President-elect Donald Trump’s deal with Carrier rescued some 800 jobs at a cost of some $7 million in additional subsidies. It was immediately attacked, often by commentators whose devotion to the free market had never previously been detected. In reality, the Ricardian free trade doctrine is an abstraction that does not work well in the real world, just as was Thomas Mun’s mercantilism. If Trump structures some new rules, and doesn’t just proceed case-by-case, he may develop new economic truths that will serve us better.

For the newly-free-marketer critics of Trump’s Carrier deal, I have one question. If the $8,750 per Carrier job (actually $875 per year for 10 years) Carrier subsidy is so obviously bad, why was the net $11.2 billion loss ($162,000 for each one of the company’s 69,000 U.S. jobs) or the gross $49 billion cost ($710,000 per job) of the 2008-09 General Motors bailout acceptable, It would seem to me that a high-skill manufacturing job at Carrier is just about as valuable as a high-skill manufacturing job at GM, so if those jobs can be saved for the United States at a small fraction of the cost per job of the GM bailout, that is surely desirable.

Of course, in principle one would not subsidize companies to put jobs in particular places. Similarly, David Ricardo’s Doctrine of Comparative Advantage is correct in claiming that if widgets can be made cheaper in country A and grommits in country B, then country A should specialize in widgets and country B in grommits, whatever the effect of that decision on the inhabitants of each country. But both statements are of a mathematical ideal, in a world economy with no friction, no nationalism, no subsidies by other countries and no externalities. In the real world, friction, nationalism, foreign subsidies and externalities all exist, so a hard no-subsidies rule and pure Ricardian optimization do not work very well.

That’s not to defend the opposite positions, of government subsidization based on political criteria and rampant Smoot-Hawley protectionism. The failure of General Motors was a huge political embarrassment, but a subsidy of hundreds of thousands of dollars per U.S. job was unnecessary and unjustifiable. Even worse was the $185 billion bailout of AIG, where relatively few jobs were involved, and the collateral activity of bailing out the CDS market and providing a spurious $13 billion to Goldman Sachs has weakened Wall Street’s incentives for decent behavior even further. If subsidies and tariffs are decided on a political basis, they will be badly decided and economically very costly.

There is thus a logical Trumpian position on bailouts and subsidies: if through a bailout or subsidy of less than say 10% of the salaries of the workers involved, a company can be bailed out of prevented from leaving the United States, that bailout is probably justified. At that level, the tax and social security contributions payable by the workers, and the unemployment, retraining and disability benefits avoided, almost certainly add up to more than the cost of the subsidies. In addition, there would seem little problem in the President or local Governors jawboning companies that are seeking to outsource production from the United States. Adding a Public Relations hit to the other costs of outsourcing seems a reasonable thumb to place on the corporate decision makers’ scales.

For trade as a whole, the trade-offs are more complex but equally comprehensible. Ricardian optimization, allowing the forces of global commerce to place manufacturing in the countries in which it can be carried out most cheaply, ignores a number of problems. For one thing, the Ricardian optimum is not stable. It may be attractive to source in Brazil one year but the following year, when Brazil has elected a leftist who bashes business, the equation may be different. Even simple movements of exchange rates, which can often be of 20% or more in less than a year, can flip the optimum from one country to another.

There are thus “menu changing” costs that should not be ignored. It may be cost-effective at present to move production of a particular item to China, but with Chinese wage costs increasing much more rapidly than those in the U.S., who is to say that the move to China will go on being cost-effective over the life of a new factory. Foxconn, the giant Taiwanese electronics fabricator, has found itself moving production out of China over the last few years, as Chinese costs escalate and other production locations become more attractive.

It may be objected that companies are able to take these decisions on their own, using the criteria of long-term profit maximization as their guide. Unfortunately, in the last two decades of “funny money” and stock options fueled by an ever-rising stock market, long-term profit maximization is not the goal for many corporate managements. Instead those managements, especially in the U.S., want the stock price boost that comes from a short-term fillip to earnings, whatever the long-term cost, because in the long term they will be retired.

Location and trade decisions in any case involve high levels of externalities, costs that are imposed upon the economy as a whole, but not on the company making the relocation decision. Employees who lose their jobs, even if they find another one, suffer disruption from loss of earnings during the inevitable gap, may find their skills eroded or of no interest to their new employer, and may suffer psychological or health problems due to the stress of losing their job. France has shown us that preventing companies from reallocating their workforces is horrendously expensive and itself increases unemployment (because companies are reluctant to hire.) However, it seems appropriate to discourage companies from reallocating productions due to temporary factors, or to impose moderate taxes on them for the cost of their doing so.

Moderate tariffs may thus beneficial, in encouraging domestic production when the cost disadvantage compared to importing is only minor or temporary. If trading partners remain committed to full free trade, it can also provide a country with unearned benefits – the classic case being the United States between 1862 and 1914, when it gained sector after sector of the world’s manufacturing business against Britain, which was subjected to policies of foolish unilateral free trade. That’s why the World Trade Organization is needed – the only international agency that has any useful purpose. Through it, countries can together achieve the benefits of lowering tariffs and trade barriers, without being subjected to destructive and unfair competition from their more protectionist trading partners.

There is an additional benefit from tariffs: they provide revenue. The Ricardian ideal of universal trade assumes that government is small, and that means can be found to finance it that are less damaging than tariffs. In reality, government these days is gigantic, and there appears to be little or no popular will to reduce its size. In such circumstances, moderate tariffs can be beneficial, if they prevent excessive fiscal concentration on income taxes and social security contributions. Equally, export bounties and production subsidies are doubly pernicious, because they both distort trade from the optimum and reduce the government’s revenue.

We may now have reached a position where a tariff is fiscally necessary for the United States. The budget is permanently at least $500 billion in deficit, and likely to be pushed further out of balance by Trump’s programs of infrastructure spending and defense rebuilding. In addition, the U.S. social security and Medicare systems are becoming increasingly in deficit, with trust funds (fictional though they are) likely to run out in a few years. Trump can solve this problem, by imposing a modest tariff on imports, with the proceeds being used to rebuild the social security and Medicare trust funds.

Trump’s proposed 35% tariff is far too high, but a 10% tariff would impose only modest additional costs on U.S. consumers, would go far to closing the chronic U.S. balance of payments deficit, and provided other countries reacted only with modest tariffs of their own, would be only very mildly distorting to world trade. The best precedent is the British Imperial Preference 10% tariff of 1932, which gave Britain a much pleasanter 1930s than the United States, distorted trade far less than the much higher Smoot-Hawley Tariff imposed by the U.S., and would have allowed Britain to rebuild its economy more quickly after the war had it not been disgracefully given away by Maynard Keynes in the 1944 Bretton Woods negotiations.

With the new modest tariff, Trump would have rebalanced the U.S. economy and deterred U.S. companies from unnecessary outsourcing. It would also solve the social security/Medicare deficits problem without either increasing already excessive U.S. income and payroll taxes or cutting benefits – thereby fulfilling a core Trump election promise. It would also render unnecessary many subsidies of the Carrier variety, which themselves reduce government revenues. If foreign production for the U.S. market was cheaper even after the barrier of a 10% U.S. tariff, then the outsourcing should probably go ahead – at least the fisc will gain some extra revenue to offset the job losses. However, if the production being outsourced was destined for third countries, a modest job-retention subsidy would probably remain appropriate.

By these means, a modest tariff, modest but capped job retention subsidies, and extensive Presidential jawboning, Trump would violate the principles of Whig free trade theory, and make academic economists across the entire country from Harvard to Stanford denounce his policies. He would nevertheless benefit U.S. workers, the U.S. fiscal balance and the U.S. economy in general. This column is not Cobdenite, it is LIverpudlian.

-0-
(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)