, "The Equity Premium: Consistent with GDP Growth and Portfolio Insurance" The Financial Review 41, 547-564.
15. Garcia V. F. and Liu L., 1999, "macroeconomic determinant of stock market development" Journal of Applied Economics, Vol. II, No. 1, 29-59
16. Georghiou L. G., J. S. Metcalfe, "Evaluation of the impact of European community research programmes upon industrial competitiveness", R&D Management Volume 23, Issue 2, 161-170
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
37
17. Levine, R., Renelt, D., 1992, "A sensitivity analysis of cross-country growth regressions" The American Economic Review 82, 942-963.
18. Lu Zhang, 2005, " The Value Premium", Journal of Finance, Volume 60, Issue 1, 67-103
19. Mankiw, G., Romer, D., Weil, D., 1992, "A contribution to the empirics of economic growth." Quarterly Journal of Economics 107, 407-437.
20. Moretti, E., 2004, "Workers' Education, Spillovers, and Productivity: Evidence from Plant-Level ProductionFunctions." The American Economic Review 94 (3), 656-690.
21. Myoung-jae Lee, 2002, "Panel Data Econometrics, Methods of Moments and Limited Dependent Variables."
22. Nonneman, W., Vanhoudt, P., 1996, "A further augmentation of the Solow model and the empirics of economic growth for OECD countries." Quarterly Journal of Economics 111, 943-953.
23. Porter M.E., 1990, " The competitive advantage of nations: with a new introduction"
24. Rapkin, DP and J, R. Strand, 1995, "Competitiveness: useful concept, political slogan or dangerous obsession", International Journal of Political Economy Yearbook, 8, 1-20.
25. Ritter, J. R., 2005, "Economic Growth and Equity Returns" Pacific-Basin Finance Journal 13, 489-503.
26. Roger G. Ibbotson and Peng Chen, "Long-Run Stock Returns: Participating in the Real Economy", Financial Analysts Journal ,Vol. 59, No. 1: 88-98
27. 陳冠位, 2002, "Study on the Evaluation System of City Competitiveness." ,博士 論文
28. 薛瀾, 1995, “從國際競爭力評估美國在全球競爭地位”，美歐月刊，pp.44-60。
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
38
Appendix
Sample Country List
Argentina Denmark Ireland Netherlands South Africa
Australia Finland Israel New Zealand Spain
Austria France Italy Norway Sweden
Belgium Germany Japan Peru Switzerland
Brazil Greece Jordan Philippines Taiwan
Canada Hong Kong Korea Poland Thailand
Chile Hungary Lithuania Portugal Turkey
China Mainland Iceland Luxembourg Russia United Kingdom
Colombia India Malaysia Slovak Republic USA
Czech Republic Indonesia Mexico Slovenia Venezuela