Undercapitalization and Insurance

This is a kind of off-the-cuff post about capitalization of businesses.

The simplest way to think of corporate finance is to divide investors into two categories. You have bondholders (or lenders) who get a specified rate of interest in exchange for their up-front investment. Then you have shareholders (or owners) who get everything that is left over once the business has paid its costs (raw materials, labor, taxes, and so forth, as well as principal and interest on the bonds). Another way to put it is that the company's owners are residual claimants—they don't get a defined return, they simply get the surplus after all the fixed costs are paid. This implies that the marginal dollar goes to the shareholders. That is to say, in ordinary circumstances, if the firm earns an extra dollar of profit, the dollar accrues to the shareholders. Likewise, if profits fall by a dollar, the shareholders get less than they would have. Typically the residual claimants are entitled to make decisions about the way the company will be run. (So for instance, in a public company the stockholders elect the board of directors, who hire the company's managers and have input into important decisions like mergers.)

As I said, that's a simple way to think about it, but it leaves out a lot. You can introduce various tweaks to get the description closer to reality. I'll introduce two.

First, the owners/shareholders aren't always the residual claimants. If there's a shortfall of funds so that not all the fixed costs can be paid, the lenders may not get paid in full, meaning that the marginal dollar now goes to lenders (and other creditors) rather than to shareholders. This is particularly important in light of the fact that the shareholders generally control the company (albeit indirectly).

Imagine a company that in a typical year has $50 million of fixed costs (including principal and interest on its debt) and earns revenues of $65 million. This company could pay its shareholders $15 million per year in dividends or whatever. Now imagine the company has the opportunity to embark on a project that has a 50% chance of increasing its revenue by $50 million per year and a 50% chance of reducing its revenue by $50 million per year. Of course the latter possibility will bankrupt the company because it won't be able to pay its fixed costs. But the shareholders might nevertheless find it to be an attractive project because they will either get $65 million per year or $0 per year, for an average expected return of $32.5 million per year. That's better than the $15 million they're currently getting. (Of course it depends a little on how risk averse the shareholders are, but you'd have to be pretty risk averse to turn down a coin flip that more than doubles your expected return.)

Why are the shareholders' expected returns better than the company's overall expected return? Because a large portion of the losses can be shunted off onto the company's fixed claimants (its workers, suppliers, bondholders, and other creditors), while the gains all accrue to the shareholders.

This is a known problem and lenders can protect themselves by contract. Loans will often contain restrictive covenants that prevent companies from embarking on projects like the one described. It may be more difficult for workers and suppliers to protect themselves, but it is at least theoretically possible for them to protect themselves contractually. (Some protections are written into law. Most states have various workman's liens that attach to any property improved by a contractor.)

Of course there is a category of people who can do almost nothing to protect themselves: tort claimants. Someone who is injured by a company's actions can't be protected by contract because it's impossible to identify who the parties to an accident will be before it happens.

This means that companies have a bad motive for operating on a somewhat undercapitalized basis. What I mean is that a company with lots of capital can absorb large losses caused by tort claims. Those losses come out of the pockets of the shareholders (in an economic sense stemming from the shareholders' status as residual claimants on the company, not in the sense that shareholders are personally liable for the damages). But a thinly capitalized company—one that will be insolvent if it encounters even a medium-sized loss—puts some of the risk onto its victims. If it does well and doesn't injure a lot of people, everyone will be paid in full. If it does poorly or injures a lot of people, it goes bankrupt and its fixed claimants pay the price. For the bondholders, this is generally a negotiated-for outcome. If you don't like the bond indenture's financial covenants, then don't buy the bond. When a company is known to be thinly capitalized, its bonds tend to bear a relatively high rate of interest to compensate the bondholders for the additional risk. But tort claimants get no such protections and (of course) earn no interest until they've obtained a judgment against the company (at which point I believe they would earn statutory interest).

When it is predictable that there may be large tort claims against a company, there are several policies that can protect the tort claimants. One is simply to require companies to be well capitalized. This is difficult because it is hard to know what counts as "well capitalized" and because capitalization itself is a function of the company's reported financials, which can be manipulated.

Probably the better approach is to require the company to obtain insurance policies that protect anyone injured by its operations. The premiums for those policies amount to a fixed cost of doing business (that is, at the margin they come out of the shareholders' pockets). The insurance company is itself subject to capital regulation, and it can protect itself by contract. In fact the insurance company amounts to a sort of informal third-party monitor of the insured company's activities, because it will raise its premiums if it comes to believe that the company's activities have become more risky over time. To put it another way, requiring insurance coverage means that there is a sophisticated entity charging a market price and inserting whatever contractual protections it considers prudent to control the risk that the insured company will injure people.

It's not a perfect solution. Getting the required level of coverage right is tricky. Historically, companies that used asbestos often exhausted their insurance coverage before compensating all victims. But it's often the most effective solution.