8 E.F.Fama,K.R.French Journal of Financial Economics 60 (2001)3-43 100 80 Non-Payers Never Paid Payers Former Payers 0 1925 1936 1946 1956 1966 1978 1986 1996 Year Fig.2.The percent of CRSP firms in different dividend groups.The CRSP sample includes NYSE, AMEX,and NASDAQ securities with share codes of 10 or 11.A firm must have market equity data (price and shares outstanding)for December of year t to be in the sample for that year.We exclude utilities(SIC codes 4900-4949)and financial firms (SIC codes 6000-6999).Payers pay dividends in year t:non-payers do not.The two subgroups of non-payers are firms that have never paid and former payers(firms that do not pay in year t but did pay in a previous year). for 1963-77,3.5%per year,but it is lower than the rate for 1927-62,5.4%per year.A relatively steady termination rate is consistent with the evidence in DeAngelo and DeAngelo(1990)and DeAngelo et al.(1992)that only distressed firms(with strongly negative earnings)terminate dividends.In contrast,during 1978-99,dividend payers merge into other firms at the rate of 3.9%per year. This is higher than the merger rates for 1927-62(0.6%per year)and 1963-77 (2.7%per year).Dividend payers delist at the rate of 0.9%per year during 1978-99,versus0.3%for1927-62and0.8%for1963-77. Dividend payers disappear at a higher rate during 1978-98,but the more important factor in the decline in the number of payers is the failure of new payers to replace those that are lost.Former payers (always a relatively small group)resume dividends at an average rate of 11.8%per year during 1963-77; this rate falls to 6.2%per year for 1978-99 and 2.5%for 1999.New lists surge after 1978,but the proportion paying dividends in the year of listing declines from 50.8%for 1963-77 to 9.0%for 1978-99 and only 3.7%in 1999 (Table 1). New lists feed a swelling group of firms that never get around to paying dividends.The initiation rate for firms that have never paid dividends drops from 7.1%per year for 1963-77 to 1.8%for 1978-99 and a tiny 0.7%for 1999
Fig. 2. The percent of CRSP "rms in di!erent dividend groups. The CRSP sample includes NYSE, AMEX, and NASDAQ securities with share codes of 10 or 11. A "rm must have market equity data (price and shares outstanding) for December of year t to be in the sample for that year. We exclude utilities (SIC codes 4900}4949) and "nancial "rms (SIC codes 6000}6999). Payers pay dividends in year t; non-payers do not. The two subgroups of non-payers are "rms that have never paid and former payers ("rms that do not pay in year t but did pay in a previous year). for 1963}77, 3.5% per year, but it is lower than the rate for 1927}62, 5.4% per year. A relatively steady termination rate is consistent with the evidence in DeAngelo and DeAngelo (1990) and DeAngelo et al. (1992) that only distressed "rms (with strongly negative earnings) terminate dividends. In contrast, during 1978}99, dividend payers merge into other "rms at the rate of 3.9% per year. This is higher than the merger rates for 1927}62 (0.6% per year) and 1963}77 (2.7% per year). Dividend payers delist at the rate of 0.9% per year during 1978}99, versus 0.3% for 1927}62 and 0.8% for 1963}77. Dividend payers disappear at a higher rate during 1978}98, but the more important factor in the decline in the number of payers is the failure of new payers to replace those that are lost. Former payers (always a relatively small group) resume dividends at an average rate of 11.8% per year during 1963}77; this rate falls to 6.2% per year for 1978}99 and 2.5% for 1999. New lists surge after 1978, but the proportion paying dividends in the year of listing declines from 50.8% for 1963}77 to 9.0% for 1978}99 and only 3.7% in 1999 (Table 1). New lists feed a swelling group of "rms that never get around to paying dividends. The initiation rate for "rms that have never paid dividends drops from 7.1% per year for 1963}77 to 1.8% for 1978}99 and a tiny 0.7% for 1999. 8 E.F. Fama, K.R. French / Journal of Financial Economics 60 (2001) 3}43
E.F. Fama. French Journal of Financial Economics 60 (2001)3-43 9 科 6661 9 g 今 8s 86-166 IIS 86-4661 sec P55 8 紧 86-4661 然 系 -6-8861 青 是 云 P 610 蓝 目 66696366126-99669036129-30560-71622-196169896166316101956629529-926 26-886 SIS 76-336-78-86 1080 73-186 629 百 三 器 18-0861 006 000 季 28-36 2382 28-8561 c8S 81g 8'IE 3 雪 78-8.61 209 兵 0 75-266 678.2 20 多 冷 060 0091 00 20796100-096 99 三 21女 20-8961 999 后 501 27-79579-32676-92577-19586896 女20L 6 600 125 455 浴 声 G 4 2050 20W 云 登 景 天 86-956 月 女 66-8/670-59666-596m S13 云丹 805 511796 51E 府 常 616℃ S02 贡 E 86-7960 Z'IS 8.34 厨 5 29-226m 165 0.00 EsI 片 0E8 SISII MON suuy isndwo jo suno SIS1I MON
Table 1 Counts and percents of CRSP and Compustat "rms in di !erent dividend groups Payers pay dividends in year t; non-payers do not. The two subgroups of non-payers are "rms that have never paid and former payers ("rms that do not pay in year t but did pay in a previous year). A new list is a "rm that "rst appears on CRSP or Compustat in year t. (See the appendix for more complete de"nitions of payers, non-payers, and new lists.) New Lists that Pay is the percent of newly listed "rms that pay in year t. The numbers are averages of annual values for the indicated time periods. Counts of CRSP "rms 1926}62 1963}99 1963}77 1978}99 1963}67 1968}72 1973}77 1978}82 1983}87 1988}92 1993}98 1999 All "rms 762 3,679 2,528 4,464 1,779 1,948 3,856 3,735 4,357 4,276 5,208 5,113 New lists 25 305 114 436 99 140 103 286 515 352 584 322 Counts of Compustat "rms 1963}98 1963}77 1978}98 1963}67 1968}72 1973}77 1978}82 1983}87 1988}92 1993}98 All "rms 2,919 1,823 3,702 1,024 1,600 2,847 2,883 3,301 3,566 4,831 New lists 205 69 302 61 100 45 112 283 261 511 Percents of CRSP "rms 1926}62 1963}99 1963}77 1978}99 1963}67 1968}72 1973}77 1978}82 1983}87 1988}92 1993}98 1999 Payers 74.7 47.6 65.5 35.5 71.6 64.6 60.3 58.2 36.1 29.4 23.5 20.8 Non-payers 25.3 52.4 34.5 64.5 28.4 35.4 39.7 41.8 63.9 70.6 76.5 79.2 Never paid 10.0 42.2 25.4 53.7 20.9 23.7 31.6 31.8 51.7 58.7 66.6 70.1 Former payers 15.3 10.1 9.1 10.8 7.4 11.7 8.2 10.0 12.1 11.9 9.8 9.1 New lists 3.5 7.8 5.2 9.6 5.6 7.2 2.7 7.5 11.7 8.2 11.2 6.3 New lists that pay 83.0 25.9 50.8 9.0 72.1 47.3 33.1 15.7 8.8 7.9 5.2 3.7 Percents of Compustat "rms 1963}98 1963}77 1978}98 1963}67 1968}72 1973}77 1978}82 1983}87 1988}92 1993}98 Payers 51.2 68.5 38.8 74.5 66.6 64.3 62.7 40.0 31.9 23.6 Non-payers 48.8 31.5 61.2 25.5 33.4 35.7 37.3 60.0 68.1 76.4 Never paid 39.7 23.1 51.7 19.4 23.0 26.8 28.6 49.7 58.1 67.1 Former payers 9.1 8.5 9.6 6.1 10.5 8.9 8.7 10.2 10.0 9.3 E.F. Fama, K.R. French / Journal of Financial Economics 60 (2001) 3}43 9
10 E.F.Fama.K.R. French Journal of Financial Economics 60 (2001)3-43 666 三 ￥闪 5 3 生 666 88 将 86-2661 6.16 学 阴 3 86-2661 ￥子 36-466L 26-8861 女68 8 照图 26-3861 高 运 阳 G 26-3861 70-786 母 S 28-2861 三 将 17-2161 3 Z8-8L6I 3.06 于 网 E 78-8161 只 于 18-8761 只 67-226125-8961 046 的 9 2 77-1165 E25 员 将子 771561 日 g 216 导 阳 3 于 206 26-3961 L女6 3 吕 兰 79-OR5 注 闷 39-006 8于 L706 只 8 深 G 66-376107-296166-596129-0261 IE6 66-306100-896166-296129-026m 器 6t8 66-326107-296166-596129-0261 器 F 兰 V16 于 高 房 丹 云 ￥ 726 古 器 3 Z'I8 3 2 Z aqel Ked o]onunuo 38JW sHad SuIKed jes Ked iou od isId
Table 2 What happens in year t to CRSP "rms that do and do not pay dividends in year t!1 Firms that Continue to Pay pay dividends in years t!1 and t. Firms that Stop Paying pay dividends in t!1 and not in t. Firms that Merge are delisted in year t with a CRSP delist code between 200 and 299. Delist includes all other "rms delisted in year t. The numbers are averages of annual values for the indicated time periods. What happens in year t to "rms that pay dividends in year t!1 (percent) 1927}62 1963}99 1963}77 1978}99 1963}67 1968}72 1973}77 1978}82 1983}87 1988}92 1993}98 1999 Continue to pay 93.7 91.3 93.1 90.1 94.1 91.3 94.0 90.8 88.1 89.4 91.9 89.6 Stop paying 5.4 4.4 3.5 5.0 2.6 4.5 3.3 4.1 5.7 6.0 4.5 4.2 Merge 0.6 3.4 2.7 3.9 2.6 3.5 1.9 3.9 4.4 3.8 3.2 5.5 Delist 0.3 0.8 0.6 0.9 0.8 0.5 0.5 1.1 1.7 0.8 0.4 0.5 What happens in year t to "rms that do not pay dividends in year t!1 (percent) 1927}62 1963}99 1963}77 1978}99 1963}67 1968}72 1973}77 1978}82 1983}87 1988}92 1993}98 1999 Start paying 15.0 5.2 8.8 2.7 9.9 4.1 12.3 5.0 2.5 2.4 1.5 0.9 Do not pay 81.2 86.0 84.9 86.8 84.3 90.3 80.0 86.4 86.3 85.5 89.3 82.3 Merge 0.8 3.2 2.3 3.8 2.1 2.5 2.3 3.7 2.8 3.5 4.2 7.3 Delist 2.9 5.1 3.5 6.3 3.6 2.5 4.3 4.1 7.7 8.1 4.9 9.3 Percent of non-payers in year t!1 that start paying in year t 1927}62 1963}99 1963}77 1978}99 1963}67 1968}72 1973}77 1978}82 1983}87 1988}92 1993}98 1999 All non-payers in t!1 15.0 5.2 8.8 2.7 9.9 4.1 12.3 5.0 2.5 2.4 1.5 0.9 Never paid in t!1 8.5 4.0 7.1 1.8 6.9 3.2 11.3 3.6 1.4 1.7 1.0 0.7 Former payers in t!1 16.9 8.5 11.8 6.2 14.1 6.2 15.1 9.4 6.5 5.8 4.3 2.5 10 E.F. Fama, K.R. French / Journal of Financial Economics 60 (2001) 3}43
E.F.Fama,K.R.French Journal of Financial Economics 60 (2001)3-43 11 Although mergers contribute to the decline in the number of dividend payers, they are not important in the decline in the percent of payers.During the critical 1978-99 period,non-payers merge into other firms at about the same rate(3.8% per year)as payers(3.9%per year),so mergers have little effect on the percent of firms paying dividends.Non-payers delist at a higher rate(6.3%per year for 1978-99)than payers (0.9%per year).Thus,delistings reduce the number of firms paying dividends,but they actually increase the percent of firms paying. Fig.2 gives a simple view of the factors that contribute to the decline in the percent of firms paying dividends.Terminations by dividend payers and re- sumptions by former payers have little net effect.Terminations and resumptions determine the population of former payers,which grows from 319 firms in 1978 to 466 in 1999 (Fig.1).Because the number of listed firms also grows,the proportion of all firms accounted for by former payers only rises from 8.8%in 1978 to 9.1%in 1999 (Fig.2).As a result,the decline in the proportion of firms paying dividends (from 66.5%in 1978 to 20.8%in 1999)almost matches the growth in the proportion that have never paid (from 24.7%in 1978 to 70.1%in 1999).This group (new lists that never become dividend payers)is a big factor in both the decline in the numerator of the percent of dividend payers(the number of payers)and the increase in the denominator(the number of sample firms). The rest of the paper addresses two questions raised by the declining inci- dence of dividend payers:(i)Has the population of firms drifted toward a lower frequency of firms with the characteristics typical of payers,or(ii)have firms with the characteristics typical of payers become less likely to pay dividends?We start by establishing the characteristics of dividend payers,and the declining incidence of these characteristics among publicly traded firms. 3.Characteristics of dividend payers Our evidence on the characteristics of dividend payers and non-payers is from Compustat.The time period,1963-98,is shorter than the 1926-99 CRSP period examined above,but the Compustat data cover the post-1972 NYSE-AMEX- NASDAQ period and the post-1978 period of most interest to us. On average,the CRSP sample has about 750 more firms than the Compustat sample in their shared 1963-98 period (Table 1).The difference between the samples is due to CRSP's more complete coverage and the data requirements we impose on the Compustat sample(see the appendix).But the Compustat sample does show the sharp decline in the percent of dividend payers observed in the CRSP sample.Dividend payers average 64.3%of Compustat firms in 1973-77 and 23.6%in 1993-98(Table 1).The averages for CRSP are 60.3%in 1973-77 and23.5%in1993-98. Our initial discussion of the characteristics of dividend payers focuses on the evidence from summary statistics that payers and non-payers differ in terms of
Although mergers contribute to the decline in the number of dividend payers, they are not important in the decline in the percent of payers. During the critical 1978}99 period, non-payers merge into other "rms at about the same rate (3.8% per year) as payers (3.9% per year), so mergers have little e!ect on the percent of "rms paying dividends. Non-payers delist at a higher rate (6.3% per year for 1978}99) than payers (0.9% per year). Thus, delistings reduce the number of "rms paying dividends, but they actually increase the percent of "rms paying. Fig. 2 gives a simple view of the factors that contribute to the decline in the percent of "rms paying dividends. Terminations by dividend payers and resumptions by former payers have little net e!ect. Terminations and resumptions determine the population of former payers, which grows from 319 "rms in 1978 to 466 in 1999 (Fig. 1). Because the number of listed "rms also grows, the proportion of all "rms accounted for by former payers only rises from 8.8% in 1978 to 9.1% in 1999 (Fig. 2). As a result, the decline in the proportion of "rms paying dividends (from 66.5% in 1978 to 20.8% in 1999) almost matches the growth in the proportion that have never paid (from 24.7% in 1978 to 70.1% in 1999). This group (new lists that never become dividend payers) is a big factor in both the decline in the numerator of the percent of dividend payers (the number of payers) and the increase in the denominator (the number of sample "rms). The rest of the paper addresses two questions raised by the declining incidence of dividend payers: (i) Has the population of "rms drifted toward a lower frequency of "rms with the characteristics typical of payers, or (ii) have "rms with the characteristics typical of payers become less likely to pay dividends? We start by establishing the characteristics of dividend payers, and the declining incidence of these characteristics among publicly traded "rms. 3. Characteristics of dividend payers Our evidence on the characteristics of dividend payers and non-payers is from Compustat. The time period, 1963}98, is shorter than the 1926}99 CRSP period examined above, but the Compustat data cover the post-1972 NYSE-AMEXNASDAQ period and the post-1978 period of most interest to us. On average, the CRSP sample has about 750 more "rms than the Compustat sample in their shared 1963}98 period (Table 1). The di!erence between the samples is due to CRSP's more complete coverage and the data requirements we impose on the Compustat sample (see the appendix). But the Compustat sample does show the sharp decline in the percent of dividend payers observed in the CRSP sample. Dividend payers average 64.3% of Compustat "rms in 1973}77 and 23.6% in 1993}98 (Table 1). The averages for CRSP are 60.3% in 1973}77 and 23.5% in 1993}98. Our initial discussion of the characteristics of dividend payers focuses on the evidence from summary statistics that payers and non-payers di!er in terms of E.F. Fama, K.R. French / Journal of Financial Economics 60 (2001) 3}43 11
12 E.F.Fama,K.R.French Journal of Financial Economics 60 (2001)3-43 profitability,investment opportunities,and size.The evidence from the sum- mary statistics is then confirmed with logit regressions. 3.1.Profitability Table 3 details the characteristics of firms in various dividend groups.Divi- dend payers have higher measured profitability than non-payers.For the full 1963-98 period,E/A,(the ratio of aggregate earnings before interest to aggreg- ate assets)averages 7.82%per year for payers versus 5.37%for non-payers. Among non-payers,E/A,averages 4.54%per year for former dividend payers. This is lower than the profitability of firms that have never paid dividends, 6.11%per year,which in turn is below the profitability of dividend payers, 7.82%per year. Earnings before interest,E,are the payoff on a firm's assets,but earnings available for common,Y.,may be more relevant for the decision to pay dividends.Table 3 shows that the gap between the profitability of payers and non-payers is wider when profitability is measured as Y,/BE,(aggregate com- mon stock earnings over aggregate book equity).For 1963-98,Y,/BE,averages 12.75%for dividend payers,versus 6.15%for non-payers.Among non-payers, Y,/BE,averages 7.94%for firms that have never paid dividends and only 3.18% for former payers. Low profitability becomes more common in the second half of the 1963-98 period.The plots of the decile breakpoints for E/A,in Fig.3 provide perspect- ive.Initially the breakpoints drift upward,peaking around 1979 or 1980.After the peak years,profitability declines.The decline is marginal in the higher profitability deciles,but it is large in the lower profitability deciles.The lowest breakpoint(the tenth percentile)switches from consistently positive to consis- tently negative in 1982.At least 20%of firms have negative earnings before interest after 1984.In the last three years,1996-98,negative earnings before interest afflict more than 30%of the firms. Many of the firms that are unprofitable later in the sample period are new listings.Until 1978,more than 90%of new lists are profitable (Fig.4).There- after,the fraction with positive earnings falls.In 1998,only 51.5%of new lists have positive common stock earnings.Table 3 shows that before 1982,new lists even new lists that do not pay dividends-tend to be more profitable than all publicly traded firms.After 1982 the profitability of new lists falls.The deteriora- tion occurs as the number of new lists explodes,and it is dramatic for the increasingly large group of new lists that do not pay dividends.By 1993-98 (when there are 511 Compustat new lists per year and only 5.2%pay dividends), the common stock earnings of newly listed non-payers average only 0.27%of book equity,versus 11.26%for all firms.The low profitability of new lists later in the sample period is in line with similar evidence on the low post-issue profitability of IPO firms (Jain and Kini,1994;Mikkelson et al.,1997)
pro"tability, investment opportunities, and size. The evidence from the summary statistics is then con"rmed with logit regressions. 3.1. Proxtability Table 3 details the characteristics of "rms in various dividend groups. Dividend payers have higher measured pro"tability than non-payers. For the full 1963}98 period, E /A (the ratio of aggregate earnings before interest to aggregate assets) averages 7.82% per year for payers versus 5.37% for non-payers. Among non-payers, E /A averages 4.54% per year for former dividend payers. This is lower than the pro"tability of "rms that have never paid dividends, 6.11% per year, which in turn is below the pro"tability of dividend payers, 7.82% per year. Earnings before interest, E , are the payo! on a "rm's assets, but earnings available for common, > , may be more relevant for the decision to pay dividends. Table 3 shows that the gap between the pro"tability of payers and non-payers is wider when pro"tability is measured as > /BE (aggregate common stock earnings over aggregate book equity). For 1963}98, > /BE averages 12.75% for dividend payers, versus 6.15% for non-payers. Among non-payers, > /BE averages 7.94% for "rms that have never paid dividends and only 3.18% for former payers. Low pro"tability becomes more common in the second half of the 1963}98 period. The plots of the decile breakpoints for E /A in Fig. 3 provide perspective. Initially the breakpoints drift upward, peaking around 1979 or 1980. After the peak years, pro"tability declines. The decline is marginal in the higher pro"tability deciles, but it is large in the lower pro"tability deciles. The lowest breakpoint (the tenth percentile) switches from consistently positive to consistently negative in 1982. At least 20% of "rms have negative earnings before interest after 1984. In the last three years, 1996}98, negative earnings before interest a%ict more than 30% of the "rms. Many of the "rms that are unpro"table later in the sample period are new listings. Until 1978, more than 90% of new lists are pro"table (Fig. 4). Thereafter, the fraction with positive earnings falls. In 1998, only 51.5% of new lists have positive common stock earnings. Table 3 shows that before 1982, new lists } even new lists that do not pay dividends } tend to be more pro"table than all publicly traded "rms. After 1982 the pro"tability of new lists falls. The deterioration occurs as the number of new lists explodes, and it is dramatic for the increasingly large group of new lists that do not pay dividends. By 1993}98 (when there are 511 Compustat new lists per year and only 5.2% pay dividends), the common stock earnings of newly listed non-payers average only 0.27% of book equity, versus 11.26% for all "rms. The low pro"tability of new lists later in the sample period is in line with similar evidence on the low post-issue pro"tability of IPO "rms (Jain and Kini, 1994; Mikkelson et al., 1997). 12 E.F. Fama, K.R. French / Journal of Financial Economics 60 (2001) 3}43