The relationship between pension plan funding levels and capital structure: further evidence of a pecking order.

ABSTRACT

The literature on defined benefit pension plans presumes firm managers make an economic choice to overfund or underfund their defined benefit pension plans irrespective of other firm operating conditions and/or financing practices. Theoretically, overfunding is encouraged due to the tax deductibility of the pension expense and the non-taxed status of pension plan earnings. Alternatively, underfunding might occur due to poor cash flow or the firm's ability to earn a greater after-tax rate of return on firm assets than on pension plan assets. This study demonstrates that underfunding occurs principally due to a firms' incapacity to fully fund. By examining a period when multiple defined benefit pension funding disclosure criteria existed, the results are seen not to be dependent on the form of pension disclosure requirements.

1. INTRODUCTION

The accounting, finance and economic literatures on pension funding theory and practice make the a priori assumption that firms make an economic choice with respect to funding defined benefit pension plans irrespective of firm operating conditions or financing policies. Much of the literature is either prescriptive as to what funding policy should be (Ippolito, 1985; Sharpe, 1976; and Tepper, 1981) or descriptive as to what funding policy is (Bodie, et al, 1987; Francis and Reiter, 1987 and Thomas, 1988). Neither the prescriptive nor the descriptive literature explicitly addresses the role of funding defined benefit pension plans as being integral to broader capital structure and operating policy decisions. This oversight is surprising given Stone's (1987) conclusion that firms build financial slack by overfunding defined benefit pension plans and recouping excess funds for internal financing purposes in times of need. The creation of financial slack, a central component in capital structure theory, by overfunding defined benefit pension plans suggests a direct linkage between operating decisions and capital structure decisions.

Myers' (1984) modified pecking order theory (MPOT) provides an empirically testable hypothesis for capital structure decisions. MPOT proposes that financial managers (1) avoid using external equity and risky debt, (2) set dividend policies which can be maintained by internally generated equity, (3) maintain financial slack, and, since dividends are "sticky" while investment opportunities are variable, (4) fund additional needs with risky debt before using new equity. The level of funding defined benefit pension plans fits perfectly into this pecking order framework. Pension funding is, to a major extent, contingent on discretionary managerial assumptions; consequently, pension plans could be overfunded in periods when internal funds exceed the need for investments and dividends and underfunded in those periods when internal funds are insufficient for investment and dividend needs. Overfunding is consistent with building financial slack whereas underfunding is consistent with managerial motives of imposing on employees an unsecured interest in the long term viability of the employer (Ippolito, 1985).

In a review of capital structure theory and research, Harris and Raviv note that "[t]his area is still in its infancy and is short on implications relating capital structure to industrial organization variables ..." (1991, p. 351.) This paper investigates firms' pension funding policies as being jointly determined with their capital structure and dividend policy decisions, consistent with Myers' pecking order theory. The paper begins with a review of the literature, linking the operating decision to fund defined benefit pension plans with the firm's capital structure and dividend policies. The subsequent sections discuss the methodology for investigating these linkages and a discussion of the empirical results. The paper concludes with a summary of significant findings and conclusions drawn from the results.

2. PENSION RESEARCH, CAPITAL STRUCTURE, AND DIVIDEND POLICY

Early pension theory research focused on the tax aspects of corporate pension plan funding. According to Black (1980), Feldstein and Seligman (1981), Ippolito (1985), and Tepper (1981) tax reasons justify funding defined benefit pension plans to the maximum extent possible. Pension plan payments are fully tax-deductible, plan earnings are tax-deferred to the firm and the firm earns a before-tax rate of return on the assets it places in the trust. The firm has a distinct financial incentive to over fund its defined benefit pension plans. Empirical examinations link pension funding levels to taxes, as-well-as profitability and firm risk levels, but these studies produce conflicting outcomes.

Bodie, Light, Morck, and Taggart (1987) investigate the funding level of pension plans subsequent to Statement of Financial Accounting Standards (SFAS) No. 36 (1980). Firms which had greater tax burdens, higher profitability, and less risky debt (measured by bond ratings) funded pension plans at higher levels than less profitable, more risky firms. Francis and Reiter (1987) found that firms with the highest average tax rate had higher (safer) bond ratings but lower pension funding ratios than less profitable, more risky firms, a contradiction to Bodie et al. Taxes and pension funding are more fully explored by Thomas (1988), who finds pension contributions and tax status are positively correlated. Over the period 1980 to 1984, firms that decrease pension funding also decrease in tax status from being tax-paying firms to non-tax-paying. Thomas confirms pension funding level is linked to profitability and financial slack. His results have significant implications for capital structure research.

Capital structure research provides empirical support for MPOT. Baskin (1989) finds higher debt to total assets ratios are significantly related to lower profitability, higher growth in invested capital, and larger prior period dividend payments. Claggett (1991) confirms the pecking order theory from convergence of long-term debt to total assets ratios. Firms above their industry (4 digit SIC) average favor equity, thereby lowering the leverage ratio toward the industry average, while firms below the industry average are observed to favor debt over equity. Using simultaneous equations on samples of firms from 1982 and 1987, Jensen, Solberg, and Zorn (1992) confirm that higher long-term debt to total assets ratios are significantly related to lower dividend payout ratios, lower profitability, lower research and development spending, higher fixed asset utilization, and lower business risk. These results confirm the findings of Baskin with the exception of the inverse relationship between leverage and dividend payout. Baskin proposed that larger dividends lead to higher leverage ratios due to increases in firms' cash needs. Jensen et al interpret their result as deriving from firms' unwillingness to commit to higher dividends if they have high fixed financing costs.

The empirical research on defined benefit pension plan funding suggests an interaction with both capital structure and dividend policy decisions. Bodie et al (1987) report pension funding levels significantly positively related to firm profitability, an empirically supported determinant of capital structure and dividend policy. Francis and Reiter (1987) find funding levels are positively related to capital availability, negatively related to financial leverage, and negatively related to profitability. Since both capital availability and profitability were in the same regression, the effect of capital availability potentially subsumed any effect from profitability.

Bodie et al (1987) propose a corporate finance theory wherein pension plan assets and liabilities are part of the corporate balance sheet. Firms concerned with this "extended balance sheet" make decisions on pension plan funding levels as part of the firm's financial decisions. This perspective implies firm managers could use higher financial leverage for firm assets if they overfund pension plans. Bodie et al do not, however, empirically investigate financial leverage as a possible determinant of pension plan funding level. Stone (1987) examines financial characteristics of firms terminating defined benefit pension plans. Increased likelihood of termination is hypothesized to be a function of pension overfunding and financial leverage. Stone finds a significant negative correlation between leverage and profitability.

The interconnections between pension funding policy, dividend policy, and capital structure decisions have not been directly tested. Francis and Reiter (1987) recognize the potential impacts of dividend policy on pension funding through their capital availability measure, but do not test the effect. Bodie et al do not test the impact of either capital structure or dividend policy on pension funding although they assert a financial slack motivation for pension funding.

3. RESEARCH METHODOLOGY

Corporate decisions on funding pension plans, establishing financial leverage, and setting dividend policies are potentially joint decisions and result from conformance to a pecking order decision framework. In addition to these interaction effects, potential determinants of pension funding, dividend policy, and financial leverage are obtained from previous literatures.

Tax status appears to be a principle determinant of pension funding. Bodie et al (1987), Francis and Reiter (1987), and Thomas (1988) all establish a significant positive relationship between funding level and tax rate. The pension literature also suggests pension funding levels are related to profitability, but the direction of association is unclear. Bodie et al find a significant positive relationship, while Francis and Reiter find a negative, although statistically insignificant, relationship. Other potential contributors to pension funding are financial slack (Stone, 1987), firm size (Francis and Reiter, 1987), and the ratio of accrued benefits to employees (Francis and Reiter, 1987). This latter ratio is inversely related to pension plan funding levels. Citing Ippolito (1985), Francis and Reiter hypothesize this inverse relationship as the bonding of employees to the firm by first promising large pension benefits, and then underfunding the pension plan.

The capital structure literature establishes that financial leverage is significantly related to firm profitability, firm size, and industry representation. Baskin (1989) provides a thorough review of the literature demonstrating a negative relationship between leverage and profitability. Firm size has also been demonstrated to be significantly positively related to leverage (Jalilvand and Harris, 1984 and Titman and Wessels, 1988). Titman and Wessels posit that larger firms are capable of supporting higher proportions of leverage since they tend to be more diversified and less prone to bankruptcy. Baskin (1989), Bowen et al (1982), Claggett (1991), and Titman and Wessels (1988) provide evidence of differences in leverage by industry. In particular, Titman and Wessels indicate that firms manufacturing machines and equipment (SIC values between 3400 and 4000) should have less debt. Other potential attributes of financial leverage include business risk, future growth opportunities, and tax status. The empirical evidence on business risk is mixed. Titman and Wessels (1988) find no statistical association between the two variables while Jensen et al (1992) find a negative relationship, which is significant, however, in only one of two years studied. Growth opportunities, measured by expenditures on research and development, are significantly related to leverage as confirmed by both Jensen et al and Titman and Wessels. Finally, tax burden is a potential factor in financial leverage. Tax deductions for interest expense make debt financing desirable only to the extent that the firm has tax liabilities. Therefore, financial leverage is anticipated to be positively related to tax burden on operating income.

Dividend payout ratios are hypothesized to depend on profitability, growth in sales, business risk, and investment opportunities. Miller and Rock (1985) theorize dividends are a signal about current and future profitability, a conclusion empirically substantiated by Baskin (1989) and Rozeff (1982). Dividend payout should be negatively related to recent sales growth based upon Rozeff (1982) and Jensen et al (1992). Business risk, measured by variability in operating income, should be negatively related to dividend payout since firms experiencing greater uncertainty in revenues should be less willing to commit to a higher dividend payout, as confirmed by Jensen et al. Based on Myers and Majluf (1984), and supported empirically by Baskin (1989) and Jensen et al (1992), dividend payout should be negatively correlated with investment opportunities.

A system of equations is the preferred analytical method since the three variables are hypothesized to be jointly determined. Previously, Jalilvand and Harris (1984) and Jensen et al (1992) used systems of linear equations to demonstrate the simultaneous determination of dividend and leverage policies. The structural equations to be estimated, using variables defined in Table 1, are:

(1a) PENSION = f (LEVERAGE, DIVIDEND, TAX, BENEFITS, PROFIT, SLACK, SIZE)

(1b) LEVERAGE = f (PENSION, DIVIDEND, BUSRISK, PROFIT, R & D, SIZE, TAX, INDUSTRY)

(1c) DIVIDEND = f (LEVERAGE, PENSION, BUSRISK, GROWTH, PROFIT, INVEST)

The three stage-least-squares (3SLS) method is applied to the system of equations specified above. Data are drawn from the COMPUSTAT PC-Plus database and the sample is restricted to those firms offering defined benefit pension plans that had the available data for calculation of the variables specified in Table 1.

The endogenous variables in this study are pension funding (PENSION), financial leverage (LEVERAGE), and dividend payout (DIVIDEND). Variable definitions parallel those used in previous empirical studies. The pension funding variable uses accumulated benefits, i.e. those vested plus those not vested, for two reasons. First, the study encompasses the 1987 to 1992 period, when two separate pension funding standards were in effect. Statement of Financial Accounting Standards (SFAS) No. 36 (1980) provides for disclosure of vested and non-vested accumulated benefits while SFAS No. 87 (1985) provides for disclosure of vested, accumulated and projected benefits. Barth (1991) examines all three measures for SFAS #87 and determines that there is less error in the measure of accumulated benefits than in the other two liability measurements. The leverage variable definition follows that used by Baskin (1989) and Jensen et al (1992), among others. The definition of dividend payout is designed to measure dividends as paid from operating income prior to the payment of pension expense since it is proposed that pension plan contributions are a joint decision with dividend policy and, as such, are included in the measure of operating profit.

4. EMPIRICAL RESULTS ON ENDOGENOUS VARIABLES

Table 2 contains the results of the 3SLS regression for pension funding. Table 2 reveals that pension funding is significantly negatively related to firm profitability in all three years. This result contradicts the consensus of the tax implications for pension funding. It has been argued that firms should fund defined benefit pension plans to the maximum extent possible given the tax deductions for pension expense and the non-taxed earnings of pension plan assets. This argument implicitly assumes that the non-taxed rate of return earned on pension plan assets exceeds the after-tax rate of return on firm assets. The results of the 3SLS regression indicate that firms earning a higher rate of return on firm assets fund pension plans at a lower level; i.e., firm's managers trade off profitability on firm assets with pension assets and invest where the highest after-tax return occurs. Potentially, firms in high tax brackets are more profitable firms. Those profits are reinvested in firm assets, invested in pension assets to satisfy pension obligations, or distributed to stockholders, whichever provides the greatest benefit to stockholders. The extent to which pension plans can be overfunded, however, is restricted by law (Omnibus Budget Reconciliation Acts of 1987 and 1990). The results in Table 2 reveal that overfunded plan firms with limited opportunities for investing internally, due to low return on existing firm assets in its chosen industry, does not need the funds for additional financing of assets and, therefore, distribute profits as dividends, accounting for the significant positive relationship between dividend payout and funding level. This distribution of profits as dividends and funding of pension plans is consistent with the observed inverse relationship between pension funding level and financial leverage.

Pension funding is significantly negatively related to financial leverage in all three years. This demonstrates a pecking order for borrowing wherein a firm first taps its internal sources of funds, if it needs them, before proceeding to external financing.

Because the Pension Benefit Guaranty Corporation imposes real penalties on a firm for underfunding its pension plan, there is a lower limit on the funding status of pension plans beyond which the firm must rely on external financing. The result is the significant negative relationship between pension funding and financial leverage reported in Table 2. Table 3 confirms the inverse relationship between financial leverage and both pension plan funding levels and profitability. The significant negative relationship between financial leverage and profitability in Table 3 is the same as that reported in previous studies. The results for the 3SLS for financial leverage confirm pecking order theory. Specifically, firms requiring investment funds, but lacking funds from operations (proxied by profitability), and not having further access to discretionary pension funds, as evidenced by the underfunded status for the more highly leveraged firm, rely upon increased external debt, particularly when the firm has committed itself to high dividend payout levels. Table 4 confirms that financial leverage is significantly positively related to dividend payout ratios in all three years.

The results of the 3SLS regression for dividend payout ratios are contained in Table 4. These results confirm that dividend payout ratios are significantly positively related to both pension funding levels and financial leverage. These findings confirm a pecking order behavior. Higher dividend payout accompanies higher profitability, a statistically significant result in all three years. The dividend payout increases when firms have satisfied their pension obligations, but acts to constrain funds available for internal investment. As a consequence, the higher dividend payout firms use higher financial leverage.

5. EMPIRICAL RESULTS ON EXOGENOUS EFFECTS

The pension funding results in Table 2 confirm a negative relationship between funding levels and benefits per employee, as reported in Francis and Reiter (1987). The results, however, are significant in only one of three years. It is to the firm's advantage to promise employees large benefits and then underfund those benefits, as proposed by Francis and Reiter, only if the firm has alternative opportunities to use those funds. The lack of significance of this variable may be clue to its effect being subsumed by investment in firm assets, producing the inverse relationship between funding levels and profitability. The positive relationship between funding level and firm size indicate that older, more mature firms more fully fund their pension plans.

In Table 2 financial slack takes the correct sign in two of three years, but the results are not significant. It was hypothesized that firms would have overfunded their pension plans or accumulated extra liquid assets to create financial slack in the firm. After controlling for leverage, dividend, and profitability effects, however, the financial slack is not found to be a major determinant of pension funding. Furthermore, tax burden does not explain pension funding. Firm managers apparently trade off investment opportunities, dividend policies, and capital structure decisions such that tax payments result from managerial decisions rather than drive managerial decisions.

Table 3 contains empirical support for results reported by Jensen et al establishing higher financial leverage ratios linked to both lower research and development expenditures and lower business risk. The former is statistically significant in all three years while the latter is not significant in any year, although the sign of the coefficient is as theorized. Industry representation contributes to observed leverage, as the industry variable is negative in all three years, although significant in only one. Additionally, size is observed to be significant and positive in explaining financial leverage, in two of three years, confirming the ability of larger firms to use more leverage since they are more likely to be diversified and suffer from lower bankruptcy risk than smaller firms.

Dividend payout ratios, reported in Table 4, are not significantly related to growth opportunities, investment opportunities, or business risk. The coefficient on growth is in the hypothesized direction, but is significant in only one of three years. The positive coefficients on growth and investment confirm the observed relationship between dividend payout and financial leverage. Higher payout firms needing to invest in firm assets are constrained in their cash flows. This does not necessarily limit growth or investment, but forces the firm to borrow more. The leverage variable, therefore, captures the relationship between investment and growth, making those variables insignificant in the 3SLS regression.

6. CONCLUSIONS

This study provides statistically significant empirical support for the pecking order theory of capital structure wherein capital structure decisions are jointly determined with organizational policy variables. The empirical results suggest that the more highly leveraged firm experiences lower profitability and is constrained by a larger dividend payout, possibly due to the "stickiness" of dividends and declining profitability. In addition, this firm has exhausted its internal sources of financing by underfunding its pension plans, most likely to the extent legally possible. The results are consistent with the pecking order theory of capital structure in that firms which offer a defined benefit pension plan inherently create an opportunity for firms to use additional internal financing at the expense of the employees. This "borrowing" from employees provides the firm's managers with a source of internal financing in addition to operating income since pension benefits are, in theory, a deferral of current salary into the future. By funding this deferral at less than the full amount, the firm's managers not only improve cash flow by offering deferred salary instead of current salary, but bind the employees to the firm by essentially making them bond holders in the firm with a long term interest in the success and profitability of the firm.

This study also resolves a conflict in the literature regarding the relationship between dividend payout and financial leverage. A higher dividend payout constrains the firm's cash flow such that, in times of investment need, the firm must seek external financing, in the form of debt due to signaling costs, resulting in higher financial leverage ratios.

It should be noted that this study examined only those firms offering defined benefit pension plans. There exists the possibility that not all industries are represented since defined contribution pension plans might be the norm in certain industry groupings. The size of the sample tested, over eleven hundred firms in each of three years, potentially mitigates any industry limitation. Further research is clearly called for, since one implication of these results is that firms without defined benefit pension plans would not have the underfunding option as a source of internal financing.

TABLE 1. VARIABLE DEFINITIONS

PENSION: Natural log of the 3-year average of pension assets to
accumulated liabilities.

LEVERAGE: The three year average of the total debt to total assets
ratio.

DIVIDEND: The 3-year average of dividends to operating income plus
pension expense. The average is taken after removing highest and lowest
values in the preceding 5 years.

TAX: The 3-year average of the ratio of taxes (paid plus addition to
accrued) to operating income.

PROFIT: The three year average of the ratio of net income to total
assets.

SLACK: The 3-year average of the ratio of cash and marketable
securities to total assets.

BENEFITS: The 3-year average of the ratio of accrued benefits to number
of employees, in thousands.

SIZE: The firm's size measured by the natural log of the most recent
year's net sales.

R & D: The 3-year average of the firm's R & D expenditures divided by
total assets ratio.

BUS RISK: The standard deviation of the preceding five years' ratios of
the first differences in operating income divided by total assets.

INDUSTRY: Coded 1 if the firm's principal SIC is between 3400 and 4000;
0 otherwise.

GROWTH: The geometric mean of annual growth of sales from five years
previous to current sales.

INVEST: The firm's investment opportunities, measured as the three year
average ratio of the firm's research and development expenditures plus
its expenditures on plant and equipment divided by total assets.

TABLE 2. 3SLS RESULTS FOR PENSION FUNDING

Three stage least squares regression was applied to the system of
equations described. The following results are for the dependent
variable PENSION. The table reports regression coefficients and
their associated t-statistics (in parentheses). An * signifies a
significant result at the 0.05 level or better.

                    1987         1990         1992

CONSTANT             0.2870       0.3004       0.2361
                    (4.232) *    (3.036) *    (3.117) *

LEVERAGE            -1.0165      -0.9334      -0.8131
                    (-5.725) *   (-3.350) *   (-3.344) *

DIVIDEND             0.5238       0.5810       0.5072
                    (9.111) *    (8.705) *    (6.841) *

TAX                  0.7080       0.0002       0.0001
                    (5.686) *    (0.250)      (0.155)

PROFIT              -0.6975      -1.3303      -0.5409
                    (-3.238) *   (-3.029) *   (-2.145) *

SLACK                0.2079       -0.0034      -0.0190
                    (1.743)      (-0.021)     (-0.122)

BENEFITS            -0.4512      -0.9620      -0.1360
                    (-1.289)     (-3.057) *   (-0.616)

SIZE                 0.0106       0.0061       0.0015
                    (2.364) *    (1.562)      (0.464)

Equation R-Square   .0638        .0726        .0397

TABLE 3. 3SLS RESULTS FOR FINANCIAL LEVERAGE

Three stage least squares regression was applied to the system of
equations described. The following results are for the dependent
variable LEVERAGE. The table reports regression coefficients and
their associated t-statistics (in parentheses). An * signifies a
significant result at the 0.05 level or better.

                    1987         1990          1992

CONSTANT             0.3157       0.3090        0.2912
                    (17.513) *   (14.208) *    (11.857) *

PENSION             -0.3438      -0.1836       -0.2033
                    (-5.839) *   (-2.482) *    (-2.084) *

DIVIDEND             0.1653       0.1584        0.1624
                    (6.227) *    (5.331) *     (4.755) *

TAX                 -0.0090      -0.0002       0.0004
                    (-2.174) *   (-0.313)      (0.517)

PROFIT              -0.7253      -1.3535       -0.7602
                    (-8.188) *   (-12.593) *   (-9.927) *

R & D               -0.9495      -0.7244       -0.9467
                    (-4.562) *   (-3.745) *    (-4.740) *

BUSINESS RISK       -0.5706      -0.4507       -0.3687
                    (-1.457)     (-1.049)      (-1.402)

INDUSTRY            -0.0320      -0.0136       -0.0114
                    (-2.806) *   (-1.155)      (-0.762)

SIZE                 0.0078       0.0058        0.0034
                    (2.810) *    (2.201) *     (1.166)

Equation R-Square   .1822        .1998         .1367

TABLE 4 3SLS RESULTS FOR DIVIDEND PAYOUT

Three stage least squares regression was applied to the system of
equations described. The following results are for the dependent
variable DIVIDEND. The table reports regression coefficients and
their associated t-statistics (in parentheses). An * signifies a
significant result at the 0.05 level or better.

                            1987         1990         1992

CONSTANT                    -0.2399      -0.7493      -0.4529
                            (-2.112) *   (-4.800) *   (-3.246) *

PENSION                      1.3147       1.3119       1.7399
                            (7.348) *    (5.574) *    (9.554) *

LEVERAGE                     1.9729       2.5529       1.9199
                            (8.968) *    (6.972) *    (6.895) *

GROWTH                      -0.3196      -0.0644      -0.0886
                            (-3.298) *   (-0.534)     (-0.731)

PROFIT                       1.7422       3.8388       1.4268
                            (4.970) *    (5.942) *    (4.538) *

INVEST                       0.4006       0.3941       0.2451
                            (1.168)      (1.113)      (0.737)

BUSINESS RISK                0.9771       1.9266       0.3962
                            (0.823)      (1.590)      (0.725)

Equation R-Square           .0730        .0752        .0509

Entire system R-Square:     .1883        .1574        .1223

Sample Size (# of firms)    1267         1172         1311

REFERENCES

Barth, Mary E., "Relative Measurement Errors among Alternative Pension Asset and Liability Measures", The Accounting Review, Vol. LXVI (3), 1991,433-63.

Baskin, Jonathan, "An Empirical Investigation of the Pecking Order Hypothesis", Financial Management, Vol. XVIII (1), 1989, 26-35.

Black, Fisher, "The Tax Consequences of Long-Run Pension Policy", Financial Analysts Journal, Vol. XXXVI (4), 1980, 21-28.

Bodie, Zvi, Light, Jay O., Morck, Randall and Taggart, Robert A., Jr., "Funding and Asset Allocation in Corporate Pension Plans: An Empirical Investigation", Issues in Pension Economics, Bodie, Shoven, & Wise, editors. University of Chicago Press, Chicago, 1987

Bowen, Robert, Daley, Lane and Huber, Charles, Jr., "Evidence on the Existence and Determinants of Inter-Industry Differences in Leverage", Financial Management, Vol. XI (4), 1982, 10-20.

Claggett, E. Taylor Jr., "Capital Structure: Convergence and Pecking Order Evidence", Review of Financial Economics, Vol. I (1), 1991, 35-48.

Feldstein, Martin and Seligman, Stephanie, "Pension Funding, Share Prices, and National Savings", The Journal of Finance, Vol. XXXVI (4), 1981,801-824.

Financial Accounting Standards Board, Statement of Financial Accounting Standards No. 36: Disclosure of Pension Information, FASB, Stamford, CT, 1980. Statement of Financial Accounting Standards No. 87: Employers Accounting for Pensions, FASB, Stamford, CT, 1985.

Francis, Jere R. and Reiter, Sara Ann, "Determinants of Corporate Pension Funding Strategy", Journal of Accounting and Economics, Vol. IX (1), 1987, 35-59.

Harris, Milton and Raviv, Artur, "The Theory of Capital Structure", The Journal of Finance, Vol. XLVI (1), 1991, 297-355.

Ippolito, Richard A., "The Economic Function of Underfunded Pension Plans", Journal of Law and Economics, Vol. XXVIII (3), 1985, 611-651.

Jalilvand, Abolhassan and Harris, Robert, "Corporate Behavior in Adjusting to Capital Structure and Dividend Targets: An Econometric Study", The Journal of Finance, Vol. XXXIX (1), 1984, 127-145.

Jensen, Gerald R., Solberg, Donald P., and Zorn, Thomas, "Simultaneous Determination of Insider Ownership, Debt, and Dividend Policies", Journal of Financial and Quantitative Analysis, Vol. XXVII (2), 1992, 247-263.

Miller, Merton H. and Rock, Kevin, "Dividend Policies under Asymmetric Information", The Journal of Finance, Vol. XL (4), 1985, 1031-52.

Myers, Stewart C., "The Capital Structure Puzzle", The Journal of Finance, Vol. XXXIX (3), 1984, 575-592.

Myers, Stewart C. and Majluf, Nicholas S., "Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have", Journal of Financial Economics, Vol. XIII (2), 1984, 187-221.

Rozeff, Michael, "Growth, Beta and Agency Costs as Determinants of Dividend Payout Ratios", Journal of Financial Research, Vol. V (3), 1982, 249-59.

Sharpe, William F., "Corporate Pension Funding Policy", Journal of Financial Economics, Vol. III (3), 1976, 183-193.

Stone, Mary, "A Financing Explanation for Overfunded Pension Plan Terminations", Journal of Accounting Research, Vol. XXV (2), 1987, 317-326.

Tepper, Irwin, "Taxation and Corporate Pension Policy", The Journal of Finance, Vol. XXXIV (1), 1981, 1-13.

Thomas, Jacob K., "Corporate Taxes and Defined Benefit Pension Plans", Journal of Accounting and Economics, Vol. X (3), 1988, 199-237.

Titman, Sheridan and Wessels, Roberto, "The Determinants of Capital Structure Choice", The Journal of Finance, Vol. XLII (1), 1988, 1-19.

Dr. Aaron L. Phillips earned his D.B.A. in Finance at Southern Illinois University, Carbondale in 1986. He is currently an Associate Professor of Finance at California State University, Bakersfield and is a Certified Cash Manager (CCM).

Dr. Sharon M. Moody earned her Ph.D. in Accountancy at the University of Mississippi in 1987. She is currently a Lecturer in the Department of Finance and Accounting at California State University, Bakersfield.