Tax Position, Investment Opportunity Set (IOS), and Signaling Effect As A Determinant of Leverage
Tax Position, Investment Opportunity Set (IOS), and Signaling Effect As A Determinant of Leverage
Tax Position, Investment Opportunity Set (IOS), and Signaling Effect As A Determinant of Leverage
43
simultaneous relationship on the two decisions. In these studies, agency theory is used to
explain the simultaneity between leverage and dividend decision. This theory states that, an
agency relationship arises whenever an individual or group, called a principal, hires someone
called an agent to perform some services, where the principal delegates decision-making
power to the agent (Brigham and Daves, 2004). Further, the conflict will emerge, due to
agents personal goals that compete with principals wealth maximization. In this case,
leverage and dividend decision perform as a control mechanism to the conflict. Jensen,
Solberg and Zorn (1992) find that the level of insider ownership is strongly affecting the
simultaneity between leverage and dividend decision. Further, they find that company with
high level of insider ownership has a small dividend and also small leverage. Using the same
argument, Crutchley and Hansen (1989) find that, company uses three decisions
simultaneously in order to lessen the agency cost, they are: managerial ownership, leverage
decision and dividend decision. Noronha, Shome and Morgan (1996) study, also find the
simultaneity, on the company, which has been characterized with low growth and no block
holder.
On the contrary, Barclay, Smith and Watts (1995) argue that, not only agency theory
but also signaling theory is capable to explain the simultaneity between leverage and dividend
decisions. They have found that, three factor which are: companys tax position, investment
opportunity set (IOS), and signaling effect are capable to predict the simultaneity. Using three
method of regression analysis that is pooled regression, cross section regression and fixed
effect regression, they find that the three factors have significant effect on the simultaneity.
The consideration about tax in the companys capital structure has proposed by
Modigliani and Miller (1958), which have called irrelevance theory. On that proposition they
assume that, tax is not relevant to be considered in the companys capital structure, for there is
no such tax, either individual or corporate tax. Therefore there is no tax-shield on using
certain amount of debt. According to this view, the large amount of leverage in the companys
capital structure will not affect companys value. However, the absence of tax is unrealistic,
because it does exist, both individual and corporate (Modigliani and Miller, 1963; Miller,
1977). In this case the amount of interest inherent in the use of debt can reduce the amount of
corporate tax. As a consequence, company will use higher leverage in their capital structure,
to capture this interest-tax shield. Therefore, higher leverage will cause higher companys
value.
Besides the advantage of interest-tax shield, there is also a non-debt tax shield that
can affect the simultaneous relationship between leverage and dividend (DeAngelo and
Masulis, 1980). In this case, company with high non-debt tax shield will prefer to use low
leverage (McKie-Mason, 1990).
As an opposite of the company, there is no tax-shield on individual income. Investors
have to pay high amount of tax on their earning paid by the company (cash dividend).
However, reinvested earning which results in the form of capital gain will caused lower tax.
Thus, investors prefer to collect lower dividend or reinvest their earning to the company in
order to capture the capital gain. Chang and Rhee (1990), find that higher individual tax
compared to capital gain tax, persist the company pays low dividend and uses high retained
earning in their capital structure.
The simultaneous relationship between leverage and dividend decision is also
determined by companys investment opportunity set (IOS). The argument about IOS is
proposed by Myers (1977). On this proposition, they argue that, mainly company is a mix
between its assets in place that is tangible, and future investment option or growth option that
is intangible. Future investment option reflects companys investment opportunity (through
the choice of many projects with positive NPV). Thus, companys opportunity to grow, will
depend on the choice between: executes those projects with positive NPV or abandons them.
The choice usually called managerial discretion.
Generally high growth companies have higher investment opportunity compared to
those slow growth companies. However they dont have sufficient collateral assets that are
tangible when external financing, such as debt, is required. With this condition, high growth
44
companies will choose low leverage or high internal financing such as retained earning to
finance those potential projects. As a result the dividend paid to investors will be low.
The opposite site happens on slow growth companies, these companies usually use
high leverage and pay high dividend (Long and Malitz, 1985; Williamson 1988). Further,
Jensen (1986) argue that the tendency on using leverage and dividend in both, high and slow
growth companies, can be explained through the existence of free cash flow problem. On the
slow growth companies, which are particularly mature, debt or leverage is used as a bonding
to investors that free cash flow will extend the amount of dividend paid. This argument
directly indicates the simultaneous relationship between leverage and dividend decision.
The third factor to explain the simultaneity between leverage and dividend decision is
signaling effect. This effect appears because of the existence of asymmetric information
between managers and investors. In this case, different investors have different views on both,
the level of future dividend payments and the uncertainty inherent in those payments, and
managers have better information about future prospects than public stockholders. This
difference of view, will affect the stock market, indicated by stock price changing around
dividend announcement. Smith (1986), Venkatesh (1989), Koch and Shenoy (1999), and Kale
and Noe (1990) examine the effect of signaling on leverage and dividend decision.
Using the argument about tax, investment opportunity set and, signaling effect, this
study develops a model to examine the simultaneity between leverage and dividend. The two
stage least square (2SLS) method is used to find the existence of the simultaneity. Two
models of equations will be developed: first, the leverage equation, with leverage as an
endogenous variable, and second, the dividend decision, with dividend also as an endogenous
variable. The three factors proposed by Barclay et.al. (1995), which are: tax position,
investment opportunity set (IOS) and signaling effect then, will be use as control variable in
both equations. Another two variables will also be use in the model, to meet the identification
of equation, require in the 2SLS method. They are: assets tangibility that will be added on the
leverage equation and liquidity position that will be added on the dividend equation.
II. Theoretical Background and Hypothesis Development
2.1. Tax Position
According to the irrelevance theory, companys value will only be determined by
investment decision. This decision will only affects the left hand side of the balance sheet that
is companys assets. In this case, the tangibility of assets being used will determine that value.
2.1.1. Leverage and Tax Position
By considering the existence of tax, Modigliani and Miller (1963) revise their
argument in the irrelevance theory. On their revised proposition, they find that using a certain
amount of leverage can reduce corporate tax. Further, the interests payment inherent in that
leverage can be used as a tax shield. They conclude, this interest tax shield is the
explanation for higher leverage decision.
Besides the advantage of interest tax shield, some companies also have the benefit
from another form of debt tax shield, that is non-debt tax shield. This form of tax shield
usually provided by the government, concerning the risk inherent in the companys business.
The non-debt tax shield can be classified into two forms: tax loss carryforward and
investment tax credit. However, on the studies generated by Bradley, Jarrel and Kim (1984)
and also by Noronha et.al (1996), another non-debt tax shield can also be found in the form of
depreciation. These studies argue that, the level of depreciation reflects the amount of
companys tangible assets, which can be use as collateral assets; hence companies
characterized by high depreciation cost tend to use high leverage.
Many studies have been conducted to examine the effect of non-debt tax shield on
leverage decision; however the results are still ambiguous. Scott (1976), find positive
relationship between non-debt tax shield, which is proxies by the level of depreciation, and
leverage decision. He concludes that, the level of depreciation reflects the amount of tangible
assets, which can be use as collateral assets. Thus, company with high level of depreciation
should have more debt in its capital structure. This result is consistent with Bradley, Jarrel and
45
Kim (1984) and also by Noronha et.al (1996). However, Mc-Kie Mason (1990) find negative
relationship between leverage decision and non-debt tax shield. He argues that, for high-risk
companies (such as mining and exploration companies), which always have to face the
possibility of negative earning, tax shield is very important. Therefore in some countries, the
government gives a number of tax facilities, through some tax regulations such as: zero-tax
status and tax loss carryforward. These facilities subsequently, affect companys leverage
decision, which always tend to be low. This conclusion is consistent with the proposition
stated by DeAngelo et.al (1980) in which, the higher the benefit from non-debt tax shield, the
lower the leverage will be used.
2.1.2. Tax Position and Dividend Decision
On the level of individual income there is no tax shield. Miller (1977) argues that, for
an individual income such as dividend, investors have to pay income tax, which is remarkably
higher than those on capital gain. Therefore, investors will prefer to obtain low than high
dividend, or reinvest their earning to capture the capital gain. Chang and Rhee (1990) find
that tax on individual income affect leverage and dividend decision simultaneously. In this
case, investors preference on capital gain, insists the company to pay low dividend and use
high retained earning. Consequently, company will use low leverage.
2.2. Investment Opportunity Set (IOS)
According to the concept of investment opportunity set (IOS), the mix between
companys assets in place and investment opportunity or growth option, will determine
companys growth level. Moreover the choice to execute the investment opportunity that
consists of positive NPV projects will increase companys growth if the projects being chosen
give some benefit in the future.
2.2.1. Investment Opportunity and Leverage Decision
Facing several investment opportunity, financing decision usually perform as a
tradeoff between underinvestment and overinvestment problem. These problems are related
with companys level of assets in place and investment opportunity. High growth companies,
which are characterized with high investment opportunity usually, face an underinvestment
problem. This problem appears when companies have to operate the positive NPV projects
that require more debt financing. However, these companies generally dont have sufficiency
on both, assets in place and cash flow to cover up their debt. If equity financing - by issuing
new common stock - is chosen, then some risks will emerge such as, the existence of
asymmetric information and the raise of conflict between shareholders (particularly new
shareholder) and creditors. Therefore issuing new equity will costly. For some companies, this
situation will lead to run off those investment opportunities, however some are prefer to
continue those projects using retained earning than to leave them.
On the contrary, slow growth or mature companies, which are characterized with high
free cash flow, high assets in place, but small number of investment opportunity, have to deal
with overinvestment problem. Jensen (1986) argues that, overinvestment problem is a conflict
between managers and shareholder, which is generally, arises due to the difference view of
free cash flow. Managers, who always seek for optimal size of the company, tend to see this
free cash flow as a non-profit fund, when it has to be reinvested to the companys main
business. Thus for some managers, they have a tendency to plow this fund into another
project, which is different with companys core business. However, investors see this free
cash flow as an additional amount for dividend paid. In order to lessen the conflict, managers
then use leverage to fund any investment opportunity. This leverage is also used as a bonding
mechanism to the investors that free cash flow will be paid as an increasing dividend.
Another study to provide evidence about the effect of investment opportunity and
assets in place is generated by Long et.al (1995). This study found that company with high
investment opportunity reflected in high R&D and high advertising cost, have lower leverage
compare to those with low R&D and low advertising cost.
46
47
dividend. As a consequence if companies fail to sustain this increasing earning, the stock
price will fall. In summary, high quality companies pay high dividend.
The effect of signaling on dividend decision is proven by Kale et.al. (1990). Their
study finds that companies which characterized with stable cash flow and high earning, have a
tendency to pay high dividend. Another study on signaling effect is also generated by Koch
et.al. (1999). Their study finds significant effect of signaling on leverage and dividend
decision simultaneity. Further, they also find that, the strong effect of signaling is occurs both
on underinvestment and overinvestment companies.
Three main sources are used to obtain the samples: a) Indonesian Capital Market
Directory published from 1994 to 1998, b) records about tax regulation especially tax
facilities published by Indonesian Tax Department and c) other publication which is
relevant with this study.
3.2. Variables Measurement
3.2.1. Leverage
From the first equation, that is leverage equation, leverage decision performs as
dependent variable. This study will use market ratio of leverage, that is, the comparison
between total book value of leverage and market value of the firm, as a proxy for leverage.
This market ratio is used with the consideration on its capability to identify the future value of
tangible asset, which can be used as collateral for leverage (Barclay, et.al., 1995). Another
reason on using this ratio is because market value of leverage is capable on predicting the
cash flow stability and future debt service ability (Brigham, Gapenski and Daves, 1999). The
formula for market ratio of leverage is shown bellow:
book value of total debt
48
EAT
closing price
Market value of the firm = total debt
EPS
(4)
The term LEVERAGE then will be used to determine this variable.
3.2.2. Dividend
Similar with leverage decision, dividend decision also performs as a dependent
variable in the dividend equation. The dividend payout ratio (DPR) that is dividend per share
(DPR) divided by earning per share (EPS) will be used as a proxy for dividend decision. The
formula for dividend payout ratio is:
dividend per share
closing price
Market-to-book value of equity = EPS
common equity
(6)
The term MTBEQ will be used to determine this variable in the model.
3.2.5. Signaling Effect
The theory of signaling pronounces that signaling effect portrays companys future
quality, either low or high. Since the quality can be explained by the tendency of companys
future earning to change, Barclay et.al (1995), argue that future abnormal earning can be
employ to asses the signaling effect. The formula for future abnormal earning is:
Future abnormal earning =
EPS ( t 1 ) EPS ( t )
closing price(t)
(7)
The term FAE will be use to point out this variable in the model.
3.2.6. Asset Tangibility
49
The amount of tangible assets reflects on its book value employed by companies, will
affect the amount of leverage. The collateral feature innate in those tangible assets, produce
more leverage (Titman et.al, 1984). With this view, this research will use the assets tangibility
ratio that is book value of tangible or fixed assets divided by total assets. The formula is:
Assets tangibility =
(8)
Endogenous Variables
NDTS
Market-to-book value
MTBEQ
Leverage
TANG
Current ratio
CRATIO
Dividend
50
effect with leverage and dividend decision, therefore the associations between 14 and 24 will
be positive.
3.5. Hypotheses Development
In keeping with the explanation about tax position, investment opportunity and
signaling effect, and also with the concept of IOS spectrum on leverage and dividend
decision, this research predicted the effect of those three factors as shown bellow:
Table 3.2.
The Hypotheses
Hypotheses
Predicted Effects
H1: Leverage
Dividend
Positive
Leverage
Negative
H2b: IOS
Leverage
Negative
Leverage
Positive
Leverage
Positive
Dividend
Negative
H3b: IOS
Dividend
Negative
Dividend
Positive
Dividend
Positive
N
100
100
100
100
100
100
100
Minimum
0.003
0.027
0
0.127
-1.269
0.111
0.413
Maximum
4.258
0.951
1
16.498
2.780
0.765
7.786
Sum
55.831
43.968
35
223.057
2.720
39.683
165.825
Mean
0.55831
0.43968
0.35
2.23057
2.72E-02
0.39683
1.65825
Std. Deviation
0.63097
0.24624
0.48
2.57728
0.44369
0.16107
1.06050
100
As observed, total valid number of samples is 100 firm-years. The endogenous variable or
dependent variable, that is, dividend decision has 0.003 of minimum value and 4.258 of
maximum value. The total value, that is the summation of total number of the dividend
sample, is 55.831, with 0.63097 of standard deviation. Another endogenous variable that is
leverage decision has minimum and maximum value that is 0.027 and 0.0951, with 43.968 of
total value, and 0.43968 of mean value.
From the summary statistics we can also see the characteristics of the exogenous
variables. For the non-debt tax-shield, the maximum and minimum value is 1 and 0. The total
value of this variable is 35, results in the mean value of 0.35. It also has 0.35 of mean value
and 0.48 of standard deviation. The second exogenous variable, that is, market-to-book value
of equity, the maximum and minimum value is 16.498 and 0.127. The total number of this
51
variable is 223.057, with the mean and standard deviation is 2.23057 and 2.57728. Future
abnormal return has 1.269 minimum value and 2.780 maximum value. This variable also has
0.4439 of standard deviation and 2.72E-2 mean value.
4.2. Discussion
4.2.1. Leverage Equation
The data collected is analyzed, with the help of SPSS program. The result for
leverage equation is shown on table 4.2 bellow:
Table 4.2
2SLS Result for Leverage Equation
2SLS Results
Equation 1
Dependent Variable = Leverage
T
Sig T
Independent Variables Coefficients
(Constant)
-1.495175
-9.052
0.0000
DIVIDEND
4.611197
12.552
0.0000
NDTS
-1.454518 -11.170
0.0000
MTBEQ
-0.098155 -14.885
0.0000
FAE
0.651639
11.720
0.0000
TANG
0.178100
2.201
0.0302
F = 58.62866
Signif F = 0.0000
R Square = 0.75720
Derived from table 4.2, the join effect of the exogenous variables, which are, dividend, nondebt tax shield, market-to-book value of equity, future abnormal earning and tangibility, on
leverage decision is 75.72%, significant at the level of 0.05. The positive coefficient of
endogenous variable, that is, dividend, is consistent with the hypothesis 1 which predicts the
positive simultaneous relationship between leverage and dividend decision. Similar with the
result on the endogenous variable, the negative coefficients on non-debt tax shield and
market-to-book value of equity and positives coefficients on future abnormal earning and
assets tangibility are also consistent with the hypotheses 2a, 2b, 2c, and 2d.
4.2.2. Dividend Equation
Using the same statistical tools, the result for structural equation 2, that is, dividend
equation is presented in the table 4.3 bellow:
Table 4.3
2SLS Result for Dividend Equation
2SLS Results
Equation 2
Dependent Variable = Dividend
Independent Variables Coefficients
T
Sig T
(Constant)
-7.831348
-5.415
0.0000
LEVERAGE
17.318832
5.709
0.0000
NDTS
-2.216659
-4.767
0.0000
MTBEQ
0.760253
5.728
0.0000
FAE
-1.016480
-5.132
0.0000
CRATIO
-0.070800
-1.305
0.1950
F = 8.67238
Signif F = 0.0000
R Square = 0.31568
52
The table shows that, jointly, the exogenous variables, which are, leverage, non-debt tax
shield, market-to-book value of equity, future abnormal earning and current ratio, affects the
dividend decision for about 31.568%. The positive sign and statistical significance of leverage
indicate that leverage and dividend decision are positively interdependent. This result is
consistent with the hypothesis 1. Unlike with the endogenous variable, the exogenous
variables show conflicting result with the hypotheses. Non-debt tax shield is related
negatively to dividend that is meaning consistent with the hypothesis 3a. However, the other
exogenous variables, that is, market-to-book value of equity, future abnormal earning and
current ratio show different sign with the hypotheses 3b, 3c, and 3d, even though they are still
statistically significant.
V.
5.1.
53
non-debt tax shield, it is possible to apply another proxy, such as depreciation. The time
differences are also possible to analyze, by considering the economic crisis begin in 1998. For
this reason, the analysis can be divided into two periods, before and after the crisis. Last, the
better result may emerge, if the maturity of leverage also considered.
REFERENCES
Adedeji, A. 1998. Does the pecking order hypothesis explain the dividend payout ratios of
firms in the UK?. Journal of Business Finance and Accounting, NovemberDecember, 25 (9) dan (10): 1127-1153
Barcaly, M.J., Marx, L.M., dan Smith, C.W. 2001. The joint determination of leverage and
maturity. Working Paper, 2-27
Barclay, M.J., Smith, C.W., dan Watts, R.S. 1995. The determinants of corporate leverage and
dividend policies. The New Corporate Finance: Where Theory Meets Practice,
International Edition. Irwin McGraw-Hill: 214-229
Baskin, J. 1989. An empirical investigation of the pecking order hypothesis. Financial
Management, Spring: 26-35
Bradley, M., Jarrell, G.A., dan Kim, H. 1984. On the existence of an optimal capital structure:
Theory and evidence. Journal of Finance, July 39 (3): 857-878
Brigham, E.F., Gapenski, L.C., dan Daves, P.R. 1999. Intermediate Financial Management,
Sixth Edition. The Dryden Press, Harcourt Brace College Publishers.
Chang, R.P., dan Rhee, S.G. 1990. The impact of personal taxes on corporate dividend policy
and capital structure decisions. Financial Management, Summer: 21-31
Cooper, D.R., dan Emory, C.W. 1999. Metode Penelitian Bisnis, Edisi 5. Penerbit Erlangga.
Crutchley, C.E., dan Hansen, R.S. 1989. A test of the agency theory of managerial ownership,
corporate leverage, and corporate dividends. Financial Management, Winter: 36-46
DeAngelo. H., dan Masulis, R.W. 1980. Optimal capital structure under corporate and
personal taxation. Journal of Financial Economics, 8: 3-29.
Gaver, J.J. dan Gaver, K.M. 1993. Additional evidence on the association between the
investment opportunity set and corporate financing, dividend, and compensation
policies. Journal of Accounting and Economics, 16: 125-160
Giyartiningrum, E. 2000. Pengaruh inisiasi dividen terhadap kandungan informasi
pengumuman laba. Unpublished Tesis S-2 Universitas Gadjah Mada Jogjakarta.
Gujarati, D. 1999. Ekonometrika Dasar. Penerbit Erlangga.
Holder, M.E., Langrehr, F.W., dan Hexter, J.L. 1998. Dividend policy determinants: An
investigation of the influences of stakeholder theory. Financial Management, Autumn
27 (3): 73-82
54
Jensen, G.R., Solberg, D.P., dan Zorn, T.S. 1992. Simultaneous determination of insider
ownership, debt, and dividend policies. Journal of Financial and Quantitative
Analysis, June 27 (2): 247-263
Jensen, M.C. 1986. Agency cost of free cash flow, corporate finance, and takeovers.
American Economic Review, May, 76 (2): 323-329
Kale, J.R., dan Noe, T.H. 1990. Dividends, uncertainty, and underwriting costs under
asymmetric information. The Journal of Financial Research, 13 (4): 265-277
Kallapur, S., dan Trombley, M.A. 1999. The association between investment opportunity set
proxies and realized growth. Journal of Business Finance and Accounting, AprilMay, (3) dan (4): 505-519
Koch, P.D., dan Shenoy, C. 1999. The information content of dividend and capital structure
policies. Financial Management, Winter 28 (4): 16-35
Long, M., dan Malitz, I. 1995. The investment-financing nexus: Some empirical evidence.
The Revolution in Corporate Finance, Second Edition. Basil Blackwell Inc.: 156-162
McKie-Mason, J.K. 1990. Do taxes affect corporate financing decision?. Journal of Finance,
December, 55 (5): 1471-1493
Megginson, W.L. 1997. Corporate Finance Theory. Addison-Wesley Educational Publishers
Inc.
Miller, M.H. 1977. Debt and taxes. Journal of Finance, May: 261-274
Modigliani, F., dan Miller, M.H. 1958. The cost of capital, corporation finance and the theory
of investment. American Economic Review, June 58 (3): 261-297
Modigliani, F., dan Miller, M.H. 1963. Corporate income taxes and the cost of capital: A
correction. American Economic Review, June: 433-443
Myers, S.C. 1977. Determinants of corporate borrowing. Journal of Financial Economics, 5:
147-175
.
Myers, S.C., dan Majluf, N.S. 1984. Corporate financing and investment decisions when
firms have information that investors do not have. Journal of Financial Economics,
13: 187-221
Noronha, G.M., Shome, D.K., dan Morgan, G.E. 1996. The monitoing rationale for dividends
and the interaction of capital structure and dividend decisions. Journal of Banking
and Finance, 20: 439-454
Riyanto, B. 1999. Dasar Dasar Pembelanjaan Perusahaan, Edisi 4. BPFE Yogyakarta.
Smith, C. 1986. Investment banking and the capital acquisition process. Journal of Financial
Economics, 15: 3-29.
Soetjipto, K. 1997. Pengaruh pengumuman dividen terhadap return saham di Bursa Efek
Jakarta tahun 1995. Unpublished Tesis S-2 Universitas Gadjah Mada Jogjakarta.
Soter, D., Brigham, E.F., dan Evanson, P. 1995. The dividen cut heard round the world:
The case of FPL. The New Corporate Finance: Where Theory Meets Practice,
International Edition. Irwin McGraw-Hill: 250-261
55
56