Saturday, July 10, 2010

Problems with Mankiw's List of Ten "Principles" of Economics

Greg Mankiw is a Harvard professor of economics who writes a very popular introductory economics textbook called Principles of Economics. It is no doubt a very popular introductory economics textbook that is well written, although I certainly would prefer Paul Krugman's Economics. Also, Mankiw has a decent although infrequently updated blog here. Anyway, appropriately enough given the name of his textbook, Mankiw has devised a list of ten economics principles. There is only one problem I have with the list; many of the principles that he lists are not principles, but instead are either definitions, heuristics, or mere possibilities and also involve normative judgments. I therefore would assert that the word "principles" actually constitutes a sort of false advertising.

First, lets start with the definition of principles I am using. Consider this definition of principle from Merriam-Webster (available here):
1 a : a comprehensive and fundamental law, doctrine, or assumption b (1) : a rule or code of conduct (2) : habitual devotion to right principles  c : the laws or facts of nature underlying the working of an artificial device
Now, when I look at the first definition, the two adjectives that strike me in the definition of principle are comprehensive and fundamental. If something is not comprehensive but is instead merely a tendency or a rule of thumb, it should be called a heuristic rather than a principle. A principle should either be comprehensive or at least have a very strong tendency towards being comprehensive. 

Thus, if someone asked me whether a heuristic can rightly be called a principle, I would be inclined to say no. A real principle should have a much stronger tendency towards being comprehensive (i.e. like the law of gravity) than a mere heuristic. The word heuristic seems to be in tension against the attribute of being comprehensive. After all, as soon as you label something a heuristic, you have already conceded a more humble position for the thing than a principle. You have conceded that it is not comprehensive.

This makes me wonder about the whole decision to produce ten “principles” of economics. Why not just call them “principles, heuristics, possibilities, and definitions” or something more accurate? Let's have a look. Are the "principles" of economics that Mankiw identifies in his introductory textbook really principles or is Mankiw engaged in a sort of false advertising?

Here is the list of "principles" with commentary:
(1) People Face Tradeoffs. To get one thing, you have to give up something else. Making decisions requires trading off one goal against another.
This is the closest item in the list that actually comes closest to being a principle. Maybe. People usually face tradeoffs, although one can imagine situations where there is only one decision that one could make with any desirable attributes whatsoever. On the other hand, the further statement that “making decisions requires trading off one goal against another” is simply false when interpreted as a statement about the actual prerequisites to making decisions. Often people make decisions without really considering alternatives much less calculating the costs and benefits of the alternatives even tentatively much less comprehensively. One could say that taking action (not making decisions) often requires trading off one goal against another, whether or not one is consciously aware of that fact (and assuming that the person in question has “goals” other than satisfying a momentary inclination).
(2) The Cost of Something is What You Give Up to Get It. Decision-makers have to consider both the obvious and implicit costs of their actions. 
This isn’t a principle. This is merely a definition. If I say that the cost of something is the amount of damage you impose upon the environment to get it or the cost of something is what everyone gives up for you to get it (this would factor in both externalities and internalities) these would be equally valid definitions in different contexts.

Second, the statement that “Decision-makers have to consider both the obvious and implicit costs of their actions” is clearly a normative statement disguised as a statement of fact. What if decision-makers do not consider such costs? It is possible for them to not consider such costs. What is clearly meant is “Decision-makers should consider both the obvious and implicit costs of their actions.” Even then, we can quibble, because information about costs in the real world often are far from free.
(3) Rational People Think at the Margin. A rational decision-maker takes action if and only if the marginal benefit of the action exceeds the marginal cost.
Once again, this isn’t a principle. It is a (perhaps non-comprehensive) definition of “rational” that I would venture to say that majority of people in the real world do not meet. Once again, I could reasonably come up with a different definition of rational. I would also say that there is a normative judgment lurking here. Since the Enlightenment, a major theme has been that we should aspire to be “rational” while the legal profession pushes us to be “reasonable.” Personally, if I met someone who was actually “rational” according to this definition, I do not think I would associate with them very closely. Attributes like loyalty often require that one commit to another person without constant calculation of costs and benefits. In the end, if one is wise in their selection of close friends, things tend to work out fairly well (though not always — few things are more disappointing than misplaced loyalty) without bringing in calculations of marginal costs and benefits into the picture, which would have a tendency to ruin the friendship.
(4) People Respond to Incentives. Behavior changes when costs or benefits change.
Except when they don’t. This is a heuristic more than a principle. I would say that if you really want to understand how someone is going to respond, you need to put yourself in their shoes and try to understand things from their perspective. Admittedly, this is much more difficult than merely assessing the incentives that they face, as measured in a conventional manner.
(5) Trade Can Make Everyone Better Off. Trade allows each person to specialize in the activities he or she does best. By trading with others, people can buy a greater variety of goods or services.
This is clearly true, at least if interpreted broadly (to be completely accurate, we might say that "trade sometimes can make everyone better off"). Overall, this is humbly phrased as a possibility, not even a heuristic. It is not even an argument that trade usually makes everyone better off, only that there is a possibility that it might. Is the principle here that such a possibility usually exists? In what context? This is extremely vague.
(6) Markets Are Usually a Good Way to Organize Economic Activity. Households and firms that interact in market economies act as if they are guided by an “invisible hand” that leads the market to allocate resources efficiently. The opposite of this is economic activity that is organized by a central planner within the government.
This is self-consciously phrased as a heuristic (not a principle) and it is clearly normative. Are we to believe that the “invisible hand” leads the market to allocate resources efficiently? That depends on our definition of efficiency. Efficiency is not an objective concept. It always depends on a goal. If X and Y are goals in conflict, relatively efficiency in reaching X may imply inefficiency in reaching Y. (This is what economists call trade-offs). So, to say that markets allocate resources efficiently, there is an unstated assumption about a shared underlying goal by which efficiency is measured and also a subjective judgment that a certain level of efficiency is “good enough” to be called “efficient.”

Perhaps Professor Mankiw should actually read the “Wealth of Nations.” The “invisible hand” in that book is clearly referenced in an explicitly religious context. But, shouldn’t “rational” humans evaluate the efficiency of outcomes based on facts and not faith? It is interesting that people often talk about having “faith” in the market; I think references by economists to the religious metaphor of the “invisible hand” encourage such attitudes. And these are the same people who are always talking about what “rational” people do! I would think a rational person would eschew “faith” in the “invisible hand” for an actual inquiry into the facts.

Finally, to say that markets and central planning are opposites fails to consider public-private partnerships where markets and planning are mixed together. Also, that the efficient operation of markets are supported by government and without that foundation would crumble.
(7) Governments Can Sometimes Improve Market Outcomes. When a market fails to allocate resources efficiently, the government can change the outcome through public policy. Examples are regulations against monopolies and pollution.
Again, this is a heuristic and not a principle and is clearly normative (who decides that an intervention is an improvement?). Again, talking about failing to allocate resources efficiently assumes shared goals by which outcomes are measured. Again, there is a subjective conception of a point by which we judge there to be a transition from inefficient to efficient on whatever measure of efficiency is agreed upon. 
(8) A Country’s Standard of Living Depends on Its Ability to Produce Goods and Services. Countries whose workers produce a large quantity of goods and services per unit of time enjoy a high standard of living. Similarly, as a nation’s productivity grows, so does its average income.
Again, is a definition (of what standard of living means) rather than a principle. Again, alternative definitions that are just as valid come to mind. It would be perfectly reasonable to think that a country’s standard of living should be measured not only based on quantity of goods and services, but also by the quantity and quality of leisure time (which may or may not involve market transactions) and also the quality of one’s family life, friendships, and community life. Should standard of living be measured solely by GDP?
(9) Prices Rise When the Government Prints Too Much Money. When a government creates large quantities of the nation’s money, the value of the money falls. As a result, prices increase, requiring more of the same money to buy goods and services.
Obviously, this item requires the assumption ceretis parabis but seems less objectionable than most items on the list. Notice that there are already normative judgments embedded into this item (the amount of money printed is “too much”). But many respectable economists think that a small non-negative amount of inflation is desirable; but the larger point that prices rise when the value of money decreases due to an increase in quantity seems unobjectionable. This would be a reasonable “principle,” although we probably should leave the normative judgment (“too much money”) out of it.
(10) Society Faces a Short-Run Tradeoff Between Inflation and Unemployment. Reducing inflation often causes a temporary rise in unemployment. This tradeoff is crucial for understanding the short-run effects of changes in taxes, government spending and monetary policy.
This is again stated as a heuristic (“often causes”) rather than a principle. The heuristic nature of this statement aside, this is probably one of the best items on the list, as it is free from normative judgments.
Overall, there are a few things that might be argued to be genuine principles in this list, the overall list seems to be falsely labeled. What we have here instead of a list of principles is instead a list of definitions, heuristics, normative judgments, and (maybe) a few principles. (Can a heuristic be a principle? Maybe the existence of the heuristic or a possibility can be a principle, but the object of the heuristic or possibility would not be.)