The Atlantic Home

It now seems certain that General Motors will declare bankruptcy on Monday, that the federal government will as part of the reorganization in bankruptcy acquire more than 50 percent of the common stock of the reorganized company, and that the government will invest perhaps $30 billion in the new company, bringing the total cost of the GM bailout to a shade under $50 billion.

These developments raise three questions:

1. Should GM have received any government money?

2. Should the government acquire common stock in GM?

3. Should it invest in GM?

My answer to 1 is a qualified--an importantly qualified--yes. My answer to 2 and 3 is no.

Government bailouts of failing private companies are in general a very bad thing. They weaken the discipline of the market and penalize successful firms by subsidizing their inefficient competitors. But that is in general and there is a case for bailouts in times of national economic emergency, and it is on this ground that I have defended the initial bailout of General Motors and Chrysler, last December. Last December, with the economy in a tailspin that was beginning to remind observers of the Great Depression of the 1930s, the bankruptcy of the two firms would have been experienced as a dangerous shock to the economy. The firms might have had to liquidate (see below), which would at a stroke have increased the unemployment roles by hundreds of thousands, as well as causing ripple effects throughout the chain of distribution of motor vehicles, which includes not only the manufacturers but also the parts suppliers, who supply more than 50 percent of the components of the motor vehicle, and the auto dealers. The effect on employment and therefore on incomes and business and consumer confidence would probably have been profound--and wholly negative. In short, the costs of the initial bailout were outweighed by the costs of doing nothing.

That was the macroeconomic justification for the bailout. There was another, though closely related, justification. As a result of the near collapse of the banking industry, and of financial intermediation generally, last September, lending was severely constrained for some months (it still is severely constrained, though less so). That would have made a reorganization in bankruptcy of the auto manufacturers, as distinct from a liquidation, very difficult to pull off. Bankruptcy is designed to eliminate a crushing debt, but it does not eliminate the bankrupt firm's need for money if it is to continue in business. To acquire the necessary money, the reorganized firm typically obtains what are called "debtor in possession" loans (because in a reorganization the debtor--that is, the bankrupt--remains in possession of the firm's assets), and the lenders are given priority over the old debtholders. GM and Chrysler would have needed tens of billions of dollars of such loans to keep operating, and that money would not have been available, other than from the government, in December; indeed, it seems not yet to be available.

That was a liquidity problem, rather than an allocative-efficiency problem. To understand this, one must distinguish between average and marginal cost. The average cost of producing a car is the total cost incurred by the manufacturer, including interest on its debts, divided by the number of cars it produces. That cost exceeded, and still exceeds, the price at which GM can sell an appreciable number of its vehicles. GM's marginal cost, however, is the addition to its total costs of producing one more vehicle. That cost does not include interest on existing debt, because the interest is a fixed amount rather than varying with how many cars GM builds and sells. (It is the fixity of its debt that made GM insolvent when the demand for its vehicles plummeted last fall.) As long as GM's marginal cost is less than the price it can get for its product, it is efficient that it should remain in business. So if it cannot get the cash that it requires to be able to remain in business, then an efficient production process will be shut down, and that is an inefficient result. It therefore made sense, regardless of macroeconomic consequences, for the government to step in and take the place of the temporarily constrained banks, and become GM's (and Chrysler's) banker. 

But that was then. Five months later, credit is less constrained, and so GM's inability to obtain debtor in possession financing (if indeed it is unable to do so, as I shall assume) may reflect not a liquidity problem but a judgment by the financial industry that GM has no long-run future and therefore could not repay a debtor-in-possession loan large enough to keep the company going. Moreover, these five months have seen a partial, and because gradual an orderly partial, liquidation of GM and Chryster, which has diminished the shock value of their declaring bankruptcy; and they also seen a reduction in panic as the economic situation has stabilized. Not that it has fully stabilized--indeed, it may not really have stabilized; but there is no doubt that business and consumer confidence has increased, and so bankruptcies that would have been real shockers five months ago can now perhaps be taken by the economy in stride. That certainly has been true with respect to the Chrysler bankruptcy, though Chrysler is a much smaller company than GM and the government was able to coerce the agreement of virtually all the creditors, so that the bankruptcy proceeding has proceeded at a lightning pace.

Having already shrunk a good deal, GM might if the government washed its hands of it be able to obtain enough private money to keep going. Or perhaps not; and if not, and liquidity is still a factor (that is, if lending is still inefficiently constrained because the banking industry is still undercapitalized), there is an argument for a further bailout, on the ground that without it GM might liquidate entirely and the effect on employment in both GM and its satellite firms might deal a body blow to what we hope is an incipient recovery from the depression. I am skeptical, but will assume for the sake of argument that there is a convincing case for a further bailout.

It should not, however, take the form of cash for common stock that makes the government the owner of GM. It should be a loan in exchange for preferred stock with cumulative dividends (so that it does not affect GM's current cash flow) and no fixed date of maturity (so that it does not enable the government to step ahead of other creditors). For the government to own an automobile manufacturer doesn't make any sense, for reasons too obvious to dwell on.

The government says that it's not going to interfere in management decisions. I don't believe that. Quite apart from the political pressures that the United Auto Workers, and other entities that have a financial stake in General Motors, can be expected to exert on members of Congress and on the President, the Administration seems determined to preserve General Motors in order to (1) affirm the nation's commitment to remaining a major manufacturer of motor vehicles and (2) advance the Administration's goal of reducing oil consumption and (relatedly) carbon emissions. Achieving these goals will require the Administration to intervene, directly or indirectly, in the design and production and even marketing decisions of GM's management.

Goal (1) is ridiculous. Nowhere is it written that the United States shall produce motor vehicles, any more than that it shall produce television sets, which it no longer does. If other countries, such as Japan, produce better motor vehicles (from the standpoint of price and quality--and of the health of the environment and of reducing our dangerous dependence on oil produced by unstable or hostile foreign countries) than the United States, we should import them, and reallocate the resources that go into the manufacture of motor vehicles to other productive activities.

But what makes goal (1) particularly ridiculous is that even if GM (and Chrysler) liquidated, there would be a thriving U.S. automobile industry. There would be Ford''s production, and there would be the foreign cars that are manufactured in the United States. Toyota and the other foreign producers that have factories in the United States are part of the American automobile industry; their ownership by foreigners has no significance at all.

Goal (2)--operating GM in accordance with the environmental and foreign-policy goals of the Administration (which I happen to agree with)--confuses regulation with ownership. The government can without owning the manufacturers require them to comply with rules designed to make driving less harmful to the environment and to our need to reduce our dependence on foreign oil. All that the government's owning GM will accomplish is to make the company a political plaything.

And then there is the $30 billion that the government apparently is planning to pour into GM so that it won't just limp along but will become a vibrant, revitalized producer. We are becoming accustomed to thinking of anything less than a trillion dollars as small change. But as the government's loans, investments, and guaranties mount into the stratosphere, the danger of the "depression aftershock" that I emphasized in my book and in a number of my blog entries grows, and a further $30 billion expenditure deserves critical scrutiny. We should be concerned lest GM become a kind of economic Vietnam, where the federal government throws good money after bad, year after year, in a vain quest for victory.