Corporate Finance

Which cross-sectional patterns are consistent with the trade-off theory?

• More stable firms face lower bancruptcy probabilities and thus have higher optimal leverage
• Leverage is positively correlated with asset tangibility, and negatively correlated with R&D intensiveness

Which cross-sectional patterns are not consistent with the trade-off theory?

• There is a negative relation between profitability and leverage (more profitable firms should more highly value the tax-shield benefits of debt)
• Many firms have zero leverage despite facing heavy tax burdens and low distress risk

mismeasurement of the dependent variable, leverage:

• LEVERAGE: Should leases be included in leverage ratio calculations?
• DISTRESS COSTS: The probability of distress as measured by historic default occurrence is small, leading to low costs to distress, implying that observed debt ratios are smaller than optimal debt ratios implied by trade-off models.
  • PROBABILITIES: Distress occurs in bad times, when utility of a dollar is high, so one should not use standard probabilities to calculate expected distress costs but use a risk-adjusted probability (ca. 4,5% of firm value with a BBB rating)
• NON-DEBT TAX SHIELDS: Substantial non-debt tax shields (e.g. pension contributions) that are missed by research based solely on standard financial statement data. Off-balance-sheet deductions from tax shelters (large), so firms would not find incremental debt interest deductions very valuable. 

Suppliers and customers: companies use less debt when their suppliers are dedicated (supplier sells much of its output to one customer). Customers consider their own financial distress risk as a cost to the supplier

Employees and risk: employees are exposed to unenployment risk in the event of bankruptcy (indirect cost of bankruptcy which is ultimately borne by the
company in the form of higher wages), discourages the use of debt.

Until now, consistent with MM, much of the existing literature on capital structure has assumed that capital supply is perfeclty elastic and that capital structure is solely determined by corporate demand for debt.

BUT

• CREDIT SUPPLY: There's a segmentation in debt markets between bank-dependent firms and those with access to arm's-length lenders.
  • Firms with a credit rating have a debt ratio that is significantly higher than those without. Firms without access to the bond market face a different supply schedule than those with access.
• EQUITY MARKET TIMING: Managers attempt to exploit deviations of security prices from fundamental value.

• Traditional models take the types of securities as given and ask whether an optimal mix of these securities enhances firm value.
  • Newer literature derives optimal contracts between manager/owner and investor. The resulting empirical questions go beyond studying leverage ratios and focus on more detailed features and types of contracts, such as covenants and the potential for renegotiation.
• The contracting literature highlights the fact that not all debt is equivalent, as is often implicitly assumed in constructing leverage ratios (e.g., collateral and asset redeployability).

• Contribution to firm value of optimal capital structure is modest for most firms.
• Large deviations from optimal capital structure might be costly, but firms might have little incentive to engage in capital structure policy.
• Empirical evidence suggests that for up to half of all public firms, the value contribution of optimal capital structure is modest.
• For some firms, the optimal financing choice is quite valuable because some capital structure situations (e.g. excessive debt) can have disastrous effects.

• One explanation for the relatively poor performance of traditional trade-off model is that firms might be moved away from target leverage ratios. In this case, proxies for optimal leverage determinants will have low explanatory power.
  • However, if firms actively manage leverage towards a target, we should see evidence of a return to a target following shocks to leverage.
    • BUT Firms respond slowly (at best) to deviations between actual and target levels.

Dynamic trade-off models suggest that empirical evidence might be more consistent with the trade-off theory than previously recognized. 

1. Non-dividend paying firm
2. Profitable firm

Problem

Empirically determining the effect of one variable is very difficult!

Problem of endogenity -> you might run a multiple regression and there's another variable you didn't consider , maybe 'cause you didn't know about it... e.g. when testing whether coffee is healthy, everything else about the persons is important too (do they do a lot of sports, do they smoke etc) 'cause that might influence on the one hand the overall health of the test subject and on the other hand the consumption of coffee etc...

-> Watch out that you measure the right effect!!

Solution

Exogenous shock

People who drink coffee don't randomly do so, they select whether they do it or not (depending on their attitude towards health etc.). If everybody was forced to drink 3 cups of coffee today, the self-selection problem would be gone

Heider & Ljungqvist

They applied this solutions to the financing problem: Does the trade-off theory work in practice?

1. Run a regression
     Problem: Tax rate is not randomly distributed across companies

2. Use of an exogenous shock: the North Carolina government raised its corporate tax rate

3. Comparison of firms affected (North Carolina) and firms unaffected (South Carolina)

---> READ PAPER AND LOOK AT LECTURE SLIDES (5 DECKBLATT)

• Pay out a dividend
• Purchase shares

Payout Policy is affected with whether payouts to investors affect the market value of the firm

MM Proposition III: 

With perfect capital markets and no taxes the value of the firm is independent of its payout policy

Solution to example:

You invest in A and B and pay out the difference (10) to the investors

Arguments

1. Firms and investors can borrow at the same rate
2. Without taxes, there's no difference between dividends or purchasing shares (most important argument)
3. Risk of firms and investors is the same

Example: company with $100 cash and 100 share and 1 shareholder, 1 manager

In the 2nd year, lazy manager wins $100 for the company in the lottery (thus $200 cash, $2/share), he has 2 options:

1. Pay a dividend of $100: $100 equity and $100 cash dividend
2. Pay no dividend: $200 equity

Investor might get upset when there's no dividend BUT he can just pay himself a dividend by selling 50 shares and receiving 50x$2=$100 and he still owns 50 shares à $2=$100

-> In an MM world, the manager doesn't have to care about the shareholder, he can do that himself

1. Taxes
2. Agency Costs

• (In the US) the tax system makes a distinction between dividends and share repurchases
• Dividends are classified as ordinary income. Ordinary income is taxed at a rate which depends on an individual’s tax bracket. The tax is paid in the year in which the dividend is received.
• Share repurchases lead to capital gains. Capital gains are taxed at the capital gains tax. The capital gains tax is paid when shareholders sell their shares.
• Tax rates vary over time. In the past, capital gain taxes were typically lower than income taxes

• less
  more
• As a result, shareholders tend to sort across firms depending on a firm's payout policy.
  low
  high
  high

Also, individuals can offset capital gains with losses on other investments. This can generate individual variation in one's preferences for payout policy -> optimal share depends on tax bracket and on your investments, your losses etc.

• It's a lot of effort AND in reality there are pretty high transaction costs when selling shares
• Pension funds are not taxed at all and are among the largest investors in the stock market -> related to tax, they don't have a preference for either dividends or repurchases
• They prefer dividends cause they pay less taxes on dividends than on capital gains

Some observers have argued that regular dividends reduce the scope for management to invest in negative NPV projects

Companies which pay low dividends will be more attractive to highly taxed individuals; those which pay high dividends will have a greater proportion of pension funds or other tax-exempt institutions as investors.

These financial institutions are sophisticated investors who monitor carefully the companies that they invest in and they bring pressure on poor managers to perform.

Well-managed companies are happy to have financial institutions as investors, but poorly-run companies prefer to have passive investors.

Signaling: Well-managed companies want to signal their worth! They can use dividends for that!

Those shareholders who pay taxes do not object as long as the effect is to encourage institutional investors who are prepared and able to put the time an effort into monitoring the management.

A credible signal is an action that is less costly for a good agent than for bad agent. Good agents will be able to undertake the action so that they can be recognized, but it will be too costly for bad agents to mimic them.

In a sense, the good agent can be though as saying: “You must believe I am a good agent. If I were a bad agent, the costs of undertaking an action would have been so large compared to the benefits, that I would not have taken it.”

Examples:

• Job application: everyone says that he/she is a good employee. One signal is education, it's a credible signal but it's costly (time and opportunity costs...)
  • You should try to build up a signal that's very expensive for the bad agent but cheap for yourself
• Bargaining: you misuse signals and start acting
• Company ratings: you have to pay for it but it's a signal to the investors
• Insurance company uses your past driver data as a signal to price your premium

How can a firm do signaling with dividends?

-> A firm’s dividend policy can be used to signal information about the firm’s prospects to the market.

• Managers want the market to think highly of their firm, but cannot simply tell the market that their firms have good prospects. Such claims would not be credible.
  • Paying a high dividend is an example of a credible signal when external finance is difficult to obtain
• The cost of the signal is that by increasing the dividend the firm might have to pass up some positive NPV projects.
  • If a good firm (one with high future cash flows) pays a high dividend, it will only pass up a few positive NPV projects.
  • If a bad firm (one with low future cash flows) pays a high dividend, it will pass up a lot of positive NPV projects

-> Thus, it is costlier for a bad firm to pay a high dividend

Is signaling by dividends a good thing for the firm?

• A firm may increase dividends even when it means foregoing positive NPV investments
• The market will react to the increased dividend by raising the firm’s stock price.
• The act of raising the dividend may thus destroy firm value while raising the stock price
• Perhaps it is better to simply wait until the information finds its way to the market by other means (e.g., future earnings announcements)

BUT some information will never get to the investors, such as behavior of the manager -> there0s always private information -> you need signals!

And signals are costly but if they are free, they're useless because everyone could use them!

No, it's not!

In 1984, ITT announced a 64% cut in its dividend. The firm claimed that it was conserving cash to enable it to fund profitable investments. The market reacted by lowering ITT’s stock price by 32%. The value of the firm’s equity fell
by roughly $1 billion

-> this cannot be only because of dividend cut BUT it's because the cut is a signal.
Thus, firms usually avoid negative signaling effects and try to keep dividend payouts constant (no immediate adjustments, also not on the upside)

-> This market response is typical: stock prices usually rise after dividend increases and fall after dividend decreases

1. No. E.g., in 2007, more than 2/3 of firms that paid a dividend also repurchased stock
2. when they have accumulated a large amount of unwanted cash or wish to change their capital structure by replacing equity with debt
3. No. They incur a commitment to maintain the payout in the future
4. Repurchases are more volatile. They usually fall sharply during crises / recessions
5. When companies offer to repurchase their stock at a premium, this may indicate an undervaluation of the stock

Which type of company is more likely to pay a dividend and to engage in share repurchases? Which type is less likely to do so?

• Many smaller, high-growth firms do not pay a dividend. They are also less likely to engage in share repurchases.
• Many older, low-growth firms pay a dividend. They also engage in large share repurchases
  • Why is that so?
    • If you're a high-growth company, you don't have cash to "waste" for dividends but you have to invest in order to grow
      • Paying dividends if you're a high growth company might even be a bad signal, it looks as if you didn't have enough projects in your pipeline that you could invest in
    • If you're a low-growth (value-) company and you have a lot of cash but not a lot of investment opportunities, you have a higher propensity to misuse the cash. Thus, an old firm uses dividend payments as signals

Why are firms trying to keep paying the same dividend?

• Cutting dividends is a very bad signal
• Your clientele might have chosen its firms because of the dividend policy that matches its desire -> Regularity is important! Otherwise investors get annoyed if they have to rebalance their portfolios every year

The results of Fama and French suggest that a lower propensity to pay is at least as important as changing firm characteristics (i.e., more small, less profitable, highgrowth firms in the sample).

-> Can share repurchases explain this trend, i.e., are dividends substituted for by share repurchases?

This seems not to be the case:
-- Share repurchases jump in the 1980s.
– In general, firms repurchasing shares also pay dividends.
– The primary effect of these increased repurchases is to increase the already high earnings payouts of dividend paying firms!

Denis and Osobov (2008)

• The propensity to pay dividends is higher among larger, more profitable firms, and firms for which retained earnings comprise a large fraction of total equity.
• Although there are hints of reductions in the propensity to pay dividends in most of the sample countries over the 1994-2002 period, they are driven by a failure of newly listed firms to initiate dividends when expected to do so.
• Dividend abandonment and the failure to initiate by existing nonpayers are economically unimportant (except in Japan).
• In all sample countries, aggregate dividends have not declined and are concentrated among the largest, most profitable firms.
• Overall, they argue that their findings support agency cost-based lifecycle theories (older firms with more free cash flow pay out more).
• -> Side note: They find no evidence (except for the US) of a positive relation between relative prices of dividend paying and non-paying firms and the propensity to pay dividends (catering theory: firms pay dividends depending on the market’s sentiment, you tailor your policy towards your clients)

---> SEE LECTURE SLIDES!!!

1. Run a regression (e.g. logit regression with dependent variables 0 + 1)
2. Take the positively correlated parameters you found
3. Look at the actual firm characteristics and multiply them with the coefficients

The principal (shareholder):

• Does not have the expertise to run the firm
• Is usually a large number of dispersed individuals / institutions which prevents them to act in concert and exert control on a daily basis
• Provides the financing

The agent (manager):

• Runs the firm for the principal

“Being the managers of other people’s money…it cannot well be expected that they should watch over it with the same anxious vigilance […]” (Smith, 1776, p. 700). 

The problem with this separation of ownership (by the shareholders) and control (by the managers) is that the interest of the two groups may not necessarily coincide at all times.

• The shareholders of a firm simply want the value of their shares to be as high as possible, i.e., they want to be compensated for providing risk capital to the enterprise.
• The managers, while certainly not inherently adverse to a high share price, may have other interests.

1. Managerial shirking (i.e., playing golf instead of working)
2. Managerial consumption of perks (plush offices, corporate jets, etc.)
   -> Whether such expenses / investments hurt (or benefit) shareholders is often difficult to assess!
3. Manager's desire to remain in power
   • As we all, top managers do not like to lose their jobs. If the right management team is not the one that is currently in power, a serious conflict of interest between managers and shareholders may result.
   • Managers tend to invest in projects/assets that they are experts in because company won't fire them since they're the experts; otherwise, someone else might be!
4. Mangerial risk aversion
   • The manager has usually all of his human capital tied up in the firm and, due to restricted stock and option grants, also the majority of his financial wealth. Hence, a manager may be overly risk-averse and unwilling to take on projects that are worthwhile from the shareholders’ point of view.
   • Manager might hold too much cash as a cushion against bad times while shareholders would prefer the money to be paid out
5. Free cash flow and empire building
   • Free cash flow is the cash flow generated by a firm in excess of the amount required to fund all available positive NPV projects. What to do with this FCF presents serious potential conflicts of interest.
   • Some managers spend that money on expensive cars etc. But that's only possible if a lot of FCF exists!

• Companies, in which the CEO uses a company jet for private purposes, underperform by approximately 4% p.a. on average (sample of 237 large US firms from 1993 to 2002).
• On the day of the announcement of the private jet use by the CEO, the stock price drops by 1.1% on average.
• The negative announcement effects exceed the costs of the private (and even total) airplane use by far.
• Hence, shareholders seem to perceive the news of the private jet use by the CEO as a signal for a bad corporate governance structure more generally

In the case of Armand Hammer, the price of Occidental stock increased from $28 to $31 per share, representing a total gain in shareholder value of approximately $300 million…

If a valuable managers (no/less agency problems) falls ill, the price drops!

The effects are much higher if the CEO was also the founder because a founder usually has more power (CAR = Cumulative abnormal return)

1. Bond managers contractually to behave in shareholders’ best interest

• This is only a partial solution, at best, as it would require to spell out every possible eventuality and specify what action a manager should take in that situation. Moreover, often is even ex-post difficult to assess what would have been the value-maximizing decision at the time.
• Contracts may provide crude guidelines for manager behavior and help to ensure that managers will not take certain actions which harm the principal or that the principal will be compensated if he does take such actions.
• Purpose of a manager is to flexibly adapt to new situations; contracts make that harder, they reduce the real options of a firm!

2. Monitoring of managers

• Monitoring requires effective monitors who present credible threats tomanagement. The obvious monitor would be the shareholders. But: most shareholders lack the necessary expertise and the incentive to monitor!
• More effective monitors are:
  • The board of directors
  • Large shareholders
  • Competing management teams / companies (they might take you over and replace the management because they know that the reason for the low firm value is only the management)
    -> Managers need to be aware that monitors are powerful!

3. Alignment of incentives between managers and shareholders

• By providing performance dependent, and in particular stock-based, compensation, managers become shareholders and the interests between principal and agents are (better) aligned.
• But if manager does very bad and economy rises (external effects), he earns a lot anyway and if he destroys firm value, you still pay him ('cause you give him shares)
• You could use stock options with a strike price above the current stock price, so the manager only earns something if he performs really good.
  • But option values not only rise with the value of the underlying but also with volatility!
    • Manager could simply increase volatility (take on very risky projects...) which might be bad for the company

1. The monitoring expenditures by the principal (monitoring the behavior of the agent, budget restrictions, compensation policies, etc.)
2. The bonding expenditures by the agent (e.g., contractual guarantees to have the financial accounts audited by a public account, limitations on the managers‘ decision making power, etc.)
3. The residual loss

• The principal and the agent will incur positive monitoring and bonding costs (non-pecuniary as well as pecuniary).
  • Still, there is a divergence between the agents’ decisions and those decisions which would maximize the welfare of the principal.
    • The Dollar equivalent of the reduction in welfare experienced by the principal due to this divergence: “residual loss”.

-> To raise new capital (primary offering) and/or to allow existing shareholders to cash in by selling part of their holdings (secondary offering)

But there are further motives (see picture)

.