Determinants Of Corporate Borrowing

Stewart C. Myers, Journal of Financial Economics 5 (1977) 147-175

Introduction

The existence and determination of optimal capital structure is an ongoing topic of research in corporate finance. In a perfect market setting with no frictions Miller and Modigliani showed in their famous Proposition 1 that the market value of a firm is independent of its capital structure, in other words capital structure does not matter. Of course corporations do not operate in a perfect world, and one of the most obvious violations is corporate taxes and the tax deductibility of interest.

If interest is tax deductible, it appears to be cheaper to finance a firm with debt than with equity. Consider a simple example, a firm taxed at rate T is able to borrow or lend from investors the same amount of funds, F, at the same expected rate of return, r. The firm has earnings before taxes, interest and dividend payments of C, how much remains after interest and the government are paid?

Debt financing: (E – rF)(1-T) = E(1-T) – rF + rFT
Equity financing: = E(1-T) – rF

Clearly the residual earnings of debt financing are larger than that of equity financing because of the term rFT, which is called the debt tax shield. Therefore, it seems advantageous that a firm be financed completely with debt in order to maximize these residual payments. However, we never observe in practice firms with 100% debt financing, which leads to the main question of this paper, why do firms not borrow as much as possible?

Prior to this paper, many researchers utilized other market frictions and uncertainties like personal taxes, uncertainty of the ability to fully realize tax shields, and bankruptcy costs to explain this phenomenon. Myers reasoning in this paper is that the existence of debt causes managers to reject positive NPV projects that do not earn quite enough to provide payments to shareholders. Students of finance know that managers are tasked to maximize the value of shareholder wealth, which usually requires them to undertake all positive NPV projects. Myers says that suboptimal investment policy by managers is a cost that occurs whenever debt holders are uncertain whether or not they will be paid. Therefore, at some level of debt, the costs of these bad decisions are more expensive than the benefit of the tax shields and thus it is optimal for firms to limit their borrowing to less than 100%.

We will utilize an example to explore how and when debt leads to suboptimal decisions by managers acting in the interest of shareholders. We will consider three versions of our example to illustrate the suboptimal decision making.

Example 1: All Equity financing

Consider a project that pays out an amount dependent on what state of the world, s, occurs, for example what the temperature is or what the price of oil is. On day 1 managers must decide whether or not to invest the amount I in order to obtain the payout V(s). The managers have no money with which to invest now, so if they decide to invest, they must issue shares to obtain I to “buy” the project. The shareholders will receive all proceeds of the investment if the managers invest in the project. If on day 1 the managers know exactly what the payout of the project, V(s), will be, their decision is simple: invest if V(s) ≥ I. In figure 1, sa is exactly the breakeven point where V(s)=I.

Investment rule: Invest if V(s) ≥ I

Figure 1: All equity

MyersFig1.JPG?imgmax=512

Example 2: Some debt financing

Now consider the case where the firm has some money (but not the total I) to invest from debt issued before the value of the project is revealed. The payment promised to the debt holders (principal and interest) is P, where the debt matures after the state of the world is revealed. Now for the shareholders to receive any proceeds of the project, V(s) must be greater than I+P, since the bondholders are paid first. Note that this debt is “risky” because there are states of the world where the bondholder will not receive P. In this case the investment rule for managers has changed to invest if V(s) ≥ I+P, and in the new graph, sb is exactly the new breakeven point where V(s)=I+P. A simple numerical example may help.

The investment required is I= $10, of which $6 is available from existing debt, and an additional $4 would be required from shareholders. The promised payment to bondholders is P=$8. It is clear that if V(s) = $13, the bondholders receive their promised $8 and the shareholders would receive the residual $5. This $5 is greater than the $4 the shareholders would have to pay for the shares, thus the shareholders are interested in this project and would invest. If however, V(s)=$11, which is still a positive NPV project (NPV=V(S) – I=$1), bondholders would receive their required $8, but the shareholders would receive the residual $3 in exchange for their $4 investment in shares. Obviously the potential shareholders aren’t interested in this project, and prefer the managers default on the loan, in which case the bondholders gain control of the firm. Unlike in all equity financing, now shareholders will only invest if sb ≤ s.

After default the bondholders control the firm, with assets consisting of the original $6 loaned the firm. If the bondholders also still have the ability to execute the project after the managers default, the bondholders would follow the same rule as with the all equity financing in example 1, invest if V(s) ≥ I, which again means invest if s ≥ sa. The bondholders then are the ones who receive the benefits of the project if sa ≤ s ≤ sb.

Therefore the value of the project under this set-up is the same as under the all equity rule, because it will always be executed as long as NPV>0, the only change is who will realize the investment cash flows. The value of the firm is divided as shown below between the value of the debt and the value of the equity. The purple is the value to the debtholders if the firm defaults, if the managers execute the project, the bondholders receive the blue representing payment P. The yellow is the value of the option for the shareholders.

Manager Investment rule: Invest if V(s) ≥ (I +P)
After Default Bondholder Investment rule: Invest if V(s) ≥ I

Figure 2: Value with Debt

MyersFig2.JPG?imgmax=512

If the debt matures before the option to invest expires, no project value is lost, but what if the option expires as soon as the managers reject the project? If this were the case, then the option to invest in the project has no value to anyone if s ≤ sb because the equity holders do not want to invest, and the debt holders cannot invest. Note that the shareholder’s investment rule is still to invest if V(s) ≥ I+P. In this case, a positive NPV project is essentially wasted in some states of the world and therefore the overall option to undertake project must lose value. Our firm, which consists only of the option to invest in the project, is therefore also worth less in time 0 with any level of debt financing than it would be under all equity financing. The difference between the two is the purple triangle of value lost between sa and s b. It is apparent from figure 2 that raising P, which occurs as the use of debt increases, lowers the value of our firm at an increasing rate. We also are willing and able to make our promised payment P in fewer states of the world, which at first the debt holders may be willing to accept because they would be receiving more money in the remaining states of the world. But at some level of promised payment P, they are no longer willing to make this tradeoff and the value of the debt also decreases.

Figure 3 shows the changing value of the firm and the debt with increasing levels of P. Again as noted, with any level of P>0, the value V is less than the all-equity financing value. Also the value of the debt reaches a maximum at some level P, therefore no investor would be willing to issue the firm debt with payments at a level higher than the level P which maximizes the value of the debt to the debtholders.

Figure 3: V maximized at P=0 => Debt=0

MyersFig3.JPG?imgmax=512

Tradeoff between Costs and Benefits of Debt

We have shown debt to have a negative effect on the value of the firm, but initially we discussed how the tax shields provided by debt (absent any other costs) have a positive effect on firm value. The combination of these two effects justifies the existence of a mixture of debt and equity in capital structure, as observed in the market place. With low levels of debt, the value loss of managers refusing positive NPV projects may be offset by the value gained from the debt tax shield. But as debt is increased, the value lost from suboptimal decision making overshadows the gains of the tax shield. Therefore as you can see in Figure 4, an optimal level of debt for a firm exists, and is less than the full value of the firm. Thus, we have answered the main question posed by the paper.

Figure 4: With corporate taxes, an optimal level of debt maximizes the firm value

MyersFig4.JPG?imgmax=512

The paper also utilizes this set-up to explain other observable firm financing behavior. For example, note from figure 4, it can be shown under feasible assumptions that the optimal debt value is always less than the maximum that lenders would be willing to loan the firm. Another insight is that the problem of suboptimal decisions pertains to projects that give management discretion whether or not to invest in the future. If we consider the value of a firm as the value of its existing assets combined with the value of future growth options, this theory would predict that the suboptimal decision making would be most severe for the growth options, as assets in place do not require future investment. Therefore mature firms with few growth options would face lower potential for suboptimal decisions that hurt bondholders and thus their assets would support a higher level of debt than firms whose assets are primarily growth options. Similarly firms with valuable growth opportunities would have higher value with less debt in their capital structure.

If the problem with debt is managers making decisions that aren’t optimal for the value of the firm (rather are optimal for current equity holders), there should always be something that could be done to better align management decisions with optimizing firm value. However, this avoidance is costly. Consider a preemptive debt contract that forces management to undertake all positive NPV projects. The problem with this approach is management determines NPV, so if the project is not in shareholder’s best interest, they could tweak assumptions to make the NPV appear negative. Bondholders would not know about this manipulation unless they paid for monitoring of management behavior. Many other potential solutions, like mediation, would require similar costly monitoring.

Remember the problem also did not exist if the debt matures before the option to invest expires, so if firms shorten debt maturity they may be able to avoid this problem. While shorter term de bt appears to be the best solution, it requires continuous and gradual renegotiation, and maintaining such a relationship with debt holders may be costly.

Conclusion

In conclusion, we observe that firms with risky debt in their capital structure will follow a different investment rule than those with all equity financing. But at lower levels of debt the suboptimal decisions of debt financed firms are less costly to the firm than gains realized by the debt tax shield. Therefore, firms have an optimal level of debt financing that is related to the debt support optimal for the type of assets in their portfolio, whether they be growth assets or assets in place. Growth assets support less debt than equity assets because managers still have the opportunity to make bad decisions when they have the option to invest in the future.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License