The Relationship between Dividend Payout and Price to Earnings

CNV Krishnan Yifei Chen1

August 9, 2017

Abstract

Using a large database of all S&P 1500 index firms spanning the 88-quarter period from 1995
through 2016, we document that current period dividend payout is significantly and positively
correlated with next period Price-to-Earnings ratio (PE) for high market cap firms and
manufacturing firms, and significantly negatively correlated for high book-to-market firms.
Market cap (firm size), book-to-market ratio (a proxy for market perception of growth potential)
and industry matter for determining PE levels as a function of payout levels. However, once the
PE levels are determined, current period dividend payout change is significantly and negatively
associated with next period PE change. We find evidence supporting an argument that an
increase in current period payout signals reduced investment opportunities and increased risk
that reduce future PE ratios. Thus, modeling determinants of PE must take into account industry,
size, risk and market perception of growth potential for a firm.

Keywords: Price-to-Earnings ratio, PE ratio, dividend payout ratio, industry, market cap, firm
size, book-to-market ratio, growth opportunities, firm risk.

1CNV Krishnan is professor of Banking and Finance at Weatherhead School of Management, Case Western
Reserve University, email: [email protected] Yifei Chen is graduate student at Weatherhead School of
Management, Case Western Reserve University, email: [email protected]
1

Electronic copy available at: https://ssrn.com/abstract=3016040

1. Introduction

Standard text book models show that the Price-to-Earnings ratio (PE), ,

increases as the payout ratio increases for any given growth rate, PE decreases as risk increases,
and PE increases as the growth rate increases (see Appendix A). Investors focus on PE ratios,
classifying into high and low PE stocks. Hough (2011), for example, claims that low PE stocks
outperformed high PE stocks in the 2000’s. We examine whether any such standard model holds
across the cross section of firms.
Extant literature has studied both PE ratios and dividend payouts extensively. Risk matters.
Henne, Ostrowski and Reichling (2009) argue that stock performance generally improves with
increasing dividend yield, but this result is actually based on risk reduction rather than higher
return, in the German market. Ferson (2008) mentions that the argument that a shock to expected
return on equity, ceteris paribus, changes the asset value may overstate the effect to the extent
that a shock that changes the required return also changes the expected future cash flows. Growth
matters. Ang and Zhang (2011) find that growth opportunities account for approximately 95% of
the variation and 80% of the level of PE ratios. Riahi-Belkaoui and Picur (2001) allege that firms
with high investment opportunities are “PE valued”. Profitability matters. Benartzi, Michaely,
and Thaler (1997) and Grullon, Michaely, and Swaminathan (2002) find no evidence that dividend
changes predict abnormal increases in earnings. Penman (1996) fails to find that the current return
on equity is a good indicator of PE ratio. Our goal in this paper is to use a large panel of firms to
check the determinants of PE ratio levels and changes, in particular the current period dividend
payout levels and changes.
We start with examining the determinants of PE levels, and find, in univariate tests, that
industry matters. High-tech and healthcare firms have higher PE ratios, on average, and
manufacturing firms the lowest, broadly reflecting future growth opportunities. Market
capitalization (which is a proxy for firm size) and book-to-market ratio (which is a proxy for (the
reverse of) growth options) matter. The higher the market capitalization, the higher the PE,
showing the market prices in market dominance, and the lower the book-to-market ratio the
higher the PE, reflecting future growth opportunities.
In multivariate regressions, after controlling for industry and time fixed effects, we find
that current period dividend payout is significantly and positively related to next period PE.
Results are significant when we use current earnings or trailing earnings. When we run
2

Electronic copy available at: https://ssrn.com/abstract=3016040

regressions on groups of firms divided by industries, market cap, book-to-market ratio and
recession and non-recession years, we find that results are more complicated. For Consumer and
High Tech industries, the correlation between current period payout and next period PE tends to
be significantly negative, while it is significantly positive for Manufacturing and other industries.
In general, we find a positive relationship between current period payout and future PE for high
market cap firms. So, a simple dividend discount model is perhaps too simple to use across all
industry segments, market capitalization, book-to-market ratios, and other measures of firm risk.
When we examine changes, the change in PE ratio is significantly and negatively
associated with prior changes in payout. This negative relationship not only appears overall, but
also in almost every industry and type of companies: all of the significant associations between
current period change in payout and future change in PE ratio are negative. We examine the
impact of payout changes on future risk, using the following regression specification, and find
that there is indeed a significant and positive relationship. Thus, given PE levels, an increase in
current period payout signals reduced positive growth opportunities in future and increased risk,
and hence results in a decrease in next period PE.

2. Data and Descriptive Statistics

We use all S&P 1500 firms’ (including S&P 500, S&P Mid Cap 400 and S&P Small Cap 600)
data from the Compustat Quarterly from 1995 through 2016, entailing 88 quarters and 132,000
observations in all.
We define book equity (BE) as total shareholders’ equity plus deferred taxes and
investment tax credit (item TXDITCQ) minus the book value of preferred stock (item PSTKQ).
We prefer the shareholders’ equity numbers (SEQQ). In case this data are not available, we
calculate shareholders’ equity as sum of common and preferred equity (items CEQQ and PSTKQ).
If neither of the two are available, we define shareholders’ equity as the differences of total assets
and total liabilities (items ATQ and LTQ). Trailing earnings is the average net income of the past
four quarters. We use 5 industry groups defined by Ken French, which are Consumer,
Manufacturing, High Technology, Healthcare and Other
(see http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html)
We also segregate years by recession and non-recession years. The definition of recession
years is taken from the Business Cycle Dating Committee of the National Bureau of Economic
Research (NBER) based on the behavior of various indicators of economic activity (see
3

Electronic copy available at: https://ssrn.com/abstract=3016040

http://www.nber.org/cycles/cyclesmain.html ). The full list of variables used is shown in
Appendix B.

3. Univariate Results

Figure 1 and Figure 2, respectively, show the time series plots of quarterly average
dividend payout and quarterly average PE overall, as well as by industries, market capitalization
and book-to-market categories, where the recession quarters are 2001 Q2 - 2001 Q4 and 2008 Q1
- 2009 Q2, as determined by NBER, are shaded. Panel A of Figure 1 shows that dividend payout
ratio tends to spike up during recession periods. Panel B shows that manufacturing firm payouts
tend to be higher, Panel C shows the higher the market cap the greater the dividend payout, while
the final panel shows that the higher the book to market ratio, the higher the payout. This is
confirmed by the pairwise differences in Table 1. Manufacturing firms have significantly higher
payout that other firms that have significantly higher payouts, on average, than consumer firms.
Healthcare and high-tech firms have the lowest payouts on average, perhaps reflecting their
future growth potential. The middle panel shows the higher the market capitalization the higher
the payout reflecting firm maturity, while the last panel shows that the higher the book to market
ratio the higher the payout reflecting lack of growth potential. So, in summary, manufacturing
firms payout more, larger firms payout more, and firms with lower future growth potential
payout more, on average.
Panel A of Figure 2 shows that PE ratios, on average, tend to be lower during recessions
because of price depressions. There is no clear pattern for PE ratios emerging when we look by
industries, although healthcare stocks in recent years tend to have higher PE ratios. However,
Panel C shows that low market cap firms (small firms) generally tend to have lower PE ratios,
while Panel D shows that the higher the book to market ratio, the lower the PE ratio, reflecting
the fact that the lower the future growth prospects the lower the PE ratio. This is corroborated by
the pairwise differences among the groups of firms by industries, market capitalization and book
to market in Table 2. High tech and healthcare firms do tend to have higher PE ratios, and
manufacturing the least. The higher the market cap the higher the PE, and the lower the book to
market ratio the higher the PE.
The takeaway is that payout and PE ratios move in opposite directions when we compare
by broad industry categories, in the same direction when we examine by market cap (firm size),
and in the opposite direction when we examine by book to market ratio (firm growth). Overall, a
4

Electronic copy available at: https://ssrn.com/abstract=3016040

cursory examination of panels A of Figures 1 and 2 show that payout and PE ratios tend to move
in opposite directions.

4. Multivariate Results

Table 3 examines the determinants of the PE ratio, using the 2 variants of the following regression:

(1) /

where is a proxy for the inverse of growth options and is the market beta
calculated by running regressions between individual returns and market returns by moving
windows. We also control for year and industry fixed effects. In specification 1, we use price and
EPS as of time (t+1), and in specification 2, we use price as of time (t+1) and EPS as of time t,
lagged earnings computed as the average earnings of the past 4 quarters.
We find that current period dividend payout is positively and significantly related to next
period PE (at 1% significance level) after controlling for year and industry fixed effects and firm
beta. Contrary to univariate results, PE ratio is significantly lower for higher market cap firms,
after controlling for other determinants. In line with the univariate results, PE ratio is significantly
lower for higher book to market firms, because of lower perceived growth opportunities. These
results are significant whether we use current earnings or trailing earnings.
When we run regressions by groups divided by industries, market cap, book to market
ratio and recession years, we find that results are more complicated. Table 4 reports regression
results run separately over 90 groups overall, made up of 5 industry groups, 3 groups of market
capitalization, 3 groups of book to market values, and whether or not recession period. Only the
specifications for which payout ratio is significant in explaining future PE at 5% or 1% levels are
shown. Panel A shows significant negative relations, and Panel B significant positive relations.
For Consumer industry and High Technology industry, the correlation between payout
and PE tends to be significantly negative, while it is significantly positive for Manufacturing and
Other industries. In general, the positive relation between payout and future PE is for high market
cap firms. There are no discernable relations for when the relation between payout and PE is
positive or negative with regard to book-to-market and recession or expansion years, although in
most of the regressions, book to market ratio has a significantly negative correlation with future
PE.
5

Electronic copy available at: https://ssrn.com/abstract=3016040

 The results are somewhat similar when we use trailing earnings instead of current
earnings, in Table 5. For Consumer and High Tech industries, the correlation between payout and
future PE tends to be significantly negative, while it is significantly positive for Manufacturing
and other industries. In general, we find positive relationship between payout and future PE for
high market cap firms and for low book-to-market firms.
To directly examine the impact of changes on changes, we regress change in PE on change
in payout. We have seen that the correlation between payout and PE levels tend to be significantly
positive for high market cap firms and manufacturing firms, and tends to be negative for high
book to market firms. Therefore, the market cap, book-to-market ratio and whether a firm is in
the manufacturing sector or not matters for determining PE levels as a function of payout levels
and we control for this using the following 4 different specifications of the following regression
equation, where ∆ is the difference between dividend payout at t and t-1 and
/ is the difference in PE ratio between times t+1 and t.

/
∆

∆
∆
∆

Table 6 shows that, irrespective of the specification used, next period PE change is significantly
and negatively associated with current period dividend payout change. The previous results have
shown that, on average, dividend payout level is significantly and positively related to PE level.
But once the levels are determined, any change in payout negatively affects future changes in PE.
As before, we run the regression specification separately over 90 groups, made up of 5 industry
groups, 3 groups of market capitalization, 3 groups of book to market values, and whether or not
recession period. Only the specifications for which ∆ is significant at the 5% and
1% levels are shown in Table 7. We find that the above documented negative relation is robust:
it not only in the overall sample, but also in almost every industry and firm type subsample.
Finally, we regress change in volatility in the next period on change in payout, to check
the impact of current period payout on future risk, using the following regression specification:

Electronic copy available at: https://ssrn.com/abstract=3016040


∆

and find that there is indeed a significant and positive relationship (Table 8). Thus increased
payout in the current period signals increased risk in future (perhaps due to reduced growth
opportunities) and, and hence the future PE decreases.
In summary, current period dividend payout change is significantly and negatively
associated with future PE change likely because an increase in current period payout signals
reduced investment opportunities which reduces future PE. Moreover, the relation between
payout and PE levels and between payout changes and PE changes depends on industry, firm
size, perceived growth opportunities (proxied in this paper by book to market ratios), and other
measures of firm risk.

5. Conclusion

Using a large database of all S&P 1500 index firms spanning the 88-quarter period from
1995 through 2016, we document that in terms of dividend payout, manufacturing firms payout
more, large firms payout more, and firms with lower future growth potential payout more, on
average. In terms of PE ratios, in univariate tests, we find that high tech and healthcare firms have
higher PE ratios, on average, and manufacturing firms the least, perhaps reflecting their future
growth potential. The higher the market capitalization, the higher the PE ratio; and the lower the
book to market ratio the higher the PE ratio. So, payout and PE ratios move in opposite directions
when we compare by industry, in the same direction when we examine by market cap, and in the
opposite direction when we examine by book to market ratio, consistent with a perceived growth
potential story. In multivariate regressions, we find that current period dividend payout is
positively and significantly related to next period PE (at 1% significance level) after controlling of
the year and industry effects. After controlling for other factors, future PE ratio is significantly
and negatively associated with market capitalization and higher book to market ratio. Results are
consistently significant when we use current earnings or trailing earnings.
After we run regressions according to groups divided by industries, market cap, book to
market ratio and recession years, we find that for consumer and high technology industries, the
correlation between current period payout and next period PE tends to be significantly negative,

Electronic copy available at: https://ssrn.com/abstract=3016040

while it is significantly positive for manufacturing and other industries. In general, for high
market cap firms, the relation between payout and future PE is positive.
When we examine changes, future PE change is significantly and negatively associated
with current period change in payout. We also find that current period payout change is
associated with future increased risk. This implies that future PE decreases consistent with
increased risk due to perceived lack of growth options stories. Any model of the determinants of
PE is complicated, and should take into account industry, size, risk and market perception of
future growth for a firm.

Electronic copy available at: https://ssrn.com/abstract=3016040

 References
Ackert, L. F., & Hunter, W. C. (2001). An Empirical Examination of the Price-Dividend Relation
with Dividend Management. Journal of Financial Services Research, 115-129.
Ang, A., & Zhang, X. (2011). Price-Earnings Ratios: Growth and Discount Rates. New York:
Columbia Business School.
Arnott, R. D., & Asness, C. S. (2003). Surprise! Higher Dividends = Higher Earnings Growth.
Financial Analysts Journal, 70-87.
Eaton, R. V. (1999). Stock Price Adjustment to the Information in Dividend Changes. Review of
Quantitative Finance and Accounting, 113–133.
Ferson, W., (2008), Stock Return Predictability, The New Palgrave Dictionary of Economics,
2nd edition, 2008. edited by Steven N. Durlauf and Lawrence E. Blume.
Grullon, G., Michaely, R., & Swaminathan, B. (2002). Are Dividend Changes a Sign of Firm
Maturity? Journal of Business, 3, 387-424.
Henne, A., Ostrowski, S., & Reichling, P. (2009). Dividend yield and stability versus
performance on the German stock market: a descriptive study. Springer, 225-248.
Hough, J. (2011, October 9). Peeling Back the Market's P/E. Retrieved from Wall Street Journal:
https://www.wsj.com/articles/SB10001424053111904491704576573241520154756
Penman, S. H. (1996). The Articulation of Price-Earnings Ratios and Market-to-Book Ratios and
the Evaluation of Growth. Journal of Accounting Research, 34(2), 235-259.
Riahi-Belkaoui, A., & Picur, R. D. (2001). Investment Opportunity Set Dependence of Dividend
Yield and Price Earnings Ratio. Managerial Finance, 27(3), 65-71.
Shiller, R. J. (1980). Do Stock Prices Move Too Much to be Justified by Subsequent Changes in
Dividends? . National Bureau of Economic Reserach, 1-40.
Singh, S., Jain, P., & Yadav, S. S. (2016). Analysis of Price Multiples. In S. Singh, P. Jain, & S.
S. Yadav, India Studies in Business and Economics (pp. 127-143). New Delhi, India:
Springer Nature.
Yan, R., & Xie, C. (2012). Industry Stock Price Effect and Its Influencing Factors of Cash
Dividend Distribution: Based on Chinese Real Estat Listed Companies. In J. Kacprzyk,
Advances in Intelligent and Soft Computing (pp. 463-469). Heidelberg: Springer.

Electronic copy available at: https://ssrn.com/abstract=3016040

 Figure 1
Dividend Payout
The top Panel shows the quarter-by-quarter average dividend payout ratio defined as average of quarterly
cash dividend divided by quarterly net income, for all quarters from 1995 through 2016 where the recession
quarters are 2001 Q2 - 2001 Q4 and 2008 Q1 - 2009 Q2 as determined by NBER, are shaded. The second
panel shows the average payout ratio quarter-by-quarter for the 5 industry groups. The third panel shows
the average payout ratio quarter by quarter by three groups by firms’ market capitalization. The last panel
shows the average payout ratio quarter by quarter by three groups by firms’ book to market value. All
variables are defined in Appendix B.

10

Electronic copy available at: https://ssrn.com/abstract=3016040

 11

Electronic copy available at: https://ssrn.com/abstract=3016040

 Figure 2
PE Ratio

The top Panel shows the quarter-by-quarter average PE ratio defined as average of quarterly stock price
divided by quarterly EPS for all quarters from 1995 through 2016 where the recession quarters are 2001 Q2
- 2001 Q4 and 2008 Q1 - 2009 Q2 as determined by NBER, are shaded. The second panel shows the average
payout ratio quarter-by-quarter for 5 industry groups. The third panel shows the average payout ratio
quarter by quarter by three groups by firms’ market capitalization. The last panel shows the average payout
ratio quarter by quarter by three groups by firms’ book to market value. All variables are defined in
Appendix B.

12

Electronic copy available at: https://ssrn.com/abstract=3016040

 13

Electronic copy available at: https://ssrn.com/abstract=3016040

 Table 1
Dividend Payout
The top panel shows the differences in mean Payout ratios for 5 Fama-French industry groups. The middle
panel shows the differences in mean Payout ratios for three groups of market capitalization. The bottom
panel shows the differences in mean Payout ratios for three groups of book to market value. All variables
are defined in Appendix B.

Average Dividend Payout by Industries

Mean(X) Mean(Y) Mean(X-Y) t-stat df p-value

Manuf = 0.31 Other = 0.22 0.09 40.4 50337 < 2.2e-16

Other = 0.22 Cnsmr = 0.20 0.02 11.6 48357 < 2.2e-16

Cnsmr = 0.20 Hlth = 0.06 0.14 71.4 25381 < 2.2e-16

Hlth = 0.06 HiTec = 0.06 0.00 0.12 13760 0.90

Average dividend payout by Market Cap

Mean(X) Mean(Y) Mean(X-Y) t-stat df p-value

High MVE = Mid MVE =

0.08 45.7 72370 < 2.2e-16
0.27 0.19
Mid MVE = Low MVE =
0.06 34.8 34.771 < 2.2e-16
0.19 0.13

Average dividend payout by Book to Market

Mean(X) Mean(Y) Mean(X-Y) t-stat df p-value

High BM = Mid BM =
0.08 36.6 56125 < 2.2e-16
0.27 0.19
Mid BM = Low BM =
0.02 12.7 71775 < 2.2e-16
0.19 0.17

14

Electronic copy available at: https://ssrn.com/abstract=3016040

 Table 2
PE Ratio

The top panel shows the differences in mean PE for 5 Fama-French industry groups. The middle panel
shows the differences in mean PE ratios for three groups of market capitalization. The bottom panel shows
the mean differences in mean PE ratios for three groups of book to market. All variables are defined in
Appendix B.

Average PE by Industry

Mean(X) Mean(Y) Mean(X-Y) t-stat df p-value

Hlth = 78.03 HiTec = 76.13 1.90 2.51 20282 0.01

HiTec = 76.13 Cnsmr = 66.22 9.91 17.34 30784 < 2.2e-16

Cnsmr = 66.22 Other = 59.80 6.42 19.11 43835 < 2.2e-16

Other = 59.80 Manuf = 59.38 0.42 1.41 54063 0.16

Average PE by Market Cap

Mean(X) Mean(Y) Mean(X-Y) t-stat df p-value

High MVE = Mid MVE =

3.66 12.75 72783 < 2.2e-16
71.91 68.25

Mid MVE = Low MVE =

15.84 34.77 34.771 < 2.2e-16
68.25 52.41

Average PE by Book to Market Ratio

Mean(X) Mean(Y) Mean(X-Y) t-stat df p-value

High BM = Mid BM =
-21.19 -71.15 67201 < 2.2e-16
44.47 65.65

Mid BM = Low BM =
-19.75 -68.65 69341 < 2.2e-16
65.65 85.40

15

Electronic copy available at: https://ssrn.com/abstract=3016040

 Table 3
Dividend Payout and PE Ratio

This table reports the regression coefficients and the associated heteroscedasticity consistent t-statistics
(along with adjusted R2) of 2 specifications of:

/
 
All variables are defined in Appendix B.
 

P/E Ratiot+1
P/E Ratiot+1
(trailing earnings)

2.74 *** 4.72 ***

Dividend Payoutt
(4.46) (8.69)
-1.382 -4.05 ***
Recession
(-1.43) (-4.73)
-22.74 38.96.
Volatility
(-0.93) (1.79)
-0.48 *** -1.04 ***
Log(Market Cap)
(-4.35) (-10.70)
-34.43 *** -35.66 ***
Book to Market
(-70.80) (-83.03)
3.69 *** 4.14 ***
Beta
(7.95) (10.03)

Industry Fixed Effects Yes Yes

Year Fixed Effects Yes Yes

Adjusted R2 0.78 0.84

*, **, and *** denote significant at the 10%, 5% and 1% level respectively.

16

Electronic copy available at: https://ssrn.com/abstract=3016040

 Table 4
Dividend Payout and PE Ratio by Groups

This table reports regression coefficients and the associated heteroscedasticity consistent t-statistics (along
with adjusted R2) of
/

run separately over 90 groups overall, made up of 5 industry groups, 3 groups of market capitalization, 3
groups of book to market values, and whether recession period. Only the specifications for which Dividend
Payout is significant at 5% or 1% levels are shown. Panel A shows significant negative relations, and Panel
B significant positive. All variables are defined in Appendix B.

Panel A
Rank Rank Dividend LogMarket Book to Adjusted
Industry REC Volatility Beta
MVE BM Payout Cap Market R2
-11.0 *** 429.9 *** 2.2 *** -55.3 *** 9.4 *** 0.06
Cnsmr High Low 0
(-4.4) (4.6) (3.3) (-10.1) (5.0)
-21.2 *** -44.9 6.2 *** -40.6 *** 1.7 0.03
Cnsmr Mid Low 0
(-5.5) (-0.3) (3.7) (-5.2) (0.7)
HiTec High High 0 -49.7*** -272.1 8.6 *** -14.9 6.1 0.10
(-5.5) (-0.9) (4.1) (-1.4) (0.8)
HiTec High Low 0 -17.2 *** 376.1 *** -1.0 -34.5 *** 13.1 *** 0.07
(-4.8) (3.8) (-1.3) (-4.3) (5.3)
HiTec High Mid 0 -24.5 *** -298.7 * 0.6 -84.4 *** 6.09 0.07
(-4.5) (-1.8) (0.5) (-6.9) (1.5)
Other High Low 0 -10.9 *** 13.1 -1.7 * -88.4 *** 5.7 ** 0.09
(-3.5) (0.1) (-2.4) (-14.0) (2.8)

17

Electronic copy available at: https://ssrn.com/abstract=3016040

 Panel B
Rank Rank Dividend LogMarket Book to Adjusted
Industry REC Volatility Beta
MVE BM Payout Cap Market R2
21.3 *** 77.7 -1.8 * -35.3 *** 0.3 0.02
Manuf High Mid 0
(6.2) (0.6) (-2.0) (-4.6) (0.1)
35.0 ** 639.8 -7.8 * -12.2 -18.8 * 0.04
Manuf High Mid 1
(2.7) (1.5) (-1.9) (-0.4) (-1.9)
11.7 ** -85.7 4.3 ** 33.5 *** 2.8 0.03
Manuf Mid Low 0
(3.0) (-0.6) (2.7) (4.4) (1.3)
7.4 ** -67.7 4.0 ** -24.2 *** -0.1 0.01
Manuf Mid Mid 0
(2.5) (-0.7) (3.1) (-3.8) (-0.1)
12.8 *** 221.9 * 1.4 * -7.8 ** -5.0 ** 0.02
Other High High 0
(4.6) (2.4) (2.1) (-3.0) (-2.6)
10.0 *** -257.2 ** 1.8 -26.4 *** 8.3 *** 0.06
Other Low High 0
(3.6) (-2.7) (1.3) (-8.6) (5.0)
19.4 *** 100.2 14.2 *** 16.7 * -3.3 0.04
Other Low Low 0
(3.9) (0.5) (5.0) (1.6) (-1.1)
15.1 *** 148.0 * 5.6 *** -15.0 *** 2.6 0.03
Other Mid High 0
(6.0) (1.8) (4.9) (-5.6) (1.6)
*, **, and *** denote significant at the 10%, 5% and 1% level respectively.

18

Electronic copy available at: https://ssrn.com/abstract=3016040

 Table 5
Dividend Payout and PE Ratio by Groups: Using Trailing Earnings

This table reports regression coefficients and associated heteroscedasticity consistent t-statistics (along
with adjusted R2) of
/

run separately over 90 groups overall, made up of 5 industry groups, 3 groups of market capitalization, 3
groups of book to market values, and whether recession period. Only the specifications for which Dividend
Payout is significant at 5% or 1% levels are shown. Panel A shows significant negative relations, and Panel
B significant positive. The earnings here are trailing earnings. All variables are defined in Appendix B.

Panel A
Rank Rank Dividend LogMarket Book to Adjusted
Industry REC Volatility Beta
MVE BM Payout Cap Market R2
-7.4 *** 503.6 *** 0.8 -60.4 *** 9.2 *** 0.08
Cnsmr High Low 0
(-3.5) (6.4) (1.5) (-12.9) (5.7)
-32.8 *** 547.4 * 6.5 ** -24.9 * -18.4 *** 0.10
Cnsmr High Low 1
(-5.0) (2.2) (2.9) (-1.9) (-3.3)
-15.2 *** 162.0 * 1.6 -55.3 *** 0.9 0.05
Cnsmr Mid Low 0
(-4.8) (1.7) (1.1) (-8.4) (0.5)
-16.9 *** 854.9 *** -0.4 -36.9 *** 12.7 *** 0.13
HiTec High Low 0
(-5.0) (9.0) (-0.6) (-4.9) (5.4)
-30.1 ** 1107.8 * 8.7 * -17.4 -8.1 0.06
HiTec High Low 1
(-2.9) (2.4) (2.5) (-0.7) (-0.6)
-20.0 *** -67.1 0.3 -100.1 *** 4.8 0.08
HiTec High Mid 0
(-3.7) (-0.4) (0.2) (-8.5) (1.2)
-47.5 ** 1404.6 *** 1.6 -211.6 *** -36.1 *** 0.32
Hlth High Low 1
(-3.2) (3.7) (0.5) (-8.4) (-3.3)
-24.3 *** -482.6 *** 5.1 * -149.2 *** 2.8 0.13
Hlth Mid Low 0
(-3.4) (-3.3) (2.1) (-10.0) (0.7)

19

Electronic copy available at: https://ssrn.com/abstract=3016040

 Panel B
Rank Rank Dividend LogMarket Book to Adjusted
Industry REC Volatility Beta
MVE BM Payout Cap Market R2
Cnsmr High High 0 16.2 ** 261.4 -1.4 -27.6 *** -16.6 *** 0.07
(3.0) (1.3) (-0.9) (-4.1) (-4.1)
Cnsmr High High 1 143.0 *** 454.7 5.2 -69.7 -39.0 0.30
(4.2) (0.3) (0.4) (-1.5) (-1.4)
Cnsmr Mid Mid 0 8.5 *** -368.6 *** 0.6 -37.8 *** 6.1 *** 0.05
(3.5) (-4.7) (0.5) (-6.9) (4.0)
20.8 ** -279.3 * 0.6 -77.0 *** 1.0 0.07
Hlth Mid Mid 0
(2.6) (-1.8) (0.2) (-6.2) (0.27)
23.8 *** 95.3 -0.1 -25.6 *** 1.7 0.07
Manuf High High 0
(9.5) (0.8) (-0.1) (-8.8) (0.9)
38.1 *** 356.8 * 0.2 -34.0 * -24.1 *** 0.15
Manuf High Low 1
(4.5) (1.6) (0.1) (-2.0) (-3.7)
24.0 *** 235.8 * -2.7 *** -45.4 *** 1.3 0.04
Manuf High Mid 0
(7.5) (2.1) (-3.3) (-6.4) (0.7)
32.5 *** 534.9 * -3.0 -49.6 ** -8.8 0.07
Manuf High Mid 1
(4.2) (2.1) (-1.2) (-3.2) (-1.5)
12.0 *** -270.2 ** 5.5 *** -18.7 *** 6.1 *** 0.06
Manuf Mid High 0
(5.3) (-3.1) (4.8) (-6.9) (3.9)
29.3 *** 113.6 -13.6 ** -59.0 *** -4.1 0.08
Manuf Mid Mid 1
(3.9) (0.5) (-2.8) (-4.0) (-0.6)
21.3 *** 337.9 *** 1.0 * -0.8 -7.6 *** 0.04
Other High High 0
(8.1) (4.0) (1.6) (-0.3) (-4.2)
11.1 *** 312.5 *** -0.6 -61.0 *** -4.7 ** 0.06
Other High Mid 0
(4.5) (4.3) (-1.1) (-12.8) (-3.0)
22.7 ** -446.3 * -0.6 -52.3 ** 3.1 0.09
Other High Mid 1
(2.62) (-1.8) (-0.2) (-3.1) (0.4)
14.6 *** -249.5 ** -1.5 -23.1 *** 7.9 *** 0.06
Other Low High 0
(5.8) (-2.9) (-1.2) (-8.2) (5.3)
20.4 *** 78.1 16.0 *** 29.9 *** -3.6 0.08
Other Low Low 0
(4.7) (0.5) (6.4) (3.4) (-1.3)
22.3 *** 227.0 ** 4.3 *** -13.6 *** 0.03 0.05
Other Mid High 0
(9.9) (3.1) (4.2) (-5.6) (0.0)
*, **, and *** denote significant at the 10%, 5% and 1% level respectively.

20

Electronic copy available at: https://ssrn.com/abstract=3016040

 Table 6
Change in Dividend Payout and Change in PE Ratio

This table reports regression coefficients and the associated heteroscedasticity consistent t-statistics (along
with adjusted R2) of 4 different specifications of:

/ ∆
∆ ∆
∆
.

All variables are defined in Appendix B.

/ / / /

-69.91 *** -51.30 ** -73.33 *** -51.48 **

∆Dividend Payout (-20.81) (-2.74) (-11.44) (-2.75)

-1.90 -1.87 -1.87 -1.87

Recession (-1.76) (-1.76) (-1.76) (-1.76)

-119.95 *** -118.37 *** -118.58 *** -118.45 ***

Volatility (-4.63) (-4.58) (-4.59) (-4.59)

-0.06 -0.11 -0.11 -0.11

Log Market Cap (-0.52) (-0.93) (-0.95) (-0.93)

1.63 ** 1.56 ** 1.55 ** 1.56 **

Book to Market (2.97) (2.86) (2.83) (2.87)

0.45 0.35 0.36 0.35

Beta (0.90) (0.72) (0.73) (0.71)

∆Dividend Payout -2.01

Log Market Cap (-0.90)

∆Dividend Payout
1.32
Log Market Cap (0.99)
Book to Market
∆Dividend Payout
-1.68
Log Market Cap (-0.73)
Manufacturing

Industry Fixed Effects Yes Yes Yes Yes

Year Fixed effects Yes Yes Yes Yes

Adjusted R2 0.02 0.01 0.01 0.01

*, **, and *** denote significant at the 10%, 5% and 1% level respectively.

21

Electronic copy available at: https://ssrn.com/abstract=3016040

 Table 7
Change in Dividend Payout and Change in PE Ratio by Groups

This table reports the regression coefficients and associated heteroscedasticity consistent t-statistics (along
with adjusted R2) of

/ ∆

run separately over 90 groups overall, made up of 5 industry groups, 3 groups of market capitalization, 3 groups of
book to market values, and whether or not recession period. Only the specifications for which ∆Dividend Payout is
significant at the 5% and 1% levels are shown. All variables are defined in Appendix B.

Industr Rank Rank ∆Dividend LogMarket Book to Adjusted

REC Volatility Beta
y MVE BM Payout Cap Market R2
-77.8 *** -152.3 1.5 * 9.8 -4.9 * 0.02
Cnsmr High Low 0
(-6.2) (-1.4) (1.8) (1.5) (-2.1)
-86.6 *** -62.0 0.1 19.0 * -1.2 0.02
Cnsmr High Mid 0
(-4.3) (-0.4) (0.1) (1.7) (-0.3)
-69.5 ** -105.7 -0.3 18.1 * 1.3 0.01
Cnsmr Low Mid 0
(-3.0) (-0.7) (-0.1) (1.7) (0.5)
-100.8 *** 35.9 -0.3 11.2 0.8 0.01
Cnsmr Mid Low 0
(-4.6) (0.3) (-0.2) (1.2) (0.3)
-67.7 *** -172.9 0.5 11.1 2.8 0.01
Cnsmr Mid Mid 0
(-3.6) (-1.4) (0.3) (1.2) (1.1)
-67.6 ** -173.8 * -1.3 * 23.8 ** -1.6 0.01
HiTec High Low 0
(-3.2) (-1.8) (-1.6) (3.0) (-0.6)
-70.2** 11.5 1.6 -5.9 1.7 0.01
HiTec High Mid 0
(-2.9) (0.1) (1.4) (-0.5) (0.4)
-126.6 ** -207.4 1.9 21.7 -4.5 0.01
HiTec Low Low 0
(-2.8) (-1.2) (0.5) (1.2) (-1.4)
-126.8 *** -128.9 0.7 13.2 0.1 0.02
Hlth High Low 0
(-5.5) (-0.9) (0.8) (1.3) (0.02)
-343.4 ** 320.0 18.3 -12.4 -6.8 0.11
Hlth Mid Mid 1
(-2.9) (0.5) (1.5) (-0.4) (-0.3)
-81.0 *** -144.9 -2.3 * 4.5 6.0 * 0.03
Manuf High High 0
(-5.2) (-0.7) (-1.7) (0.7) (1.8)
-66.3 *** -353.5 *** 0.3 16.5 * 3.8 * 0.02
Manuf High Low 0
(-5.2) (-3.9) (0.4) (2.2) (2.1)
-65.5 *** -230.9 * -0.5 18.1 * 1.7 0.02
Manuf High Mid 0
(-4.2) (-1.6) (-0.6) (2.0) (0.8)
-93.6 *** -56.4 -2.3 11.0 0.712 0.02
Manuf Low Mid 0
(-4.6) (-0.4) (-1.2) (1.1) (0.3)
-114.3 *** 125.9 -2.9 1.3 1.2 0.03
Manuf Mid High 0
(-5.7) (0.7) (-1.2) (0.2) (0.4)
-50.5 ** -355.0 ** -1.4 7.9 0.71 0.01
Manuf Mid Low 0
(-2.6) (-2.7) (-0.8) (0.9) (0.3)
-72.0 *** -25.3 -0.02 21.5 ** 2.0 0.02
Manuf Mid Mid 0
(-5.1) (-0.2) (-0.0) (3.1) (1.2)
-60.7 *** -55.0 -0.5 -2.0 3.0 0.02
Other High High 0
(-5.9) (-0.6) (-0.8) (-0.8) (1.6)
22

Electronic copy available at: https://ssrn.com/abstract=3016040

 -83.4 *** -156.9 -0.8 6.7 -0.1 0.02
Other High Low 0
(-5.2) (-1.5) (-1.1) (1.0) (-0.0)
-71.6 *** -98.1 -0.6 15.5 ** 0.6 0.02
Other High Mid 0
(-6.0) (-1.3) (-1.1) (3.1) (0.4)
-70.9 *** -28.2 1.1 -5.0 0.2 0.01
Other Low High 0
(-5.1) (-0.3) (0.7) (-1.5) (0.1)
-53.9 *** -78.8 0.4 5.2 * -3.0 * 0.01
Other Mid High 0
(-4.6) (-0.9) (0.4) (1.8) (-1.8)
-81.7 *** -140.9 -1.8 -1.3 -0.5 0.01
Other Mid Low 0
(-3.6) (-1.2) (-0.9) (-0.1) (-0.2)
-50.6 *** -174.7 * -1.2 12.8 * 1.0 0.01
Other Mid Mid 0
(-3.5) (-2.3) (-1.1) (2.5) (0.7)
*, **, and *** denote significant at the 10%, 5% and 1% level respectively.

23

Electronic copy available at: https://ssrn.com/abstract=3016040

 Table 8
Change in Dividend Payout and Change in Volatility

This table reports the regression coefficients and the associated heteroscedasticity consistent t-statistics
(along with adjusted R2) of:

∆

All variables are defined in Appendix B.

0.001***
∆Dividend Payout (4.0)

0.001***
Log Market Cap (4.6)

-0.001
Book to Market (-1.4)

-0.001***
Beta (-22.2)

Adjusted R2 0.08

*, **, and *** denote significant at the 10%, 5% and 1% level respectively.

24

Electronic copy available at: https://ssrn.com/abstract=3016040

 Appendix A
Dividend Payout and PE Ratio

A basic derivation (see, for example, Professor A. Damodaran’s online teaching notes
http://pages.stern.nyu.edu/~adamodar/New_Home_Page/invfables/peratio.htm) shows that
if the PE ratio is stated in terms of expected earnings in the next time period, then

So, PE ratio is an increasing function of the payout ratio and the growth rate and a decreasing
function of firm risk. We can state the payout ratio as a function of the expected growth rate and
return on equity.

Payout ratio

Substituting back into the equation above,

The price-earnings ratio for a high growth firm can also be related to fundamentals. In the special
case of the two-stage dividend discount model, this relationship can be made explicit fairly
simply. When a firm is expected to be in high growth for the next n years and stable growth
thereafter, the dividend discount model can be written as follows:

 (1+g) n 
EPS0 Payout Ratio(1 +g)
1

 (1+k e,hg ) n 
 EPS0 Payout Ration (1 +g) (1 + gn )
n

P0  + n
k e,hg - g (k e,st - gn )(1 + k e,hg )
where,
EPS0 = Earnings per share in year 0 (Current year)

g = Growth rate in the first n years

ke,hg = Cost of equity in high growth period
ke,st = Cost of equity in stable growth period
Payout = Payout ratio in the first n years
gn = Growth rate after n years forever (Stable growth rate)

Payout Ration = Payout ratio after n years for the stable firm
25

Electronic copy available at: https://ssrn.com/abstract=3016040

Divide both sides of the equation by EPS0:

Here again, we can substitute in the fundamental equation for payout ratios.
   
1- g 1 +g1 (1+g) n  
n
1- g n  1+gn 1+g n 
P0  ROE hg   (1+k e,hg )   ROE st 
= + n
EPS0 k e,hg - g (k e,st - gn )(1+k e,hg )

where ROEhg is the return on equity in the high growth period and ROEst is the return on
equity, from which the implications arise.

26

Electronic copy available at: https://ssrn.com/abstract=3016040

 Appendix B
Variables Descriptions

Variable Description Formula/ Source

Income Before Extraordinary Compustat
IBADJQ Items - Adjusted for Common
Stock Equivalents
NIQ Net Income (Loss) Compustat
ATQ Assets - Total Compustat
Common/Ordinary Equity - Compustat
CEQQ
Total
LTQ Liabilities - Total Compustat
Preferred/Preference Stock Compustat
PSTKQ
(Capital) – Total
SEQQ Stockholders Equity Quarterly Compustat
Deferred Taxes and Investment Compustat
TXDITCQ
Tax Credit
DVY Cash Dividends Compustat
DVPSPQ Dividends per Share– Quarter Compustat
PRCCQ Price Close – Quarter Compustat
CSHOQ Common Shares Outstanding Compustat
= SEQQ + TXDITCQ - PSTKQ,
BEQ Book Value of Equity If SEQQ missing, SEQQ = CEQQ + PSTKQ,
If also missing, SEQQ = ATQ - LTQ
DIV Cash Dividend = max(Delta(DVY), DVPSPQ * CSHOQ)
Previous one year of the DATADATE of stock
Volatility Past Volatility
return standard deviation
Previous one year daily beta by running: RET –
Beta Daily past Beta
RF = alpha + beta * MKTRF
EPS0 Earnings per Share = NIQ / CSHOQ
EPSt Trailing EPS = avg(NIQ) / CSHOQ for past four quarters
PE0 Price/Earnings Ratio = PRCCQ / EPS0
PEt Price/Trailing Earnings Ratio = PRCCQ / EPSt
Dividend = DIV / IBADJQ if IBADJQ > 0
Dividend Payout Ratio
Payout
PE01 P/E Ratio for t+1 = lead(PE0)
PE02 P/E Ratio for t+2 = lead(PE01)
PEt1 Trailing P/E Ratio for t+1 = lead(PEt0)
PEt2 Trailing P/E Ratio for t+2 = lead(PEt1)

27

Electronic copy available at: https://ssrn.com/abstract=3016040

 DPR_1 Dividend Payout for t-1 = lag(DPR)
Fama French 5 Industry C Cnsmr, HiTec, Hlth, Manuf, Other
Industry
classfication
MVE Market Value of Equity = PRCCQ * CSHOQ
BMR Book to Market Ratio = BEQ / MVE
Divide each quarter’s MVE into 3 equal groups:
Rank MVE Ranking by MVE
“High”, “Mid”, “Low”
Divide each quarter’s BMR into 3 equal groups:
Rank BMR Ranking by BMR
“High”, “Mid”, “Low”

28

Electronic copy available at: https://ssrn.com/abstract=3016040