Mar 19, 2006 - relationship between R&D spending and stock markets, Chan et al. .... Khanna, Madhu, Wilma Rose H. Quimio, and Dora Bojilova, (1998) ...
An Empirical Study on the Impact of Environmentally Friendly News on Stock Prices in Japan Shohei Nagayama1 and Fumiko Takeda2 March 19, 2006 To be presented at Asia-Pacific Economic Association 2006 Annual Meeting This paper investigates how stock prices respond to announcements of corporate environmental measures by using a standard event study methodology. Examining 862 cases of Japanese listed companies from December 15, 1996 to December 15, 2004, we find that news on environmental research and development (R&D) had strongly positive impacts on stock prices. In particular, environmental R&D announcements of high-technology industries (chemicals, electronics, and automobiles) affected stock markets more positively than did those of non-high-technology industries. In contrast, the announcements of investment based on environmental accounting tended to decrease stock prices. Keywords: Environmental management, Environmental accounting, Research and development, Event study, Valuation, Corporate social responsibility JEL Classifications: L60, M14, Q29
1
Department of Technology Management for Innovation, University of Tokyo.
2
Corresponding author.
Department of Technology Management for Innovation, University of
Tokyo. 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-8656 Japan. Tel & Fax: +81-3-5841-1191. takeda＠tmi.t.u-tokyo.ac.jp. 1
E-mail:
1. Introduction There has been a long debate on whether the environmental performance of firms can have positive influence on stock markets. Traditional views are based on recognizing a tradeoff between social benefits and private costs. That is, environmental protection favors the society but imposes a significant economic burden on private firms. However, recently, several economists have argued that firms engage in “profit-maximizing” environmental management, expecting benefits that exceed additional costs, by generating innovation,3 raising their reputation,4 charging higher prices for their products, recruiting highly skilled workers, and so on. In Japan, more and more companies have attempted to engage in environmentally friendly activities after the agreement of the Kyoto Protocol. According to the Ministry of the Environment, the proportion of companies that have introduced an environmental accounting system has increased from 13.2% in FY 2000 to 28.2% in FY 2004. In addition, the percentage of companies that publicly disclose the data and activities related to environmental friendliness and other corporate social responsibility (CSR) has risen from 21.3% in FY 1999 to 47.4% in FY 2004. A number of recent empirical studies have attempted to show the relationship between environmental and financial performance by using an event study methodology.5 Most of these studies have reported that stock markets react positively to environmentally friendly news and negatively to environmentally unfriendly news. A first strand of studies examines the impact of specific pollution news on stock prices (Blacconiere and Patten, 1994; Hamilton, 1995; Konar and Cohen, 1997; Khanna et al., 1998). Blacconiere and Patten (1994) analyzed the effect of the 1986 chemical leak at 3
For example, Porter and van der Linde (1995) claimed that in a dynamic model well-designed environmental regulations could generate innovation that might offset costs incurred by them. 4 A number of research papers on corporate social responsibility (CSR) have supported the idea that CSR positively affects consumers’ brand and product evaluations above economic considerations (Maignan, 2001; Mohr et al., 2001). For instance, Maignan (2001) conducted a consumer survey and found that U.S. consumers were highly conscious about firms’ economic responsibilities, while French and German consumers placed a high value on corporate legal and ethical standards. In addition, several studies have shown that CSR has a positive influence on even unrelated product evaluations (Klein and Dawar, 2003). For example, Klein and Dawar (2003) showed that attributions influenced by CSR reduced the negative impact of public announcements of defective and/or dangerous products. 5 For a detailed survey of the relevant literature from 1993 to 2001, see, for example, Koehler (2003). 2
Union Carbide’s plant, showing that the event pulled down not only its stock price but also that of 47 other chemical firms. Hamilton (1995) and Konar and Cohen (1997) provided evidence that stock markets reacted negatively to the first edition of Toxic Release Inventory (TRI) data released by the Economic Planning Agency (EPA) in June 1989. Khanna et al. (1998) also found that repeated annual release of TRI data negatively affected stock prices during the years 1990-1994. A second set of studies investigates stock market responses to the announcements of firm-specific environmental news (Wingender and Woodroof, 1997; Yamashita et al., 1999; Dasgupta and Laplante, 2001; Filbeck and Gorman, 2004; Grand and D’Ella, 2005; Gupta and Goldar, 2005). Wingender and Woodroof (1997) found a positive correlation between announcements of energy management projects and stock prices. Yamashita et al. (1999) examined the relationship between environmental conscientiousness (EC) scores6 and stock returns in the U.S. Their results indicate that only a weak increase was found for firms with high EC scores. Filbeck and Gorman (2004) more comprehensively analyzed the stock price reaction of companies listed in the S&P500 to ten types of news related to environmentally friendly policies or actions, among which the impact of receiving an environmental award was significantly positive. Dasgupta and Laplane (2001), Grand and D’Ella (2005), and Gupta and Goldar (2005) showed that stock prices of developing countries also tended to drop after the release of environmentally unfriendly news.7 Our study is in line with the second strand of studies. We investigate how stock prices respond to announcements of corporate environmental measures by using a standard event study methodology. Compared to most of the prior studies that focus on a particular kind of news or ratings, we search for more varied types of environment-related measures announced by firms and concentrate on the analysis of 6
Yamashita et al. (1999) used EC scores published in Fortune magazine (1993). Using the stock market data of Argentina, Chile, Mexico, and the Philippines between 1990 and 1994, Dasgupta and Laplane (2001) showed that news of citizen’s complaints tended to decrease stock prices, while the announcements of superior environmental performance increased them. Grand and D’Ella (2005) focused on Argentina and found that its companies’ stock prices dropped after the release of negative news, whereas they did not react significantly to positive news. Gupta and Goldar (2005) provided evidence from India that the announcements of poor environmental performance based on the “green leaf rating,” which is the environmental rating evaluated by the Center for Science and Environment, India’s environmental NGO, generally decreased stock prices of pulp and paper, automobile, and chlor alkali firms. 7
3
differences in stock market reactions to these announcements. Examining 862 cases of Japanese listed companies from December 15, 1996 to December 15, 2004, we find that environmental research and development news had strongly positive impacts on stock prices. In contrast, the announcements of investment based on environmental accounting tend to decrease stock prices. The rest of this paper is organized as follows. Section 2 describes the data. Section 3 explains the event study methodology. Section 4 provides the results and discussions. Concluding remarks are offered in Section 5. 2. Data Using Nikkei Telecom 21, a search for announcements of environmental measures (kankyo-taisaku in Japanese) from December 15, 1996 to December 15, 2004 results in 862 samples from 271 firms listed in the Tokyo Stock Exchange. Figure 1 shows that the amount of sample data has increased significantly since the beginning of 2000. Daily stock returns of individual firms and the index TOPIX are computed as follows, using Toyo Keizai’s Kabuka CD-ROM 2005: Rit =
T − Tt −1 Pit − Pit −1 , and Rmt = t , Tt −1 Pit −1
where Pit refers to the stock price of the ith firm at time t , Rit is its rate of return, Tt presents TOPIX at time t , and Rmt is its rate of return. We divide the announcements of environmental measures into four groups, as follows: (1) environmentally related research and development (R&D), (2) statements of costs and benefits expected from environmental measures based on environmental accounting, (3) new establishment or exploration of an environmentally related business, and (4) other environmental measures. The last category, other environmental measures, includes: (i) reduction of CO2 or NOx emissions, (ii) elimination or reduction of toxic substances, (iii) introduction of “green procurement,” (iv) recycling raw materials, (v) waste management, and so on. Among these measures, “green procurement” can be defined as the selection of products and services that have a lesser hazardous effect on human health and the environment. This procurement may cover the purchase of raw material, parts, supplies, equipment, and/or services. Recently, major Japanese companies such as SONY, Canon, NEC have disclosed information on the use of hazardous substances and have required their suppliers to ban or reduce their
4
use. 3. Methodology The evaluation is performed using the standard event study methodology as described by MacKinlay (1997). This methodology is based on the assumption that markets are efficient, in the sense that current stock prices reflect all publicly available information. In other words, only unexpected events would deliver new information about the future profitability of a firm and change stock prices. Here, the event is defined as the date when the firm announces its environmental measures. We first choose the event window, which is the period over which stock prices react to the event. We define the event day as t = t0 , the initial date of the event window as t = t1 , and the final date of the event window as t = t2 . Considering that investors can often receive information by online news providers prior to newspaper publication, we choose a 4-day window setting, in which t1 = −1 and t2 = +2 . In comparison, we then examine the longer-term effect of the event by using a 21-day event window with t1 = −10 and t2 = +10 . The following market model is estimated for each announcement: Rit = α i + β i Rmt + γ i dit + ε it , where dit is a dummy variable, which takes 1 at the event window and 0 otherwise, and ε it is the zero mean disturbance term. The estimation window is set at 140 transaction days prior to the event window.
By estimating the equation above, we
obtain the estimated parameters αˆ i , βˆi , and γˆi . The abnormal return for the stock of firm i in period t is defined as: ARit = Rit − (αˆ i + βˆi Rmt ) .
The cumulative abnormal return (CAR) is given by summing up abnormal returns over the event window, as follows: t2
CARi (t1 , t2 ) = ∑ ARit . t = t1
Since we assume the null hypothesis H0 that each event does not affect the mean or variance of returns, the CAR is normally distributed with a mean 0 and variance σ i2 (t1 , t2 ) .
2
The average cumulative abnormal return ( CAR ) and its variance σ i (t1 , t2 ) are
5
calculated by averaging the CAR and its variance σ i2 (t1 , t2 ) across N firms in the same category, as follows: N
CAR(t1 , t2 ) = 1 N ∑ CARi (t1 , t2 ), i =1
N
VAR ⎡⎣CAR(t1 , t2 ) ⎤⎦ = σ (t1 , t2 ) = 1 N 2 ∑ σ i2 (t1 , t2 ). 2
i =1
Based on the null hypothesis H0, CAR is normally distributed with a mean zero and 2
variance σ (t1 , t2 ) . We can test H0 using J=
CAR(t1 , t2 ) 2
σ (t1 , t2 )
∼ N (0,1) .
4. Discussion Table 1 presents the CARs and the statistical significance for each event. Figure 2 shows a trend of the CARs based on the 21-day window. Both Table 1 and Figure 2 indicate that news on environmental measures on the whole does not affect stock returns of firms, partly because the whole sample includes both positive and negative events, offsetting the impact of each other. Among four groups, the announcements of environmentally related R&D have a positive influence at 1% significance level on stock prices. In contrast, news about environmental accounting tends to decrease stock prices. The announcements of the new establishment and exploration of environmental businesses do not seem to affect stock markets in a 4-day window and bring about a price decline in a 21-day window. Among other environmental measures, news on waste management and green procurement has positive impacts on stock prices in a 21-day window, while news on reduction of toxic substances has a negative influence. The fact that the announcements of environmental R&D positively affect stock markets is consistent not only with the findings of Porter and van de Linde (1995), who argued that environmental innovation, which is accelerated by environmental regulations, could offset higher costs incurred by such innovation, but also with prior studies on the stock price response to R&D spending in general (Chan et al., 1990; Eberhart et al., 2004; Xu and Zhang, 2004). Among many existing papers on the relationship between R&D spending and stock markets, Chan et al. (1990) found that increases in R&D expenditures announced by high-technology firms tended to raise
6
stock prices in the U. S. from June 1979 to June 1985, while announcements by low-technology firms had negative impacts. Like Chan et al. (1990), we next divide our 241 samples of news on environmental R&D into those announced by high-technology industries and those announced by non-high-technology industries and examine the difference in stock market reaction to the news. We employ the conventional definition that high-technology industry has a higher ratio of R&D spending to sales than do other industries. Based on this definition and the statistics published by the Ministry of Economics, Trade, and Industry, we select chemicals (including medicine), electronics, and automobiles as representative high-technology industries.8 Table 2 presents the CARs of high- and non-high-technology industries, and Figure 3 shows the trend of their CARs for a 21-day window. The table and the figure both indicate that the CARs of high-technology industries are much higher than those of non-high-technology industries. This result is consistent with those of Chan et al. (1990). Among high-technology industries, the electronics industry has the strongest impact on stock markets compared to chemicals and automobiles. That is, investors expect higher profits from environmentally related R&D for high-technology industries, especially the electronics industry, than from non-high-technology industries. We find that the news of costs and benefits expected from environmental measures based on environmental accounting has a significantly negative influence on stock markets when considering both 4-day and 21-day windows. This may due to the fact that in many cases, costs of investment exceed benefits, when firms evaluate them by environmental accounting. Among 163 events of environmental accounting, 96 events contain information on both costs and benefits, of which the only 8 events report benefits over costs, while the remaining 88 events announce the costs exceed benefits. In the third category of announcements, that is, those about the new establishment and exploration of environmentally related business, the effects of the news of such establishment or exploration are not statistically significant in a 4-day window, but they become significantly negative in a 21-day window. Like the announcement of investment based on environmental accounting, investors may not regard these activities as profitable at least in the short run. In particular, the profitability of expanding an environmental business overseas may be too uncertain to be welcomed by investors. According to the Ministry of Economics, Trade, and Industry, the ratio of R&D expenditures to sales is 6.32%, 7.35%, and 6.04% for the chemicals, electronics, and automobiles industries, respectively, in FY 2000. These figures are much higher than that of all industries (3.75%) and manufacturing on average (4.36%). 8
7
Other environmental measures on the whole do not affect stock markets in a 4-day window, but they have a negative influence in a 21-day window, although these measures may enable firms to avoid the possibly increased costs imposed by tightening environmental regulations. The exceptions are announcements about waste management and green procurement, which affect stock markets positively and significantly in a 21-day window. As we see in Section 2, green procurement does not impose additional costs on firms employing this measure, as it occurs in the form of requests to suppliers to make environmentally friendly efforts. To conclude this section, examining 862 events in Japan from December 15, 1996 to December 15, 2004, we found that investors were likely to appreciate news on environmentally related R&D positively, but negatively to announcements about costs and benefits based on environmental accounting, the new establishment and exploration of environmental businesses, and other environmental measures except for waste management and green procurement. Based on these results, we conclude that Japanese investors seem to believe that environmental R&D can enhance firms’ innovation and competitiveness, while the effects of many other measures are still uncertain or evaluated negatively in terms of profitability. In previous studies on stock price response to environmentally related news in the U.S. and other countries, investors tended to punish the companies that were environmentally unfriendly or appreciate the firms that were awarded for their environmental considerations, but this trend is not yet seen to the same degree in Japan. 5. Concluding remarks This paper investigated how stock prices responded to announcements of corporate environmental measures by using a standard event study methodology. Examining 862 cases of Japanese listed companies from December 15, 1996 to December 15, 2004, we found that environmental research and development plans had strongly positive impacts on stock prices. In particular, environmental R&D announcements of high-technology industries (chemicals, electronics, and automobiles) affected stock markets more positively than did those of non-high-technology industries. This result is consistent with the study of Chan et al. (1990). In contrast, the announcements of investment based on environmental accounting tended to decrease stock prices. This may due to the fact that in many cases, costs of investment exceeded benefits, when firms evaluated these factors using environmental accounting. The effect of announcements of the new establishment and exploration of
8
environmental business was significantly negative, possibly due to a high level of investor uncertainty. Among other environmental measures, news on waste management and green procurement positively affected stock markets, probably because of the low costs of implementation. These results indicate that so far, Japanese investors seem to behave more profit-consciously than environment-consciously. However, in the future, as the entire society becomes more sensitive to environmental activities of firms, like investors in other advanced countries, Japanese investors may also come to evaluate more positively firms’ environmental activities.
9
References Blacconiere, Walter G., and Dennis M. Patten, (1994) “Environmental disclosures, regulatory costs, and changes in firm value.” Journal of Accounting and Economics 18, 357-377. Bosch, J. C., E. Woodrow Eckard, and Insup Lee, (1998) “EPA Enforcement, Firm Response Strategies, and Stockholder Wealth: An Empirical Examination.” Managerial and Decision Economics 19, 167-177. Chan, Su Han, John D. Martin, and John W. Kensinger, (1990) “Corporate research and development expenditures and share value.” Journal of Financial Economics 26, 255-276. Dasgupta, Susmita, and Benoit Laplante, (2001) “Pollution and Capital Markets in Developing Countries” Journal of Environmental Economics and Management 42, 310-335. 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, Vol. LIX, No. 2, 623-650. Filbeck, Greg, and Raymond, F. Gorman, (2004) “The Stock Price Reaction to Environmentally-Related Company News.” Journal of Business and Public Affairs, Fall 2004, 25-31. Grand, Mariana Conte, and Vanesa V. D'Elia, (2005) "Environmental news and stock markets performance: Further evidence for Argentina," CEMA Working Papers 300, Universidad del CEMA. Gupta, Shreekan, and Bishwanath Goldar, (2005) “Do stock markets penalize environment-unfriendly behavior?: Evidence from India.” Ecological Economics 52, 81-95. Hamilton, James T., (1995) “Pollution as News: Media and Stock Market Reactions to the Toxics Release Inventory Data” Journal of Environmental Economics and
10
Management 28, 98-113. Khanna, Madhu, Wilma Rose H. Quimio, and Dora Bojilova, (1998) “Toxics Release Information: A Policy Tool for Environmental Protection.” Journal of Environmental Economics and Management 36, 243-266. Klein, Jill and Niraj Dawar, (2003) “Corporate social responsibility and consumers’ attributions and brand evaluations in a product-harm crisis” International Journal of Research in Marketing 21, 203-217. Koehler, Dinah A., (2003) "Capital Markets and Corporate Environmental Performance - Research in the United States", Fall 2003. Working Paper 03-36-DK, Wharton Risk Management and Decision Process Center. Konar, Shameek, and Mark A. Cohen, (1997) “Information As Regulation: The Effect of Community Right to Know Laws on Toxic Emissions.” Journal of Environmental Economics and Management 32, 109-124. MacKinlay, A. C. (1997) “Event studies in economics and finance.” Journal of Economic Literature 35, 13-39. Maignan, Isabelle, (2001) “Consumers’ Perceptions of Corporate Social Responsibilities: A Cross-Cultural Comparison.” Journal of Business Ethics 30, 57-72. Mohr, Lois A., Deborah J. Webb, and Katherine E. Harris, (2001) “Do Consumers Expect Companies to be Socially Responsible?: The Impact of Corporate Social Responsibilty on Buying Behavior.” The Journal of Consumer Affairs, Vol. 35, No. 1, 45-72. Porter, Michael E., and Claas van de Linde, (1995) “Toward a New Conception of the Environment-Competitiveness Relationship.” Journal of Economic Perspectives Vol. 9, No. 4, 97-118. Wingender, John R., and Eric A. Woodroof, (1997) “When Firms Publicize Energy Management Projects Their Stock Prices Go Up.” Strategic Planning for Energy and the Environment, Vol. 17 (1), 38-51.
11
Xu, Ming, and Chu Zhang, (2004) “The explanatory power of R&D for the cross-section of stock returns: Japan 1985-2000.” Pacific-Basin Finance Journal 12, 245-269. Yamashita, Miwaka, Swapan Sen, and Mark C. Roberts, (1999) “The Rewards for Environmental Conscientiousness in the U. S. Capital Markets.” Journal of Financial and Strategic Decisions, Vol. 12, No. 1, 73-82.
12
Figure 1: Number of the announcements of corporate environmental measures
180 Number of events
160 140 120 100 80 60 40 20
13
04 20 ∼
∼
20
03
/1 2
/1 2
/1 2
/1
/1
5
5
5 /1
5 02 20 ∼
∼
20 ∼
20
00
01
/1 2
/1 2
/1
/1
5
5 /1 /1 2 99
19 ∼
19 ∼
∼
19
97
98
/1 2
/1 2
/1
/1
5
5
0
Figure 2: Cumulative abnormal returns based on a 21-day window
2.0% Environmental R&D 1.5% 1.0% 0.5%
All sample
0.0% Others
-0.5% -1.0%
Establishment and exploration Environmental accounting
-1.5% -10
-8
-6
-4
-2
0
14
2
4
6
8
10
Figure 3: Cumulative abnormal returns of high- and non-high-technology industries
2.5%
High-technology industries
2.0% All indutries
1.5% 1.0%
Low-technology industries 0.5% 0.0% -0.5% -10
-8
-6
-4
-2
0
2
4
6
8
10
5.0% Electronics 4.0% 3.0%
Automobiles All industries
2.0% 1.0% 0.0% Chemicals -1.0% -2.0% -10
-8
-6
-4
-2
15
0
2
4
6
8
10
Table 1: Cumulative abnormal returns
Announcement type Total
Number of observations 862
4-day window CAR (J-statistic) 0.119% (1.55)
(1) Environmental R&D
241
0.859%
(2) Environmental accounting
163
-0.437%
41
0.337%
417
-0.112%
(3) Establishment and exploration (4) Other measures
(5.77) ***
21-day window CAR (J-statistic) -0.051% -(0.67) 1.323%
(8.90) ***
-1.382%
-(7.24) ***
(1.10)
-1.134%
-(3.70) ***
-(1.04)
-0.219%
-(2.04) **
-(2.29) **
(i) Reduction of emissions 140 -0.051% -(0.29) 0.012% (ii) Reduction of toxic substance 79 -0.062% -(0.24) -1.307% (iii) Recycling raw materials 73 -0.313% -(1.23) 0.217% (iv) Waste management 57 0.167% (0.58) 0.805% (v) Green procurement 29 0.570% (1.29) 1.185% (vi) Others 39 -0.968% -(2.68) *** -2.206% Note: *** and ** indicate 1% and 5% statistically significance level, respectively.
(0.07) -(5.02) (0.86) (2.80) (2.68) -(6.10)
*** *** ** ***
Table 2: Cumulative abnormal returns of high- and low- technology industries
Indutry
4-day window 21-day window Number of J-statistic J-statistic observations CAR CAR High-technology 130 1.41% 6.834 *** 1.99% 9.639 *** Chemicals 26 0.99% 2.017 ** -0.07% -0.141 Electronics 46 2.87% 8.827 *** 3.57% 10.963 *** Automobiles 58 0.44% 1.385 1.65% 5.267 *** 111 0.21% 0.999 0.55% 2.547 ** Non-high-technology Note: ***, **, and * indicate 1%, 5%, and 10% significance level, respectively.
16
