This article was downloaded by: [University of Arizona] On: 19 April 2015, At: 16:41 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Applied Economics Letters Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/rael20 The weekend effect, good news, bad news and the Financial Times Industrial Ordinary Shares Index: 1935–94 Zainudin Arsad & J. Andrew Coutts Published online: 05 Oct 2010. To cite this article: Zainudin Arsad & J. Andrew Coutts (1996) The weekend effect, good news, bad news and the Financial Times Industrial Ordinary Shares Index: 1935–94, Applied Economics Letters, 3:12, 797-801, DOI: 10.1080/135048596355628 To link to this article: http://dx.doi.org/10.1080/135048596355628 PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Applied Economics Letters, 1996, 3, 797–801 The weekend effect, good news, bad news and the Financial Times Industrial Ordinary Shares Index: 1935–94 ZAINUDIN ARSAD and J. ANDREW COUTTS* Research Student, Department of Statistics, Herriot Watt University, and *Department of Accounting and Finance, Sheffield University Management School, 9 Mappin Street, Sheffield S1 4DT, UK Downloaded by [University of Arizona] at 16:41 19 April 2015 Received 3 November 1995 In recent years much evidence has been documented of the existence of regularities in stock price returns, and consequently the notion of market efficiency has been questioned. The primary objective of this paper is to investigate the ‘weekend’ effect for a large sample of daily returns from the Financial Times Industrial Ordinary Shares Index (FT 30). Empirical results lead us to tentatively suggest that a weekend effect has existed for the FT 30, but that this regularity has not been persistent. We then partition the Monday returns into positive and negative returns, and find that whilst the weekend effect holds for the Mondays, with negative returns, it fails to hold for Mondays which exhibit positive returns. These results support the results of earlier research. Finally we conclude that these results do not contest the notion of market efficiency. I. INTRODUCTION In recent years much empirical research in financial economics has focused on regularities or seasonalities in financial markets. Since the pioneering study by Cross (1973) researchers have investigated financial markets across the world, in the hope of confirming existing, or documenting new regularities. See Mills and Coutts (1995) for a review of the international evidence. One of the most prevalent anomalies appears to be a weekend effect, where stocks display significantly lower returns over the period between Friday’s close and Monday’s close. A number of recent articles (Board and Sutcliffe, 1988, Kim, 1988, Yadav and Pope, 1992 and Mills and Coutts, 1995) have confirmed the existence of this so-called ‘weekend effect’, for various UK indices. It is the aim of this paper to contribute to this debate by investigating the weekend effect in the Financial Times Industrial Ordinary Shares Index (FT 30), over a sixty year period: July 1935 through to December 1994. II. DATA AND METHODOLOGY The FT 30 was the first major UK share index on the London International Stock Exchange, and its computation began on 1 July 1935. The original purpose of the index was to measure 1350–5851 Ó 1996 Routledge market movements over the short term and not to provide any estimates of market return or to act as a benchmark portfolio. The index comprises 30 heavily traded ‘blue chip’ securities chosen to provide a representative spread across British Industry and commerce. Sutcliffe (1993) notes that these 30 securities account for almost 30% of the market value of the securities quoted on the London Stock Exchange, and to this extent mirrors movements of the whole market quite effectively. This index is ideal for investigating stock market anomalies as it reflects a broad industrial base, and as its constituent securities are frequently traded, the problem of ‘thin trading’ is eliminated. The index is available on a daily basis form 1 July 1935 through 31 December 1994, giving a total of 14 888 observations after holidays have been excluded. Daily returns are calculated as R t Pt Pt 1 , The sample period of sixty years covers the vast majority of the economic events of the twentieth century. Our ‘close to close’ data does not, however, contain information about the payment of dividends, with the obvious implication that we cannot extend our analysis to consider previous research which has concentrated explicitly on the role of dividends in return anomalies; see for example Brennan (l970), Blume (1980) and Litzenburger and Ramaswamy (1979, 1980). However, as Mills and Coutts (1995) document, the exclusion of dividend payments may not 797 Downloaded by [University of Arizona] at 16:41 19 April 2015 798 Z. Arsad and J. A. Coutts necessarily invalidate the results of our analysis. They cite the work of Lakonishok and Smidt (1988) and Fishe, Gosnell and Lasser (1993), who conclude that any dividend bias which occurs from not employing dividend adjusted returns is relatively small and will not impact on the statistical significance of any results. However, Philips-Patrick and Schneeweis (1988) found conflicting evidence; consequently, we need to be cautious when interpreting our results and attempting to draw firm inferences. The data consists of the daily closing values of the FT 30 Index from 1 July 1935 through to 31 December 1994, the data were filtered to exclude holidays which left a data set consisting of 14 887 returns, where daily returns (R t ) are given by R t Pt Pt 1 . As well as investigating the weekend effect for the whole sample, the data were separated into twelve sub-samples of five years: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 1 1 1 1 1 1 1 1 1 1 1 1 July 1935–31 December 1939 January 1940–31 December 1944 January 1945–31 December 1949 January 1950–31 December 1954 January 1955–31 December 1959 January 1960–31 December 1964 January 1965–31 December 1969 January 1970–31 December 1974 January 1975–31 December 1979 January 1980–31 December 1984 January 1985–31 December 1989 January 1990–31 December 1994 III. ’GOOD’ NEWS, ‘BAD’ NEWS AND THE WEEKEND EFFECT These twelve sub-samples not only allow us to test for the weekend effect over short periods of time, they will also indicate whether there is a consistent weekend effect. The following regression was run, for the whole sample and the twelve sub-samples, to test whether there is any statistically significant difference among index returns on different days of the week: Rt 1 D1t 2 D2t 3 D3t 4 D4t 5 D5t t 1 where R t is the return on the FT 30, D1t a dummy variable which takes the value 1 if day t is a Monday and 0 otherwise, D2 t is a dummy variable which takes the value 1 if day t is a Tuesday, and 0 otherwise; and so on. The OLS coefficients 1 to 5 are the mean returns for Monday through Friday respectively. The stochastic term is indicated by t . The null hypothesis tested is: H0 1 2 3 4 5 stock price returns are non-normal and display leptokurtic ‘fat tailed’ properties, see for example, (Fama 1965, Badrinath and Chatterjee, 1988 and Agganval, Rao and Hiraki 1989), we also report the non-parametric KruskallWallis-H statistic. This statistic makes no distributional assumptions about stock price returns. Finally the Fstatistics is also reported. From Table 1 we see that for the entire sample period, 1935–94 and for the twelve sub-samples, the mean return for Mondays is negative. We note further that for the whole sample and for six of the twelve sub-samples there is a significant weekend effect. Hence we can tentatively conclude that, in general, the FT 30 Index displayed a weekend effect over the period 1935–94, although this effect was not persistent. 2 against the alternative that all ’s are not equal. If the null hypothesis is rejected then the stock returns must exhibit some form of day-of-the-week seasonality. Table 1 contains summary statistics for daily index returns through different time periods. First, we report tstatistics: however, as previous research has suggested that The above weekend effect fails to take into account the environment of the market in which the stocks were traded. If the market has increased (decreased) on a particular day, then this may be interpreted as the consequence of a positive (negative) information flow, in other words, the market can be identified as either a good or bad news environment. A consequence of the weekend effect is that the market return will not just be lower in ‘bad’ news environments, it should ceteris paribus be lower in both good and bad news environments. Table 2 presents the daily average, after the entire sample has been partitioned on the basis of positive and negative return days. The results for bad news are consistent with the results contained in Table 1: more bad news is present on Monday, and the average return is significantly lower than other weekdays. However, the good news sample is not consistent with the results contained in Table 1. The average Monday return is not significantly different to that of Thursday and Friday. A further result of interest is that Monday’s mean is not the lowest of the daily returns amongst the good news sample, rather, it is Friday which displays the lowest mean return. If we assume that negative returns are associated with bad news, then our results suggest that there is more bad news arriving on Mondays compared to the rest of the week, as more negative returns occur on a Monday than other days. This result offers support to the conclusions of Patell and Wolfson (1982) and Penman (1987), who find that good news announcements are more probable during trading hours and that a higher proportion of announcements occur after the close of trade on Friday than other days. Finally, they also concluded that more unanticipated negative earnings announcements take place on Mondays or over the weekend when trading is closed. 799 The weekend effect in the FT 30 Downloaded by [University of Arizona] at 16:41 19 April 2015 Table 1. Summary statistics on the day of the week effect in the FT 30 Index Mean Std. Dev. F-stat K-W stat Observations 1935–94 Monday Tuesday Wednesday Thursday Friday 0 129 0.052 0.066 0.035 0.074 1.055 1.049 0.976 0.995 0.953 6 79*** 2 72*** 3.60*** 1.89** 4.01*** 19.96*** (0.000) 98.62*** (0.000) 2923 2870 2904 2901 2798 1935–39 Monday Tuesday Wednesday Thursday Friday 0 193 0 026 0 002 0.048 0.022 0.973 ().777 0.841 0.854 0.679 3 39*** 0 46 0 04 0.88 0.40 2.85** (0.023) 11.88*** (0.019) 213 214 227 232 228 1940–45 Monday Tuesday Wednesday Thursdav Friday 0 028 0.019 0.024 0.059 0.073 0.594 0.457 0.472 0.540 0.502 0 85 0.57 0.76 1.83* 2.28** 1.47 (0.210) 4.98 (0.290) 246 247 258 257 256 1945–49 Monday Tuesday Wednesday Thursday Friday 0 054 0 037 0 050 0.023 0.078 0.591 0.493 0.612 0.598 0.516 1 49 0 037 0 050 0.023 0.078** 2.58** (0.036) 8.25* (0.084) 242 240 254 255 253 1950–54 Monday Tuesday Wednesday Thursday Friday 0 040 0 022 0.052 0.131 0.069 0.390 0.428 0.544 0.454 0.454 0 040 0 022 0.052* 0.131*** 0.069** 5.51*** (0.000) 26.76*** (0.000) 240 241 258 258 253 1955–59 Monday Tuesday Wednesday Thursday Friday 0 099 0.027 0.207 0.09 0.021 0.816 0.725 0.691 0.830 0.892 1 94* 0.53 4.18*** 1.82* 0.42 4.94*** (0.001) 27.58*** (0.000) 244 246 258 258 252 1960–64 Monday Tuesday Wednesday Thursday Friday 0.140 0 060 0.082 0.090 0 014 0.817 0.833 0.872 0.743 0.682 2 75*** 1 17 1.66* 1.83* 0 28 3.76*** (0.005) 21.73*** (0.000) 242 241 255 258 254 1965–69 Monday Tuesday Wednesday Thursday Friday 0 094 0.025 0.122 0.064 0 055 0.914 1.220 1.070 0.869 0.778 1 47 0.39 1.84* 1.05 0 89 1.84 (0.118) 13.91*** (0.008) 241 243 256 257 252 1970–74 Monday Tuesday Wednesday Thursday Friday 0 327 0.197 0 090 0 178 0.058 1.405 1.499 1.375 1.417 1.466 3 55*** 2.14** 1 10 1 99** 0.64 4.94** (0.001) 16.20*** (0.003) 238 237 255 257 254 t-stat 800 Z. Arsad and J. A. Coutts Downloaded by [University of Arizona] at 16:41 19 April 2015 Table 1 continued. Summary statistics on the day of the week effect in the FT 30 Index Mean Std. Dev. K-W stat Observations 1975–79 Monday Tuesday Wednesday Thursday Friday 1980–84 0.262 0.305 0.095 0 029 0.256 1.834 1.718 1.524 1.611 1.583 2 45** 2.84*** 0.91 0 28 2.46** 4.64*** (0.001) 23.95** (0.000) 235 234 255 257 252 Monday Tuesday Wednesday Thursday Friday 0 019 0.076 0.151 0.006 0.143 1.150 1.251 1.202 1.174 1.043 0 25 0.99 2.07** 0.09 1.95** 1.09 (0.361) 6.08 (0.194) 235 234 255 257 252 1985–89 Monday Tuesday Wednesday Thursday Friday 0 235 0.086 0.184 0.025 0.196 1.291 1.264 0.998 1.123 1.022 3 14*** 1.15 2.58*** 0.35 2.71*** 5.62*** (0.000) 21.63*** (0.000) 233 236 256 256 250 1990–94 Monday Tuesday Wednesday Thursday Friday 0 054 0.023 0.018 0.097 0.028 1.013 0.878 0.823 0.998 0.967 0 80 0.33 0.27 1.50 0.42 0.66 (0.620) 13.91*** (0.008) 193 191 208 210 206 t-stat F-stat ***, ** and * denote statistical significance at the 1%, 5% and 10% level respectively, for a two-tailed test. IV. CONCLUSION positive or negative, to reflect either a good news or bad news market environment, we find very strong evidence for the existence of the weekend effect in a bad news environment. However in the case of the good news environment, we find that the weekend effect no longer exists, this offers confirmatory evidence of previous studies. Finally these results lead us to conclude that the lack of a persistent weekend effect does not offer a challenge to the notion of market efficiency. This paper has investigated the existence of a weekend effect in the FT 30 employing 14 887 daily logarithmic returns for the period July 1935 through December 1994, and also for twelve five year sub-samples. The initial conclusion being that although the weekend effect exists for the entire period, it does not exist for all of the twelve five year sub-samples. When the Monday returns are partitioned on the basis of whether they are Table 2. Summary statistics for the ‘good’ and ‘bad’ news market environment Day Positive returns Std. Dev. Observation Monday Tuesday Wednesday Thursday Friday 0.664 a 0.738 b 0.703 b 0.671 a 0.653a 0.706 0.766 0.693 0.742 0.776 1235 1499 1599 1562 1576 F-statistics 3.25** Negative returns 0 811 0 714 0 715 0 722 0 650 Std. Dev. Observation 0.867 0.804 0.729 0.794 0.689 1450 1349 1287 1303 1229 7.27*** Within a column a different superscript represents a significant difference at the 95% level; group means with the same superscript are not significantly different from each other. ** denotes significance at the 95% level, *** denotes significance at the 99% level. The weekend effect in the FT 30 Downloaded by [University of Arizona] at 16:41 19 April 2015 REFERENCES Aggarwal, R., Rao, R. P and Hiraki, T. (1989) Skewness and kurtosis in Japanese equity returns: empirical evidence, The Journal of Financial Research, 12, 253–60 Badrinath, S. G. and Chatterjee, S. (1988) On measuring skewness and elongation in common stock return distributions: the case of the market index, Journal of Business, 61, 451–72. Blume, M. (1980) Stock returns and dividend yields: some more evidence, Review of Economics, 62, 567–77. Board, J. L. G. and Sutcliffe, C. M. S. (1988) The weekend effect in UK stock market returns, Journal of Business Finance and Accounting, 15, 1611–26. Brennan, M. (1970) Taxes, market valuation and corporate financial policy?, National Tax Journal, 412–27. Cross, F. (1973) The Behaviour of stock prices on Mondays and Fridays Financial Analysts Journal, Nov.-Dec., 67–9. Fama, E. F. (1965) The behaviour of stock market prices, Journal of Business, 38, 34–105. Fishe, R. P. H., Gosnell, T. F. and Lasser, D. J. (1993) Good news, bad news, volume and the Monday effect, Journal of Business Finance and Accounting, 20, 881–92. Kim, S-W. (1988) Capitalizing on the weekend effect, Journal of Portfolio Management, 14, 59–63. Lakonishok, J. and Smidt, S. (1988) Are seasonal anomalies real? a ninety year perspective, Review of Financial Studies, 1, 403–25. 801 Litzenburger, R. and Ramaswamy, K. (1979) The effect of personal taxes and dividends on capital asset prices: theories and empirical evidence, 7, 163–95. Litzenburger, R. and Ramaswamy, K. (1980) Dividends, short selling restrictions, tax-induced investor clienteles and market equilibrium, Journal of Finance, 35, 459–82. Mills, T. C. and Coutts, J. A. (1995) Calendar effects in the London Stock Exchange FT-SE Indices, European Journal of Finance, 1, 79–94. Patell, J. M. and Wolfson, M. A. (1982) Good news, bad news and the intraday timing of corporate disclosures, Accounting Review, 67, 509–27. Penman, S. H. (1987) The distribution of earnings news over time and seasonalities in aggregate stock returns, Journal of Financial Economics, 18, 199–228. Philips-Patrick, F. J. and Schneeweis, T. (l988) The ‘weekend effect’ for stock market index and stock index futures, Journal of Futures Markets, 8, 115–21. Sutcliffe, C. M. S. (1993) Stock Index Futures: Theories and International Evidence, Chapman and Hall, London. Tang, G. Y. N. (1996) Day-of-the-week effect on skewness and kurtosis: a direct test and portfolio effect, European Journal of Finance, forthcoming. Yadav, P. K. and Pope, P. F. (1992) Intraweek and intraday seasonalities in stock market risk premia: cash and futures, Journal of Banking and Finance, 26, 233–70.
