UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGM Digitized by the Internet Archive in 2011 with funding from University of Illinois Urbana-Champaign http://www.archive.org/details/allocationsofper726leec Faculty Working Papers College of Commerce and Business Administration University of tllinoit at Urbana-Champaign FACULTY WOPKING PAPERS College of Commerce and Business Administration University of Illinois at Urbana-Champaign November 4, 1980 ALLOCATIONS OF PERMANENT AND TRANSITORY EARNINGS BEIli/EEN RETAINED EARNINGS AND DIVIDEND PAY>ENTS Cheng F. Lee, Professor, Department of Finance Walter J. Primeaux, Jr., Professor, Department of Business Administration #726 Summary Based upon the pernianent incoine hypothesis theory developed by Friedman, accounting earnings are decomposed into two components, i.e., Che permanent component and the transitory component. It is shown that the allocation of earnings between retained earnings and dividends pay- ments may be in accordance with either permanent earnings or current earnings. This theory is then used to test whether the dividend payments decision of the electric utility industry is in terms of the "information content" or the "partial adjustment" hypothesis. Presentation Paper presented at the Southern Financial Association Meeting at Washington, DC, November 5-7, 1980. I. Introduction Earnings of a firm are allocated between retained earnings and dividends by a financial decision. Retained earnings are internal sources of funds which provide additional financial capital which may be used either for expansion or as a financial reserve against future contingencies; dividends are generally distributed to stockholders to satisfy their need for liquidity or for other uses according to their preference functions. It is well-known that earnings of a firm can be classified into either a permanent component or a transitory component. A firm's permanent earning power creates the permanent component and the transitory component is composed of income of temporary nature. Modigliani and Miller (1958, 1961, 1963, 1966) have argued that a firm's market value is determined by its e:qjected (or permanent) earnings, not its transitory component of income. The transitory component of a firm's earnings originates from a temporar>' change in market conditions, a temporary change in accounting method or any other non-permanent change which would cause earnings to fluctuate over tine. Several practical methods exist to determine a firm's dividend policy. [See Ueston and Erigham (1973)]. Theoretically, finance schol- ars have attempted to explain a firm's dividend payment behavior in terms of three different hypotheses: (1) information content, (2) par- tial adjustment or (3) the residual theory. It is well known from the finance literature that dividend policy can affect a firm's internal source of funds and cost of capiual. -2- In addition, the forecasting of dividends is of importance to the security analyst. Therefore, the allocations of earnings between re- tained earnings and dividend payments are generally a serious concern of financial managers. The main purposes of this paper are (1) to develop some theories to explain how firms generally allocate permanent earnings and transi- tory earnings between dividends payments and retained earnings and (2) to develop a method for decomposing the current earnings into permanent and transitory components. The implications of each of these income components for a firm's dividend policy and payments decision are also developed. The first section is the introduction. The second section modi- fies Friedman's (1957) permanent income hypothesis to describe the role of permanent earnings and transitory earnings in the dividend determina- tion process. The relationship between accountings earnings and economic earnings are also discussed. The third section employs models to decom- pose the current earnings into permanent and transitory components de- fined according to the methods proposed by Darby (1972, 197A). The fourth section, uses disaggregated earnings and dividends data of the electric utility industry to determine whether permanent earnings or current earnings data should be used to describe dividend payment be- havior in that business. The final section summarizes the results and provides some concluding remarks. -3- II. Theoretical Deteraination of Firm's Feraanent and Transitor\' Earnings In the development of the consumption function, which is one of the key concepts in Keynesian economics, several important theories were developed to explain how consumers adjust consumption expenditures to accommodate changes in their levels of income. One of these theories is the Permanent Income Hypothesis developed by Milton Friedman (1957). The Permanent Income Hypothesis explains that consumption is not a function of current income but a function of permanent income. Total income, Y, is composed of tvo components, Y + Y , where Y is permanent income and Y is transitory income. Transitory income is not fully anticipated and it may be positive or negative. That is, a prize would constitute a positive transitory income component while a loss of income from temporary illness or layoff would constitute a negative component of permanent income. Friedman explains that these transitory elements would not affect consumption expenditures. The Permanent Income Hypothesis is readily adaptable to finance theory and a new theory of dividend pa>'ments by business can be devel- oped. The income of interest here is the income of the business firm and dividends are analogous to consumer consiimption e:q)enditures. The level of permanent income earned by a firm determines the per- manent dividends it can pay out to stockholders. Permanent income is essentially an average of current, past, and future earnings of the firm. Current income is divided into two comconents : When Friedman received the ^Icbel prize in economics, this work was cited as cne of his major contributions. -4- Y = Y + Y P t where: Y = current income of the firm Y = permanent income of the firm Y^ = transitory income of the firm Transitory income may be postive or negative and current income will differ from permanent income by the amount of transitory income. A business earns transitory income, which is really unanticipated earn- ings, from windfall profits from any source. For example, oil companies are now earning transitory income from the increase price they receive from selling products made from crude oil produced domestically. Firms incur negative transitory income if they experience an uninsured cata- strophic event such as the destruction of a plant by a disaster of any kind or an unexpected strike by employees. The transitory components of income, positive and negative, should cancel out over the permanent income time horizon. Transitory' components, however, are always present during shorter time periods. Professor Eisner (1967, 1978) has developed a permanent income theory for investment decision. If a firm investment essentially de- pends upon interr^l sources of funds, then the nature of retained earn- ings is an important factor affecting the decision to undertake long- term or short-term investment. Retained earnings can conceptually be decomposed into two components, i.e. periranent and transitory components. Dividends can also be divided into two components: permanent divi- dends and transitory dividends: D = D + D^ -5- where: D = current dividends paid by the firm. D = permanent dividends paid by the firm. d|_ = transitory dividends paid by the firm. Permanent dividends are only one component of dividends and total dividends may be larger than permanent dividends, depending upon the level of transitory dividends. Permanent dividends are dividends which the business firm systematically pays based on its permanent earnings. All income is either paid out in dividends or retained by the business in the form of retained earnings. Y = Y + Y P t Y - (Dp+D^) - E^ = where: Y = current income of the firm. Y = permanent income of the firm. Y^ = transitory income of the firm. D = permanent consumption of the firm. D = transitory consumption of the firm. E" = retained earnings of the firm. Y and D are "random" or "chance" variations in income and dividends. The existence of transitcry dividends will depend upon the finan- cial manager's use of either (1) information content, (2) partial adjust- ment or (3) the residual theory to determine their firms' dividend pay- ments over time. If either the "partial adjustment" or the "residual theory" is used to determine the dividend pajTaent behavior then transi- tory dividends are not independent of transitcry income. A transitory decline in income does generally cause transitory dividends to decline. Siir.ilarly, a transitory increase ir income does cause dividend pa>Tnents to increase. Retained earnings may also increase when unexpected changes in inccire take place. Windfall income would be paid out as dividends, or be kept as retained earnings. -6- If the "information content" is used to determine the dividend pay- ment behavior, then most of transitory income will become retained earn- ings instead of dividends. In sum, whether current earnings or permanent earnings should be used to determine a firm's dividend behavior will de- pend upon whether or not a transitory dividend component exists. This issue will be empirically tested in the fifth section of this paper. III. Models for Decomposing Current Earnings Into Permanent and Transitory Earnings Components The models used to compute permanent income as proposed by Friedman (1957) can be classified into the traditional approach and Darby's (1974) modified unbiased method. The modified method can be defined as (3.1) Yp^ = 8Y^ + (1 - B)(l + C)Ypj._^ where Y and Y , are permanent income in period t and t-1 resDec- pt pt-1 tively; Y is the current income in period t; 6 is the adjustment coefficient and C is the trend rate of income growth. To estimate the permanent income series, we need £, C and Y . po Darby (19 7A) has shown that the unbiased weight of current income in the determination of permanent income of about .10 en an annual basis and .025 on a quarterly basis. The initial value Y and trend rate C can po be taken from estimating the income trend regression (3.2) logY_ = a, -f a„t + u^ After a, and a„ are estimated, the Y and C can be defined as ^ ^ ♦ po (3.3) Y = el and po c = a^ Note that this is only one of several methods to estimate C and Y po The estimated Y and C can be used in equation (3.1) to repeatedly estimate Y , It should also be noted that estimated a. is the pt 2 earnings growth rate estimate. Both quarterly and annual earnings and dividend data from fort^/- two electric utility firms were used to do the empirical investigations. The operating data covered the period of 1962-1978. 2 IV. Current Earnings, Permanent Earnings and Investment Analysis Accounting earnings contain a transitory component which dees not represent the true earning power of the firm. Hence, the trans- itory component of earnings should not be used to determine the business' future value. Security analysists of Value Line have generally used only the permanent component of earnings to forecast the e:cpected future market value of common stock. Mcdigliani and Miller (1958, 1961, 1963, 1966) [M&M] have shown that expected earnings should be used instead of current earnings to determine the value of a firm. In estimating the cost of capital for the utility industry, M&M (1966) used the instru- mental variable approach to remove the trar^itory component associated with current earnings. One difficulty of using the instrumental variable 2 Seasonal components were removed by using X-11 multiplicate decomposing method which was developed by the Department of Commerce. -8- approach involves the selecticn of the appropriate explanatory variables for specifying the regression equation. A more desirable approach for determining the percanent ccnponent of earnings was previously set out in section III. To estimate permanent income, we should estimate the initial value of permanent income and the trend rate of income growth. The exact pro- cedures used to develop these estimations are described in equations (3.2) and (3.3). After these equations are estimated, they may be used to estimate either annual or quarterly permanent income. The weights used to estimate the annual and quarterly permanent earnings are .10 and .025, respectively as suggested by Darby (197*^). The growth rates of both annual and quarterly earnings for firms in the sample are presented in Table 1. Th.e current and permanent earn- ings developed from quarterly data are shown in Table 2. This table in- dicates thiat the permanent earnings per share are always smaller than current earnings per share. The coefficients of variation for both current and permanent earnings were calculated to investigate the degree of fluctuation of current earnings per share compared with permanent earnings per share. These coefficients are presented in Table 3. The results show that the coefficient of variation for permanent earnings is smaller than that statistic for current earnings in most of the cases. The coefficient of variation was also calculated to examine the variation of dividends per share. These results, presented in col- umn 3 oT Table 3, show chat permanent earnings per share is generally less volatile than current earnings oer share or dividends ser share. -9- V. Current Earnings, Permanent Earnings and Dividend Payment Behavior Dividend payment decision theory and practice is one of the most important topics for study by finance scholars. Lintner (1956), Fama and Babiak (1968) and others have defined the dividend payment equation as: (5.1) D^ - D^_^ = a^ + a^(Dj^ - D^^^_^) + u^^ (A) and ht = '^i^it ^^^ where D and D ^ are dividend per share for i firm in t and t-1 period respectively; D. is the target dividends for 1 firm i » t in period t and a. is the "partial adjustment coefficient." Substitut- ing (4.1.B) into (Tnent behavior; only 17 of o 42 have higher R*" for permanent income V7ith quarterly data. These results imply that current earnings are generally used to determine dividend payment behavior. In other words, for firms in this sample, either the partial adjustment or the residual theory h3rpothesis provides a more suitable explanation of dividend payment behavior than the information content hypothesis. It is well-known and accepted that utility stocks are income instead of growth securities. The EPS growth rate estimates for the 42 firms in the sample can be found in Table 1. Table 1 indicates that the estimated growth rates for EPS are below 4,''o. A manager of a utility firm may be inclined to use transitory components of earnings (mentioned in Section II) to pay a transitory type of divi- dend to make his firm's dividend pa>-nent comparable to similar firms in -12- the industry. These results may not necessarily apply to firms in growth industries. VI. Summary and Concluding Remarks Milton Friedman (1957) presented a Permanent Income Hypothesis. This study uses Friedman's basic concepts of current earnings, permanent earnings and transitory earnings and examiines how well they explain dividend payment behavior of the 42 electric utility firms in the sample. Earnings per share data (both annual and quarterly) were used in the analysis. The procedure employed to decompose the current earnings into trarisitory and permanent components was suggested by Darby (1972, 1974). The possible implications of the permanent component of earnings on security analysis were examined; then, the effect of the permanent earnings component on the dividend payment behavior of firms in the sample was tested. The results show that current rather than permanent income tends to describe more accurately the dividend payment behavior of firms in the sample. This unexpected result was possibly caused by the fact that the sample consisted of utility firms, which are high dividend payout firms and not growth firms. The results may be differ- ent for firms from another industry. The results also show that the technique employed here provides a very satisfactory approach to testing whether "partial adjustment," "residual theory," or "information content" is used by a firm to deter- mine its dividend payout policy. In estimating the cost of capital for the electric utility indus- try, M&K (1966, 356-353) have used the some kinds of moving average -13- methcds tc remove the transitory cciLponents cf accounting reported earn- ings. Howe\-er, they are unable to obtain satisfactory results. The permanent earnings estimation method developed in this paper may well be used to improve the quality of H&M's cost of capital estimates. _1A_ Conpany 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 TABLE 1 Growth Psace of EP? Quarterly Annualized Quarterly Annual Growth Rate Growth Rate Growth Rate .006 .024 .026 .003 .012 .015 .0003 .001 .001 .005 .020 .020 -.002 -.008 -.007 .004 .016 .018 .004 .016 .016 .01 .04 .040 .007 .028 .029 .005 .020 .024 -.001 -.004 .003 .004 .016 .018 .001 .004 .007 .007 .023 .032 .008 .032 .034 .009 .036 .037 .012 .048 .050 -.005 -.020 -.018 .010 .040 .041 .0005 .002 .002 -.001 -.004 -.004 .012 .043 .055 .007 .028 .029 .001 .OOA .005 .003 .012 .011 -.005 -.020 -.020 .004 .016 .017 .003 .012 .016 .0003 .001 -.0002 .004 .016 .021 .005 .020 .023 .005 .020 .028 .005 .020 .027 -.005 -.020 -.022 .009 .036 .039 .004 .016 .013 .001 .004 .005 .006 .024 .024 .016 .064 .069 .005 .020 .02i .007 .028 .029 .002 .008 .009 -15- TABLE 2 Average Current and Permanent Earrings and Dividends per Share (quarterly data) Current Earnings Dividends Permanent Earnings per share per share per share 1 0.49387 0.34785 0.43275 2 0.57737 0.38310 0.52548 3 0.55196 0.36656 0.54472 4 0.73413 0.50118 0.65330 5 0.68269 0.46851 0.701C2 6 0.83226 0.39450 0.75140 7 0.518110 0.3S384 0.47A46 8 0.70331 0.33084 0.55592 9 0.55421 0.38637 0. 47309 10 0.54139 0.34991 0.48056 11 0.58A63 0.38919 0.58A46 12 0.44675 0.27628 0.40365 13 0.57A22 0.36162 0.54018 lA 0.40681 0.29201 0.34557 15 0.59650 0.40290 0.50057 16 0.70400 0.49206 0.58056 17 0.48173 0.26100 0.37258 18 0.43343 0.34241 0.53174 19 0.62257 0.40685 0.50654 20 0.41722 0.31919 0.40787 21 0.47922 0.29879 0.48704 22 0.49196 0.23194 0.36920 23 0.55596 0.36897 0.47802 24 0.38419 0.27865 0.36947 25 0.52712 0.35841 0.48114 26 0.45216 0.30726 0.49576 27 0.60113 0.47099 0.54541 2S 0,39019 0.31472 0.36071 29 0.55260 0.37997 0.53562 30 0.63310 0.43019 0.56303 31 0.51221 0.33619 0.45031 32 0.56134 0.37757 0.49605 33 0.5S912 0.A0109 0.51827 34 0.46776 0.32874 0.51417 35 0.59575 0.42-^53 0.48795 36 0.42866 0.27529 0.39001 0.54956 0.29300 0.52751 38 0.34457 0.21019 0.29855 39 0.31047 0.18746 0.21757 AC 0.52268 0.36900 0.45562 41 0.61053 0.39150 0.50654 42 0.46S38 0.30557 0.43672 -16- TABLE 3 Coefficients of Variation cf Current and Permanent Earnings and Dividends Per Share Current Earnings /Share 1 0.20837 2 0.2A379 3 0.22456 4 0.23473 5 0.26996 6 0.26162 7 0.12681 8 0.33624 9 0.26291 10 0.22895 11 0.27180 12 0.22209 13 0.29440 14 0.21135 15 0.20491 16 0.23041 17 0.30325 18 0.22847 19 0.22078 20 0.19527 21 0.17451 22 0.34602 23 0.21102 24 0.21945 25 0.25065 26 0.22437 27 0.18357 28 0.16154 29 0.25670 30 0.22180 31 0.21042 32 0.22854 33 0.22289 34 0.29389 35 0.22041 36 0.23049 37 0.17858 38 0.24547 39 0.36097 40 0.31074 41 0.20408 42 C. 29038 a/X Dividend/Share 0.12698 0.42785 0,22763 0.22178 0.16693 0.20441 0.11119 C. 28482 0.18335 0.18765 0.18718 0.19274 0.20663 0.214S6 0.11000 0.20117 0.28870 0.16889 0.50178 0.14913 0.08431 0.23118 0.17191 0.18216 0.19966 0.54133 0.19784 0.11003 0.17109 0.47388 0.44064 0.24811 0.45057 0.24661 0.17681 0.40089 0.4616/. 0.18649 0.3170S 0.21260 0.23^51 0.19001 Permanent Earnings/Share 0.08268 0.04217 0.05028 0.07594 0.03747 0.07328 0.04974 0.11428 0.09785 0.77776 0.02448 0.04692 0.02860 0.09584 0.10302 0.13351 0.16396 0.05772 0.14366 0.01493 0.01511 0.16639 0.08918 0.02712 0.0449 3 0.64985 0.06375 0.05134 0.02584 0.05174 0.06631 0.09 711 0.09233 0.08464 0.12009 0.06731 0.02753 0.06615 0.22903 0. 06233 C. 084 81 0.35263 -17- TABLE 4 Empirical Results for Equations (5.2) and (5.4) (Annual Data) Conpany ^O'^O b,,b, ^2'^2 Adj R" DW Atlantic City (i) 0.14718 0.24358 -0.44622 0.3384 1.538 Electric (0.798) (3.071)** (-2.452)* (ii) -0.27607 (-1.195) 0.68376 (2.890)* -0.65174 (-2.666)* 0.3051 1.635 Carolina Fewer (i) 0.93647 0.59704 -1.50640 0.6353 2.085 & Light (2.434)* (3.039)** (-5.290)** (ii) -2.79640 (-1.606) 2.30572 (2.508)* -1.36862 (-4.757)** 0.5796 1.874 Central & (i) -0.01839 0.61186 0.90862 0.9026 0.914 Southwest Ccrp (-0.210) (10.835)** (-11.365)** (ii) 1.9F571 (1.801) -0.86103 (-1.480) -0.05667 (-0.279) 0.1635 1.855 Cleveland (i) -0.179 33 0.48863 -0.62296 0.6886 1.333 Electric Ilium (-0.821) (5.513)** (-5.188)** (ii) -0.46878 (-0.301) 0.43305 (0.572) -0,40792 (-1.031) 0.000 0.725 Colunbus & So. (i) 0.28056 0.10^59 -0.29866 0.0510 1.351 Ohio (0.711) (1.122) (-1.488) (ii) 2.18420 (1.603) -0.64407 (-1.314) -0.20046 (-1.036) 0.0812 1.889 Florida Power (i) 0.32212 0.27172 -0.76830 0.4429 1.294 & Light (1.255) (3.118)** (-3.620)** (ii) 0.70896 (0.896) -0.06344 (-0.223) -0.29981 (-1.335) 0.0298 1.328 General Public (i) 0.12021 0.09087 -0.18038 0.2789 1.818 Utilities (1.456) (1.603) (-2.733)* (ii) -0.07335 (-0.371) 0.24504 (1.443) -0.23639 (-2.292)* 0.2556 1.557 Houston (i) 0.06836 0.35221 -0.79 344 0.7443 1.557 Industries (C.560) (6.711)** (-5.591)** (ii) -1.43803 (-6.319)** 1.189 37 (7.784)** -0.97669 (-6.361)** 0.7984 0.713 Indianapolis (i) -0.01259 0.C8403 -0.08496 0.0102 1.795 Power & Light r-0.C94) (1.462) (-0.790) (ii) -0.57442 (-2.180)* 0.73224 (2.546)* -C.520CC (-2.345)* 0.2311 1.262 -18- Company ^O'^O \'h b.,b2 Adj R~ DW Kansas Gas & (i) -0.05200 0.06949 -0.04040 0.000 1.534 Electric (-0.495) (1.108) ( :-0.441) (ii) -0.55517 (-1.566) 0.55189 (1.628) ( -0.34085 '-1.443) 0.0549 1.093 Kentucky (i) -0.01991 0.19640 -0.27233 0.1278 1.168 Utilities (-C.053) (1.617) ( :-1.47) (ii) -4.17654 (-2.030) 2.32258 (2.228)* ( -0.73138 -2.590)* 0.2418 0.697 Middle South (i) 0.00515 0.43082 -0.69951 0.6223 1.284 Utilities (0.036) (4.915)** ( -4.357)** (ii) -1.55803 (-1.784) 1.44680 (2.112) ( -0.70600 -2.346) 0.1961 1.470 Minnesota Povver (i) 0.06892 0.35088 -0.60541 0.4963 1.636 & Light (0.401) (3.782)** ( :-3,779)** (ii) 1.20555 (0.60^) -0.50879 (-0.479) ( -0.07031 ' -0.251) 0.000 1.341 Cklahoraa Gas & (i) -0.144 78 0.70933 -0.35885 0.8262 1.339 Electric (-1.600) (8.332)** ( :-7.7l4)** (ii) -1.13522 (-3.802)** 1.67471 (4. 820)** ( -1.02827 :-4.832)** 0.6047 0.199 Pennsylvania (i) 0.16581 0.10737 -0.24425 0.7059 2.213 Power & Light (4.076)** (6.164)** ( :-5.393)** (ii) 0.06942 (1.411) 0.39650 (3.676)** ( -0.52495 :-3.593)** 0.4343 1.865 Public Service (i) 0.24956 0.30549 -0.55299 0.2613 1.670 Co. of Indiana (1.262) (2.195)* :-2.679)* (ii) -0.38134 (1.147) -O.OA035 (-0.162) -0.04035 :-0.588) 0.000 1.530 Public Service (i) -0.10396 0.01689 0.13289 0.4899 1.979 Co. of New (-2.420)* (0.485) (1.925) Mexico (ii) -0.18537 (-2.209)* 0.16389 (1.180) ( -0.00175 :-0.012) 0.5309 1.866 Southern Company (i) 0.25131 (0.930) 0.35008 (3.155)** ( -0.67704 '-3.557)** 0.492^ 1.568 (ii) 0.74846 (0.814) -0.07073 (-0.150) ( -0.43937 :-1.733) 0.1054 1.347 Toledc Edison Co . (i) -0.15 387 (-0.2S7) 0.57669 (1.603) -0.76781 :-2.341)* C.2195 2.684 (ii) -1.5S69S (-2.505)* 1.77835 (3.638)** ( -1.2A526 :-4.35£)** 0.5368 2.170 -19- Company ^O'^O b,,b, b2,b2 Adj R DW Union Electric (i) 0.63968 (3.156)^=* 0.27227 (2.652)* -0.87323 (-7.461)** 0.7368 0.7779 (ii) 2.Cf078 (1.118) -0.70073 (-0.578) -0.75968 (-4.432)** 0,6797 0.376 Virginia (i) 0.4S854 0,21120 -0.75203 0.5327 1.118 Electric & Power (1.514) (1.928) (-3.803)** (ii) 1.66025 (1.368) -0.41659 (-0.630) -C. 71723 (-3.008)** 0.4168 1.046 Arizona Public (i) -0.01374 0.11515 -0.14685 0.3439 2.132 Service Co. (-0.176) (2.071) (-0.966) (ii) -0.20923 (-3.310)** 0.71108 (3.573)** -0.739S6 (-2.939)* 0.5598 1.779 Central Hudson (i) 0.02414 0.04293 -0.04536 0.000 1.317 Gas & Electric (0.316) (1.001) (-0.647) (ii) -0.41060 (-2.200)* 0.53072 (2.605)* -C. 38809 (-2.482)* 0.2423 1.241 Central Illinois (i) -0.35901 0.50655 -0.36 772 0.4564 1.527 Public Service (-1.667) (3.544)** (-2.655)* (ii) -4.71264 (-2.862)* 3.86849 (3.003)** -0.98761 (-3.276)** 0.3690 0.568 Cincinnati Gas (i) -0.20511 0.29504 ^.27274 0.1846 0.843 & Elec. (-0.882) (2.268)* (1.718) (ii) -3.33704 (-1.853) 2.09805 (1.926) -0.54990 (1.926) 0.1148 0.344 Del Marva Power (i) 0.10925 0.61686 -0.989C8 0.5783 1.857 & Light (0.4A9) (3.672)** (-4.672)** (ii) -0.12543 (-0.141) 0.44269 (0.870) -0.60517 (-2.309)* 0.1880 1.158 Illinois Power (i) -0.27485 0.29615 -^.21066 0.3022 2.505 Co. (-1.380) (2.789)* (-2.437)* (ii) -1.84924 (-1.672) 1.29883 (1.829) -0.52893 (-1.965) 0.1128 1.181 Interstate Power (i) 0.05734 0.11399 -0.166 70 0.3485 2.455 Co. (1.103) (2.145) (-3.125)** (ii) -0.39783 (-2.415)* 0.71653 (3.252)'-* -0.48919 (-3.763)** 0.5136 2.357 Iowa mine is (i) 0.2S695 0.28887 -0.6S313 0.1962 1.579 Gas & Elec. (1.355) (1.73A (-2.377)* (ii) 2.28177 (1.703) -0.93029 (-1.395) -0.14974 (-0.671) 0.139 3 1.863 -20- Company ^O'^O h'h b^.b^ Adj R" DW Iowa Power i (i) 1.09078 0.85847 -1.92025 0.7999 2.016 Light (3.342)** (4.769)** (-7.743)** (ii) -2.14387 (1.974) 2.21840 (3.810)** -1.710i3 (-6.625)** 0.7A00 1.946 Long Island (i) -0.09439 0.74468 -1.06153 0.6836 1.686 Lightina (-C.533) (5.369)** (-5.692)** (ii) -1.43599 (-2.232)* 1.40953 (2.941)* -0.02154 (-3.343)** 0.3888 1.011 Louisville Gas (i) 0.12258 0.03250 -0.08522 0.6230 0.795 & Electric (4.323)** (1.8^3) (-4.659)** (ii) 0.44A56 (3.540)** -0.24873 (-2.281)* 0.08091 (1.280) 0,6604 0.833 Montana Power (i) 0.21066 0.05067 -0.17859 0.1377 1.502 Co. (2.089) (0.94-^) (-1.960) (ii) 0.16332 (1.0A3) 0.08137 (0.572) -0.17883 (-1.399) 0.1012 1.660 Niagra Mohawk (i) 0.11531 0.59780 -0.94373 0.6653 2.153 Power (0.942) (4.765)** (-5.638)** (ii) 0.75206 (1.396) -0.29468 (-0.877) -0.13442 (-0.681) 0.1319 1.585 Northern States (i) 0.49346 0.41998 -0.87523 0.3655 2.480 Power (2.211)** (2.857)** (-3.262)** (ii) 0.00911 (0.036) 0.79841 (3.049)** -0.92190 (3.450)** 0.3977 2.103 Public Service (i) 0.36820 C. 32198 -0.82923 0.2671 1.895 Co of Colo. (0.928) (1.410) (-2.713)* (ii) -1.73234 (-1.933) 1.87563 (2.862)* -1.1954 7 (-3.96A)** 0.4817 1.706 Rochester Gas & (i) -0.26109 0.1^128 -0.02883 0.1580 1.699 Electric (-1.508) (2.147) (-0.223) (ii) -3.00923 (-3.167)** 1.74874 (3.145)** -0.63107 (-2.592)* 0.3522 1.098 Sierra Pacific (i) -C.02A72 0.35113 -0.54244 0.7189 2.681 Power Co. (-0.276) (6.288)** (-4.271)** (ii) -O.S954i (-2.78^)* 1.18153 (3.361)** -0.62486 (-2.828)* 0.3919 1.631 Tucson Gas & (i) -0.C6986 -0.01691 G. 20611 0.4567 1.866 Electric (-1.9£7) (-0.250) (1.426) (ii) -0.08724 (-2.106) 0.17213 (0.790) -G. 01779 (0.073) 0.4791 1.554 -21- Ccmpany ^O'^O Washington Water (i) 0.23643 Power (1.5SA) (ii) -1.12299 (-1.912) Wisconsin (i) 0.05044 Electric Power (0.293) (ii) -1.96150 (-4.556)*^ Wisconsin Public (i) 0.210C5 Service (1.521) (ii) 0.63600 (0.661) \'h ^2 '^2 0.37334 (5.573)** 1.0^221 (2.792)* -0.70077 (5.633)** -0.54064 (-3.153)** 0.43419 (4.137)** 1.56294 (5.229)** -0.70348 (-3.820)** -0.82728 (-4.877)** 0.41973 (4.481)** -0.29979 (-0.476) -0.81197 (-4.359)** -C. 08169 (-0.356) Adj R 0.7125 0.3910 0.5128 0.6364 0.5693 0.0000 2.306 1.69^ 1.087 0.260 0,653 1.188 (i) represents coefficients for regression equations using current earnings (Equation 5.2). (ii) represents coefficients for regression equations using permanent earnings (Equation 5.4) . * denotes significance at 5% level. ** denotes significance at If' level. -22- TABLE Empirical Results for Equation (5.2) ar.d (5.4) (Quarterly Eata) Company (i) ^O'^O 0.01602 b,,b, 0.04688 ^2 '^2 -0.10878 Adj R~ DW Atlantic City 0.0370 2.084 Electric (1.038) (1.959) (-1.896) (ii) -0. 02427 (-1.023) 0.19072 (2.127)* -0.16412 (2.212)* 0.0466 1.99A Carolina Power (i) 0.31^72 0.51354 -1.60342 0.8057 1.830 & Light (4.525)** (4.709)** (-16.A52)** (ii) -0.61575 (1.556) 2.31855 (3.051)** -1.58416 (-1^.982)** 0.7716 1,814 Central & (i) 0.00196 0.41342 -0,62635 0.5580 1,6^4 Scuthv.-est Corp rc.i€2) (8.S41)** (-9,098)** (ii) 0.17066 (2.275)* -0.272S5 (-1.902) -0,05668 (1.206) 0.0707 2.119 Cleveland (i) 0.00310 0.17135 -0,25740 0.2618 1.878 Electric Illur (0.152) (4.754)** (-4,615)** (ii) 0.05649 (0.743) -0.05115 (-0.345) -0,04700 (-0,715) 0.0029 2.235 Columbus & So. (i) 0.02990 0,05794 -0,14607 0.0911 1.880 Ohio (1.080) (2.320)* (-2,460)* (ii) 0.21216 (1.735) -0.23317 (-1.376) -0,10204 (-1.733) 0.0429 2.032 Florida Power (i) 0.03348 0.09233 -0.27282 0.1785 1.856 L Light (1.614) (3.550)** (-3.348)** (ii) 0.07202 (1.161) -0.04223 (-0.508) -0.09343 (-1.629) 0.0208 2.143 General Public (i) 0.01042 0.00123 -0.C2304 0.0002 2.330 Utilities (1.656) (0.09S) (-1.188) (ii) -0.00926 (-0.508) 0.06968 (1.154) -0.05640 (-1.679) 0.0204 2.299 Houston (i) 0.01633 0.14020 -0.34370 0.2608 1.771 Industries (1.229) (4.948)** (-4.671)** (ii) -0.G9C82 (-2.129)* 0.27S20 (2.610)* -0.18931 (-2.578)* 0.0764 1.802 Indianapolis (i) 0,015^0 0.06530 -0.13292 0.0919 1.209 Pover & Light (0.845) (2.591)* (-2.517)* (ii) -G. 136 5 6 (-5,358)** C. 86939 (6.678)** -0.58427 (-6.780)** 0,^087 0,660 -23- Company ^O'^O b,,b^ ^2'^ Adj R^ DW Kansas Gas & (i) 0.00958 0.05934 -0.11753 0.0592 1.030 Electric (0.543) (2.105)* (-2.180)* (ii) -0.26637 (-5.560)** 0.96619 (6.276)** -0.56808 (-6.363)** 0.3773 0.452 Kentucky (i) 0.00963 0.02202 -0.05612 0.0076 1.961 Utilities (0.541) (1.1^3) (-1.327) (ii) 0.12029 (0.872) -0.18892 (-0.745) -0.02359 (-0.467) 0.0 2.064 Middle South (i) 0.00735 0.11770 -0.21256 0.1A81 1.960 Utilities (0.590) (3.500)** (-3.311)** (ii) -0.02662 (-0.417) 0.12416 (0.664) -0.07991 (-1.189) 0.0 2.080 Minnesota Pcver (i) 0.02331 0.07926 -0.19022 0.1313 2.234 & Light (1.322) (3.032)** (-3.167)** (ii) 0.16633 (1.194) -0.27111 (-0.994) -0.53969 (-0.950) 0.0216 2.372 Oklahoma Gas (i) -0.00265 0.18317 -0.2487 0.1983 1.835 & Electric (-0.246) (4.168)** (-3.988)* (ii) -0.07641 (-2.476)* 0.41012 (3.016)** -0.22261 (3.071)** 0.1075 2.048 Pennsylvania (i) 0.01407 0.03133 -0.07600 0.1510 2.487 Power & Light (2.203)* (3.677)** (-3.194)** (ii) -0.001658 (-0.255) 0.17075 (2.971)** -0.20329 (-2.998)** 0.0963 2.077 Public Ser'/ice (i) 0.03246 0.269A4 -0.45225 0.3899 0.994 Co. of Ind (1.587) (6.258)** (-6.407)** (ii) -0.00532 (-0.139) 0.16251 (1.738) -0.18217 (-2.504)* 0.0609 1.201 Public Service (i) -0.004 72 0.00794 0.01823 0.0604 2.458 Co. of New (-1.329) (0.855) (0.981) Mexico (ii) -0.02425 (-2.390)* 0.12433 (2.125)* -0.07105 (-1.449) 0.1123 2.351 Southern Coxr.pan; 7 (i) 0.02674 (1.373) 0.11349 (3.297)** -0.23800 (3.616)** 0.1699 1.799 (ii) 0.G9110 (-1.514) -0.10149 (-0.894) -0.10841 (-1.807) 0.0409 2.033 Toledo Edison (i) -O.C7C0S 1.15032 -1.59972 0.7007 1.995 Co. (-0.71C) (6.013)** (-12.335)** (ii) -0.5^136 (-4.111)** 2.31767 (7.952)** -1.56250 (-14.708)** 0.7646 2.137 -24- Ccmpany ^O'^O ^'^1 ^2'^2 Adj R DW Union Electric (i) C.04181 (2.267)* 0.21484 (5.164)** -0.41449 (-5.9S9)** 0.3655 1.671 (ii) 0.35596 (1.656) -0.74297 (-2.427)* -0.16974 (-1.366) 0.1266 2,133 Virginia (i) 0.05803 0.06424 -0.29850 0.1996 1.947 Electric & (2.747)** (2.870)** (-3.999)** Power (ii) 0.19350 (1.564) -0.27783 (-1.045) -0.1961S (-2.511)* 0.1117 2.058 Arizona Public (i) -0.00456 0.01466 0.00413 0.0646 2.101 Service Co. (-0.903) (1.406) (0.1A4) (ii) -0.03619 (-3.626)** 0.30231 (3.361)** -0.25740 (-2.895)** 0.1804 1.845 Central Hudson (i) 0.00286 0.00254 -0.00229 0.0 1.994 Gas & Elec. (0.663) (0.339) (-0,163) (ii) -0.03293 (-2.252)=^ 0.15267 (2.567)* -0.100C4 (-2.463)* 0.0651 1.982 Central Illinois (i) 0.09047 0.17258 -0.57167 0.4719 0.658 Public Service (3.319)** (3.151)** (-7.564)** (ii) -0.90822 (-7.588)** 3.15609 (8.871)** -0.93920 (-13.310)**' 0.7264 0.144 Cincinnati Gas (i) -0.00025 0.08705 -0.12224 0.1352 2.030 & Electric (-0.015) (3.352)** (-2.506)* (ii) -0.23083 (-2.052)* 0.64173 (2.233)* -0.21399 (2.429)* 0.0569 1.885 Del Marva Power (i) 0.19 334 0.41275 -1.249 81 0.6155 2.443 & Light (2.228)* (2.074)* (-10.126)** (ii) 0.15893 (0.536) 0.39631 (0.656) -0.17608 (-9.703)** 0.5917 2.383 Illinois Power (i) -0.C0787 0.09725 -0.10580 0.0827 1.370 Cc. (-0.355) (2.536)* (-2.353)* (ii) -0.42133 (-4.397)** 1.15849 (4.710)** -0.44820 (-4.874)** 0.2503 0.741 Interstate (i) 0.C0925 0.00361 -0.02741 0.0187 2.251 Power Co. (2.101)* (0.421) (-1.728) (ii) -0.C3347 (-1.980) 0.21732 (2.627)* -0.13678 (-3.050)** 0.1118 2.229 Iowa- Illinois (i) 0.02S98 0.03890 -0.132G6 C.0371 1.822 Gas & Elec. (1.400 (1.372) (-2.110)* (ii) 0.19578 (1.4S6) -C. 31504 (-1.240) -0.07033 (-1.280) C.0324 1.968 -25- Company ^O'^O \'\ ^2'^2 Adj R^ DW Iowa Power & (i) 0.48120 0.33329 -1.60239 0.7758 2,247 Light (4.745)** (2.240)* (-15.059)** (ii) -0.839 39 (-2.092)* 2.77906 (3.810)** -1.67717 (-16.340)** 0.8032 2.305 Long Island (i) 0.10361 0.88290 -1.66656 0.7821 1.850 Lighting (1.592) (6.316)** (-15.318)** (ii) -0.61427 (-2.513)* 2.52449 (4.409)** -1.56577 (-13.220)** 0.7280 1.668 Louisville Gas (i) 0.00846 0.00525 -0.1869 0.3415 1.552 & Elec. (7.192)** (2.245)* (-5.841)** (ii) 0.02649 (5.335)* -0.05 336 (-3.417)** 0.01177 (1.458) 0.3992 1.691 Montana Power (i) 0.44922 0.25537 -1.50737 0.6962 2.131 Co. (4.182)** (1.633) (-12.157)** (ii) -0.07322 (-0.366) 1.32833 (3.352)** -1.54697 (-13.374)** 0.7313 2.253 Niagra Mohawk (i) 0.01702 0.12669 -0.23521 0.1800 1.834 Power (1.330) (3.588)** (-3.977)** (ii) 0.06036 (1.514) -0.09146 (-1.025) -0.04617 (-0.967) 0.0309 2.057 Northern States (i) 0.13861 0.29022 -0.73035 0.3591 2.393 Power (3.674)** (4.339)** (-6.240)** (ii) -0.3968 (0.779) 0.87483 (5.564)** -0.90947 (7.353)** 0.4410 2.037 Public Service (i) 0.14889 0.51933 -1.35131 0.6988 1.662 Co. of Colorad( D (2.177)* (3.352)** (-12.370)** (ii) -0.46475 (-2.870)** 2.21632 (5.128)** -1.47094 (-14.097)** 0.7491 1.850 Rochester Gas (i) 0.44272 0.07723 -1.67034 0.8176 1.919 & Electric (5.555)** (0.538) (-17.103)** (ii) -0.96072 (-2.237)* 2.76164 (3.367)** -1.71040 (-18.971)** 0.8444 2.160 Sierra Pacific (i) 0.00231 0.08869 -0.15120 0.1506 2.060 Power Co. (0.262) (3.667)** (-2.777)** (ii) -0.04478 (-1.591) 0.22580 (1.902) -0.10245 (-1.664) 0.0271 1.710 Tucson Gas ^ (i) -C. 00024 0.06588 -0.09083 0.2165 2.126 Electric (0.085) (4.225)** (-2.962)** (ii) -0.01684 (-2.846)** 0.265 70 (3.157)** -0.20351 (-2.776)** 0.1331 2.084 -26- Ccnpany ^0'°0 b,,b, ^2'^2 Adj R" DW Washington Water (i) 0.02290 0.107S8 -0.21538 0.2178 2.043 (1.359) (4.128)** (-3.903)** (ii) -0.06740 0.25160 -0.12870 0.0479 1.983 (-1.061) (1.605) (-2.244)* Wisconsin (i) 0.01430 0.10775 -0.20184 0.1779 1.858 Electric Power (0.936) (3.921)** (-3.544)** (ii) -0.12923 0.40463 -0.19191 0.0924 1.913 (-2.394)* (2.810)** (2.806)** Wisconsin (i) 0.0270S 0.08202 -0.21199 0.1118 2.353 Public Service (1.685) (2.801)** (-3.039)** (ii) 0.10711 -0.19629 -0.06724 0.0163 2.272 (1.204) (-0.932) (-1.185) (i) represents coefficients for regression equations using current earnings. Cii) represents coefficients for regression equations using pencanent earnings. * denotes significance at 5% level. ** denotes significance at If' level. -27- REFEREITCES 1. Ang, Janes S. (1975), "Dividend Policy: Inf ortnational Content or Partial Adjustment," Review of Econoinics and Statistics , pp. 363-390. 2. Cochran, W. G. (1970), "Some Effects of Errors of Measurement on Multiple Correlation," Journal of American Statistical Associa- tion , 65, pp. 22-34. 3. Darby, M. R. (1972), "The Allocation of Transitory' Income Among Consumers' Assets," The American Economic Review , (September 1972), pp. 928-941. 4. Darby, M. R. ("1974), "The Permanent Income Theory of Consumption — A Restatement," Quarterly Journal of Economics , (May 1974), pp. 228-250. 5. Eisner, Robert (1967), "A Permanent Income Theory of Investment," The American Economic Review , 57, pp. 363-390. 6. Eisner, Robert (1978), Factors in Business Investment , National Bureau of Economic Research, General Series No. 102. 7. Fama, E. F. and H. Babiak (1968), "Dividend Policy: An Empirical Analysis," Journal of American Statistical Association , 63, pp. 1132-61. 8. Friedman, M. (1957), A Theor:-' of the Consumption Function , Princeton, N.J,: Princeton University Press, 1957. 9. Lintner, J. (1956), "Distribution of Income of Corporations Among Dividends , Retained Earnings , and lazies , " American Economic Review , (May, 1956), 46, pp. 97-113. 10. Modigliani, F. and M. K. Miller (1958), "The Cost of Capital, Corporation Finance and The Theory of Investment," American Economic Review , 48, pp. 261-297. 11. Modigliani, F. and M. H. Miller (1963), "Corporate Income Tax and the Cost of Capital: A Correction," Am.erican Economic Review , 53, pp. 433-443. 12. Miller, M. H. and F. Modigliani (1961), "Dividend Policy, Growth and Valuation of Shares," Journal of Business , 34 (October 1961), pp. 411-433. 13. Miller, M. E. and F. Modigliani (1966), "Some Estimates of the Cost cf Capital to the Electric Utility Industry," Arierican Economic Review, 56, pp. 334-391. -28- lA. Weston, J. F. Q978), Managerial Finance , Sixth Edition, Illinois: The Dryden Press. M/E/194 I .ii<^ .f^v