Posts tagged Welfare Economics
Today's Quote of the Day...

…is from Don Boudreaux and and Burt Folsom’s Fall 1999 Antitrust Bulletin article Microsoft and Standard Oil: Radical Lessons for Antitrust Reform:

It follows that the best available evidence for whether or not a firm enjoys monopoly power is the firm’s own record at satisfying consumer demands: Do real prices in markets in which the firm offers products fall? Does output in these markets expand? Are innovations in these markets regular? If so, the firm is likely not a monopolist. Like Standard Oil, Microsoft does not behave as though it possesses monopoly power. Therefore, we argue that it, in fact, does not possess monopoly power.

JMM: Economists and antitrust lawyers have developed all kinds of statistics and metrics to try and gauge, from an objective point of view, whether or not a firm is in a monopolist position. This is also true in other realms of economic regulation: trying to identify public goods, externalities, perfect compensation, etc etc. In the end, though, our best tool is to observe that which people do. How people behave tells us a lot about the econoomic conditions we are dealing with.

Against Economic Casuistry

In any science, there is a tendency to lay down rules for conduct following some theory or discovery. Economics is no different. However, I will argue here that, in economics, this tendency is not only wrongheaded, but doomed to failure. As economists, we should resist economic casuistry.

Casuistry is “endeavour[ing] to lay down exact and precise rules for the direction of every circumstance of our behavior,” (Adam Smith, Theory of Moral Sentiments, Pg. 329). Smith is writing in the context of moral philosophy and natural jurisprudence, referring to the moralists who attempt to have a rule for every aspect of human interaction, but his definition and description work for our purposes here.

Economic textbooks, especially at the principles level, has a tendency towards casuistry. Principles books tend to lay down certain rules for economics, especially in the realm of policy. Rules like “always do cost-benefit analysis,” or “if the market has failed in a certain way, government can intervene by doing X,” or “public goods should be provided by government,” etc. Policy conversation in the real world revolve around such rules: “we’ve identified problem X, and so recommend policy Y.” For example, Trump is using national security to justify tariffs. Dani Rodrik identifies all sorts of problems and recommends policy changes in his book Straight Talk on Trade.

But the issue Smith highlights in his discussion of moralistic casuistry, namely that identifying and knowing every rule for every situation is damn near impossible, applies to economics as well. Take, for example, the rule that if a market failure occurs, there is some action the government can take to address the failure. As I have written elsewhere, properly identifying market failures is extremely difficult. A “market failure” requires a comparison to a perfectly competitive market, a comparison which may not be valid with the existence of transaction costs. Furthermore, since benefits (and thus also costs) are subjective, identifying transaction costs themselves is a difficult, if not impossible, task.

John Nye also writes on the difficulties when trying to set polices for dealing with market failures. He discusses that there are already many small Coasian bargains going on around externalities and that policy prescriptions tend to fail to take these into account.

What all this adds up to is one simple fact: identifying market failures is not as straightforward as textbook models represent. As Ronald Coase says, paraphrasing Frank Knight, these discussions ultimately come down to a discussion of morals and aesthetics.

We can do this for all supposed market failures. Identifying public goods, for example, is extremely difficult as the definition relies heavily on the scope of the problem you are looking at. National defense, the quintessential public good, is a public good at a small scale (the Army protecting Washington DC from attack also protects Maryland), but at a large scale, it becomes a private good (the Army can choose whether or not to defend Mexico). So, is national defense a public or private good? Hard to tell. Thus, hard to set rules.

Classifying economic situations and outcomes is not as simple as classifying an animal in biology. A lot of it will depend not only on the actors involved, but the judgement of the spectator, the analyst. As such, it is very loose and vague. Precise rules, like those laid forth in economics textbooks, will tend to fail.

None of this is to say one should just through up their hands and do nothing. There can, indeed need, to be rules. But they should also be loose, more negative than positive. Rules such as private property, where people can do as they wish so long as they do not harm each other. Courts that can interpret issues when conflicts inevitably arise. In short, “thou shalt not” rules rather than “thou shalt.” As economists, we must resist the natural urge to lay down 10,000 commandments, rather going for rules more like the 10 Commandments.

Operationalization of Theory is Never Straightforward

Over at EconLog, Pierre Lemieux points us to a recent op-ed by Trump economic advisor Peter Navarro. Pierre writes:

The keystone of his claims in this op-ed is the distinction between “pure free trader” and “fair, reciprocal and balanced trader.” The latter concept is at best underdetermined and at worst absurd. There have been as many definitions of “fair” as there have been political theorists and moral philosophers. Is it “fair trade” that American producers have a big advantage as they master the language of international trade better than their German or Vietnamese competitors? Isn’t that the sort of “trade barrier” that Navarro said should be compensated by a tariff in order for trade to be “reciprocal”? Reciprocity is usually a mere excuse for protectionism.

This is a point I routinely make in my classes when I lecture on supposed exceptions to free trade and operationalizing these exceptions.  In theory, it is easy to design a policy where some “unfair advantage” is corrected.  In reality, identifying these barriers is difficult.

My favorite example is rule of law.  The US, compared to many countries, is very non-corrupt.  We have a good judicial system that is relatively unbiased and even-handed.  Contracts are reasonably enforced.  Bribes aren’t really required to do business.  All this reduces the cost of doing business in the US compared to other countries, which is why many firms do business here.

Since the judicial system is run by the government, one can argue it counts as a “subsidy” to reduce costs for businesses.  Thus, an analyst could use rule of law in the US to justify tariffs against US products.

Is that also what Navarro means when he demands reciprocity?  What about our relatively educated workforce (also subsidized, BTW)?  Or our relatively good infrastructure?  All of these would justify, in theory, subsidies on American products by foreign nations given the loose, vague, and indeterminate language of Navarro.  The theory gives no guidance about what is counted and what is not, which means it is left up to the analyst.  Thus, talking about “fair trade” is not as precise as Navarro, or theory, makes it seem.

Once we begin moving away from theory and into interventionist policy making in the real world, the operationalization of the theory gets very loose very quickly. How things are defined is not straightforward and precise. Calculating “optimal” tariffs will depend crucially on the assumptions and definitions of the analyst. What Peter Navarro considers “reciprocity” may not be considered so by Donald Trump, or by you and I. Indeed, for certain definitions of reciprocity (each as theoretically legitimate as the one Navarro uses), one can argue that any Chinese trade barriers on US goods are “fair and balanced” and the actions of the Trump Administration are unfair trade!

At Cafe Hayek, Don Boudreaux highlights a similar theme put forth by our colleague Bryan Caplan and his co-author Zach Weinersmith:

The right question to ask is never “Will it be perfect?” But “Will it be better than the alternative?”

The problem with perfection is not only that men are not angels and fallible. Were it only that, then it would simply be a matter of creating an algorithm for a machine to optimize and then just internalize any externalities. The problem with perfection includes actually defining perfection. A blackboard model of equilibrium is only “perfect” in the eyes of the analyst. Assumptions we have to make to get there are enormous and context-dependent (for more on this point, see Hayek’s 1937 paper “Economics and Knowledge.” Also valuable is James Buchanan’s discussion in Chapter 1 of LSE Essays on Cost). Thus, even determining what policy should be will depend heavily on the assumptions of the analyst (note this lesson holds over from my blog post “Does Economics Imply Liberalism?”. This is all the more reason why it is important to focus on obtainable alternatives as opposed to idealized outcomes.

None of this implies radical anarchism or that policymakers should be paralyzed with fear (although I am sure people will use the arguments here to justify those stances). Rather, what this is to say is that discussions of ideal economic policy are not straightforward, that there is, as Ronald Coase wrote paraphrasing Frank Knight: “[P]roblems of welfare economics must ultimately dissolve into a study of aesthetics and morals.”