In this paper we examine the explanatory power of R&D for the cross-section of stock returns in the Taiwan stock market for the period from 1996 to 2005. Previous finance theories argued that expected stock return and the risk of return should be positively related to R&D. We find that R&D intensity is helpful in explaining the expected stock returns on average, but the association is weak. We use R&D
intensity to run the regressions that also include size, book-to-market ratio, and market beta variables. The results of the regression indicate that average stock return is moderately positively related to R&D expenditure in the entire sample.
We further divide the entire sample into three subperiods according to the index of the Taiwan stock market and run the regressions for three subperiods respectively.
The subperiod analysis reveals more about what happened in the Taiwan stock market over time. The R&D effect is negative during the bubble-forming period (1996.01-2000.03), reflecting the speculative nature of the phenomenal price appreciation during that period. In the burst-of-bubble period (2000.04-2001.09), the stock returns are negative, but the R&D effect is slightly positive, indicating that firms with high R&D expenditure lost less value than those with low or zero R&D expenditure on average. The R&D effect is negatively correlated with the stock return in the post-bubble period (2001.10- 2006.12). Considering the nature of the lag and accumulated effects of the R&D activities, we conduct a similar analysis which replaces R&D intensity with cumulative R&D and reconfirm the earlier result.
The result shows that R&D activities have some long-term effects.
Finally, we also examine the relationship between the R&D intensity and the total risk. It is plausible that firms with high R&D expenditure should endure higher risk of stock returns. We find some evidence that the R&D intensity is positively associated with the total risk, though not for every subperiod.
Previous accounting and financial studies suggest several hypotheses to account for the impact of a firm’s R&D spending on performance. Among them, the profitability hypothesis is the most popular and acceptable one. The profitability hypothesis states that R&D expenditure represents investment opportunity - that is current R&D investment potentially reflects future cash flow. In particular, the increase in R&D expenditure implies the growth of investment opportunity, and so
investors tend to positively react to news of an increase in R&D. The results reported in this paper for the R&D effect in the Taiwan stock market are somewhat identical to the profitability hypothesis, though the explanatory power is low.
References
Aboody, David, and Baruch Lev., 2000. Information asymmetry, R&D and insider gains. Journal of Finance 55, 2747-2766.
Al-Horani, A., P.F. Pope, and A.W. Stark., 2003. Research and development activity and returns in the United Kingdom. European finance review 7, 27-46.
Barron, Orie E., Donal Byard, Charles Kile, and Edward J. Riedl., 2002.
High-technology intangibles and analysts’ forecasts. Journal of Accounting Research 40, 289-312.
Barth, Mary E., Ron Kasznik, and Maureen F. McNichols, 2001. Analyst coverage and intangibles assets. Journal of Accounting Research 39, 1-34.
Bublitz, Bruce, and Michael Ettredge., 1989. The information in discretionary outlays:
advertising, research, and development. Accounting Review 64, 108-124.
Chan, L., Y. Hamao, and J. Laconishok, 1991. Fundamental and stock returns in Japan.
Journal of Finance 46, 1739-1764.
Chan, L., J. Lakonishok, and T. Sougiannis, 2001. The stock market valuation of research and development expenditures. Journal of Finance 56, 2431-2456.
Chan, S.H., J. Martin, and J. Keensinger, 1990. Corporate research and development expenditures and share value. Journal of Financial Economics 26, 255-276.
Chiao, Chaoshin, and Weifeng. Hung, 2006. Stock market valuation of R&D and electronics firms during Taiwan’s recent economic transition. The Developing Economies 44, 53-78.
Cockburn, Iain, and Zvi Griliches, 1988. Industry effects and appropriability measures in the stock market’s valuation of R&D and patents. American Economic Review 78, 419-423.
Chui, C., and J. Wei, 1998. Book-to-market, firm size, and the turn-of-the-year effect:
evidence from Pacific-Basin emerging markets. Pacific-Basin Finance Journal 6,
275-293.
Daniel, K., S. Titman, and J. Wei, 2001. Explaining the cross-section of stock returns in Japan: factors or characteristics? Journal of Finance 56, 743-766.
Demers, Elizabeth., 2002. Discussion of high-technology intangibles and analysts’
forecasts. Journal of Accounting Research 40, 313-319.
Eberhart, Allan C., William F. Maxwell, and Akhtar R. Siddique, 2004. An examination of long-term abnormal stock returns and operating performance following R&D increases. Journal of Finance 47, 623-650.
Fama, E., and K. French, 1992. The cross-section in expected stock return. Journal of Finance 47, 427-466.
Fama, E., and J. MacBeth, 1973. Risk, return, and equilibrium: empirical tests.
Journal of Political Economy 81, 607-636.
Griliches, Z., 1981. Market value, R&D, and patents. Economics Letters 7, 183-187.
Hirschey, Mark, and Jerry J. Weygandt, 1985. Amortization policy for advertising and research and development expenditures. Journal of Accounting Research 23, 326-335.
Jagannathan, R., K. Kubota, and H. Takehara, 1998. Relationship between labor-income risk and average return: empirical evidence from the Japanese stock market. Journal of Business 71, 319-347.
Kothari, S.P., Ted E. Laguerre, and Andrew J. Leone, 2002. Capitalization versus expensing: evidence on the uncertainty of future earnings from capital expenditures versus R&D outlays. Review of Accounting Studies 7, 355-382.
Kubota, K., and H. Takehara, 1996. The cross-section risk and return of Tokyo Stock Exchange firms. In: Boss, T., Fetherstone, T. (Eds.), Advances in the Pacific Basin Capital Market Research II, vol. 3. JAI, Greenwich, CT, 273-305.
Kubota, K., and H. Takehara, 1997. Common risk factors in Tokyo Stock Exchange
firms: in finding the mimicking portfolio. In: Boss, T., Fetherstone, T. (Eds.), Advances in the Pacific Basin Capital Market Research II, vol. 3. JAI, Greenwich, CT, 273-302.
Lakonishok, J., A. Shleifer, and R.W. Vishny, 1994. Contrarian investment, extrapolation, and risk. Journal of Finance 49, 1541-1578.
Pakes, A., 1985. On patents, R&D, and the stock market rate of return. Journal of Political Economy 93, 390-409.
Rapoport, J., 1971. The anatomy of the product-innovation process: cost and time. In:
Mansfield, E. (Ed.), Research and innovation in the modern corporation. Norton, New York, 110-135.
Szewczyk, Samuel, George Tsetsekos, and Zather Zantout, 1996. The valuation of corporate R&D expenditures: evidence from investment opportunities and free cash flow. Financial Management 25, 105-110.
Tsai, Kuen-Hung, and Jiann-Chyuan Wang, 2002. Chung-Hua Institution for Economic Research.
Wagner, L.U., 1968. Problems in estimating research and development investment and stock. Proceedings of the business and economic statistics section. American Statistical Association, Washington, DC, 189-198.
Xu, Ming, and Chu Zhang, 2003. The explanatory power of R&D for the cross-section of stock returns: Japan 1985-2000. Pacific-Basin Finance Journal 12, 245-269.
Table 1 Summary Report of R&D
This table reports the data availability of the Taiwan stock market and the statistics of the R&D intensity at the aggregate level during the whole sample period from 1996 to 2005. “#
of samples” is the number of sample in the year with available market data such as the monthly return and market value. RD/A and RD/BE are the measures of the R&D intensity normalized by total assets and by book value of equity, respectively.
Data Availability
Statistics of the R&D Intensity at Aggregate Level
# of Samples with RD/A*10 RD/BE*10
Samples Positive R&D Value- Equally Value- Equally Number Percent Weighted Weighted Weighted Weighted 1996 2,380 1,488 62.52% 0.02477 0.03307 0.04285 0.05331 1997 2,844 1,542 54.22% 0.02636 0.04286 0.04495 0.07055 1998 3,244 1,831 56.44% 0.02589 0.04205 0.04229 0.06928 1999 3,764 2,257 59.96% 0.02803 0.04072 0.04534 0.06728 2000 4,359 2,754 63.18% 0.03637 0.03618 0.05749 0.05999 2001 4,970 3,315 66.70% 0.03890 0.03542 0.06308 0.05954 2002 5,646 3,932 69.64% 0.03988 0.03939 0.06630 0.07040 2003 6,156 4,418 71.77% 0.04085 0.03758 0.06596 0.07143 2004 6,337 4,648 73.35% 0.04023 0.03755 0.06498 0.06428 2005 6,388 4,751 74.37% 0.04226 0.03738 0.07060 0.06080 Average 4,609 3,094 65.22% 0.03435 0.03822 0.05638 0.06469
Table 2 R&D Intensities of the Electronics and the Non-Electronics Industries
This table lists the R&D intensities of the electronics industry and non-electronics industry during the entire sample period according to the measures of RD/A and RD/ME, respectively.RD/A*10 RD/ME*10
Electronics Non-Electronics Electronics Non-Electronics
Industry Industry Industry Industry
1996 0.07258 0.01469 0.12929 0.02640 1997 0.07059 0.01214 0.12256 0.02072 1998 0.06438 0.01148 0.11142 0.01923 1999 0.06578 0.01347 0.11207 0.02222 2000 0.06364 0.01394 0.10747 0.02305 2001 0.06689 0.01429 0.11122 0.02348 2002 0.07138 0.01434 0.11874 0.02336 2003 0.07755 0.01484 0.13324 0.02486 2004 0.07686 0.01536 0.13622 0.02629 2005 0.08408 0.01644 0.14422 0.02868
Table 3 Industrial R&D Intensities in Year 2005
This table reports the R&D intensity for most of the industries in 2005.Code Industry RD/A*10 RD/ME*10
1 Cement 0.0003 0.0005
2 Food 0.0062 0.0111
3 Plastics 0.0105 0.0162
4 Textiles 0.0113 0.0212
5 Electric Machinery 0.0466 0.0928
6 Electric & Cable 0.0071 0.0138
7 Chemical, Biotech, Medical Care 0.0366 0.0534
8 Glass Ceramics 0.0129 0.0225
9 Paper & Pulp 0.0034 0.0054
10 Steel & Iron 0.0024 0.0036
11 Rubber 0.0125 0.0187
12 Automobile 0.0614 0.0907
13 Electronics 0.0841 0.1442
14 Building Material &Construction 0.0001 0.0001 15 Shipping & Transportation 0.0000 0.0000
16 Tourism 0.0000 0.0000
18 Trading & Consumers Goods 0.0000 0.0000
20 Others 0.0213 0.0377
Table 4 Sample Characteristics
This table presents the descriptive statistics for the variables used later. We report the statistics for two categories separately: one includes firms with no R&D expenditure and the other includes firms with positive R&D expenditure. The entire sample period is divided into three subperiods.
Variable No R&D firms Positive R&D firms
Mean S.D. Mean S.D.
Table 5 Correlation Analysis
This table presents the unconditional correlations between the variables used later. Return average is the cross-sectional average of the stock returns. Ln(ME) and ln(BE/ME) are the cross-sectional average of size and book-to-market ratio, respectively. β is the market beta, estimated from the most recent 36 months of return data for each month. RD/A and RD/BE are the measures of the R&D intensity normalized by total assets and by book value of equity, respectively.
Return average (%) 1 -0.0394 0.2591 0.0312 -0.1036 -0.1047
ln(ME) 1 -0.0588 0.3982 -0.0988 -0.0651
ln(BE/ME) 1 0.1132 0.2868 0.2841
Table 6 Regression of Returns on R&D Relative to Total Assets
This table reports the coefficients of Fama-MacBeth regression of returns on the firm-specific variables,
, 1 2ln( ), 3ln( / ), 4 , 5 , 1 ,
i t i t i t i t i t i t
r =θ +θ ME +θ BE ME +θ β +θRD − +ε , where ln(ME) and ln(BE/ME) are respectively the size and book-to-market of a firm, β is the market beta, RD is the R&D intensity normalized by total assets. Numbers in parentheses are t-statistics.
Intercept ln(ME) ln(BE/ME) RD/A
(continued)
*, **, and *** indicate significance at the 0.1, 0.05, and 0.01 levels, respectively.
Table 7 Regression of Returns on R&D to Total Assets: the Positive R&D Firms
This table reports the coefficients of Fama-MacBeth regression of returns on the firm-specific variables, ri t, =θ1+θ2ln(ME)i t, +θ3ln(BE ME/ )i t, +θ β4 i t, +θ5RDi t,−1+εi t, , where ln(ME) and ln(BE/ME) are respectively the size and book-to-market of a firm, β is the market beta, RD is the R&D intensity normalized by total assets. Numbers in parentheses are t-statistics.Intercept ln(ME) ln(BE/ME) RD/A
*, **, and *** indicate significance at the 0.1, 0.05, and 0.01 levels, respectively.
Table 8 Regression of Returns on R&D to Total Assets for Electronics Industry
This table reports the coefficients of Fama-MacBeth regression of returns on the firm-specific variables for electronics industry only, ri t, =θ1+θ2ln(ME)i t, +θ3ln(BE ME/ )i t, +θ β4 i t, +θ5RDi t,−1+εi t, , where ln(ME) and ln(BE/ME) are respectively the size and book-to-market of a firm, β is the market beta, RD is the R&D intensity normalized by total assets. Numbers in parentheses are t-statistics.
Intercept ln(ME) ln(BE/ME) RD
0.0408 -0.0033 0.0446 0.0008 -0.9287 0.0310 (2.53) ** (-2.92) *** (22.79)*** (0.12) (-4.17)***
(continued)
0.0406 -0.0053 0.0424 0.035 0.0298
(0.87) (-1.74)* (9.08)*** (3.51)***
0.1258 -0.0122 0.049 1.9867 0.0268
(2.21)** (-3.38)*** (7.55)*** (2.71)***
0.1252 -0.0108 0.0491 0.0202 1.9322 0.0270
(2.20)** (-2.70)*** (7.55)*** (0.76) (2.62)***
0.0006 -0.0004 0.0428 0.0003 0.0289
(0.04) (-0.38) (20.18)*** (0.22)
0.0162 -0.001 0.043 -0.8921 0.0281
(0.96) (-0.93) (18.50)*** (-4.84)***
0.0163 -0.0021 0.0442 0.0156 -0.8897 0.0286
(0.97) (-1.80)* (18.60)*** (2.43)** (-4.83)***
*, **, and *** indicate significance at the 0.1, 0.05, and 0.01 levels, respectively.
Table 9 Regression of Returns on the Cumulative R&D Relative to Total Assets
This table reports the coefficients of Fama-MacBeth regression of returns on the firm-specific variables,
, 1 2ln( ), 3ln( / ), 4 , 5 , 1 ,
i t i t i t i t i t i t
r =θ +θ ME +θ BE ME +θ β +θCRD − +ε , where ln(ME) and ln(BE/ME) are respectively the size and book-to-market of a firm, β is the market beta, CRD/A is the cumulative R&D intensity normalized by total assets. Specifically, the cumulative R&D intensity is calculated as follows: , 0.4 , 0.3 , 1 0.3 ,2
*, **, and *** indicate significance at the 0.1, 0.05, and 0.01 levels, respectively.
Table 10 Regression of Total Risk on R&D to Total Assets
This table reports the coefficients of Fama-MacBeth regression of the standard deviationσ of returns i t, on the firm-specific variables,σi t, =γ1+γ2ln(ME)i t, +γ3ln(BE ME/ )i t, +γ4RDi t,−1+εi t, , where ln(ME) and ln(BE/ME) are respectively the size and book-to-market ratio of a firm. Numbers in parentheses are t-statistics.
(continued)
ln(ME) ln(BE/ME) RD/A
Panel C: Burst-of-Bubble Period (2000.04-2001.09)
0.2296 -0.0022 0.0007
(16.14)*** (-2.40) **
0.1981 -0.0155 0.0214
(185.59)*** (-13.81)***
0.1914 1.0646 0.0028
(130.40)*** (4.31)***
0.2455 -0.0030 -0.0158 0.0226
(17.40)*** (-3.37) *** (-14.01)***
0.1675 0.0011 -0.0349 3.9126 0.0719 (10.01)*** (1.02) (-22.14)*** (14.40)***
Panel D: Post-Bubble Period (2001.10-2005.12)
0.1723 -0.0029 0.0021
(26.79)*** (-6.96) ***
0.1364 -0.0420 0.1637
(293.07)*** (-67.21)***
0.1313 -1.0810 0.0080
(205.62)*** (-12.52)***
0.1882 -0.0033 -0.0421 0.1666
(32.00)*** (-8.85) *** (-67.47)***
0.1607 -0.0018 -0.0501 0.97063 0.1970 (24.86)*** (-4.44) *** (-67.07)*** (11.55)***
*, **, and *** indicate significance at the 0.1, 0.05, and 0.01 levels, respectively.