Friday, August 22, 2008

My Tome on Economics, Markets, and Society

These comments are with reference to the two articles I included in my post entitled "On Economics, Markets, and Society" several days ago.

The main thesis of the City Journal article, "Economics Does Not Lie," is that the field of economics has reached the status, in terms of objectivity and mathematical rigor, of a science, and that beyond this, economics has shown us that capitalism is the best economic system. While my own personal views are not far from this, there are some essential nuances and differences in reasoning that cause me to make the conclusions I make, and that could just as easily cause others to come to different conclusions.

Proposition 1: Economics is a science.

In particular, economics has now reached the level of scientific rationality. With this point I am in fairly strong agreement. I will call a discipline scientific if it follows the scientific method.


For the sake of comparison, I will place economists into three camps: string theorists, experimental scientists, and engineers. The "string theorists" group is for those economists who focus on economic theory, whether it be macroeconomic theory, microeconomic theory, or any other branch of economics where the work is almost entirely abstract and mathematical with little connection to actual data. The work of these economists forms the foundation for how economists view the world and gives rise to the models that "experimental scientists" use to compare to real-world phenomena. I call these economists "string theorists" because their work is highly theoretical (but do not confuse this with impractical or irrelevant), mathematical, and unverifiable without the work of the "experimental scientists." Just as string theorists seek to create a mathematical structure that unifies the various natural laws, theoretical economists seek to create a mathematical framework that explains the economic phenomena that we observe. While there are many philosophical discussions that I could envision arising from what I have said here, I will have to postpone that until later (and more importantly, until somebody actually asks a specific question that could steer the discussion).

The economists in the "experimental science" group do work that is most analogous to the work that most scientists do. Economists make an observation about some economic event, they collect data, make a hypothesis (usually in the form of constructing a model or using an existing one), and they test the hypothesis through the use of econometrics, statistics, and other mathematical forms of reasoning. What differentiates economics from most of the sciences, however, is that economists cannot control, let alone reproduce, the experiment (Actually, this is not entirely true. Nothing in principle prevents economists from reproducing an episode of hyperinflation or a Great Depression, but limits of practicality and ethics prevent them from doing so.) However, this does not mean that an "experimental scientist's" work is shielded from peer review, or that other economists cannot reproduce their results. Rather, it means that in some instances, the raw data is not reproducible, but every other step in the analysis can be reproduced. This is also true in some instances in other sciences, such as in the irreproducibility of the big bang or of macro-evolution.

Proposition 2: Most economists are free-market purists.

Before I address the 10 propositions that economics supposedly definitively teaches us, I want to address more generally the notion that economists are, as I said above, free-market purists. The short answer is that that is simply not true. Daniel Klein and Charlotta Stern conducted a survey of members of the American Economic Association (who tend to be academic economists) to determine their views on major public policy issues. Although the economists were far more accepting of free-market principles than, say, anthropology professors, and although the economists did agree on certain issues, such as the virtue of international trade, they tended not to fall onto any one side for most other issues. There was no unanimity regarding whether or not the government should increase redistribution, or whether the minimum wage should be increased, etc. Why is there so little agreement? Greg Mankiw gives his take here. I believe there are two essential reasons for why economists do not agree.

Reason 1: Different views of reality

Most of what Greg Mankiw mentioned in his post falls under this umbrella. There is no universal consensus on the size of tax inefficiencies, or on the size of externalities from pollution. As a result, economists can extrapolate from whatever papers they have read, filter that through their worldview, and come to a personal belief about how the economy functions. Of course, an intellectually honest economist can only do this so much.

Reason 2: Different values

To me, this is the biggest factor that causes economists to disagree with one another, and it is likely highly correlated with economists' different views of reality. Consider the example of international trade. It is accepted almost universally among economists that international trade increases the total net welfare of any country that participates in it. However, it is also generally accepted that within each country there will be winners and losers. It just happens to be the case that the sum of the monetary losses will be exceeded by the sum of the monetary gains. This means that in principle (though not so much in reality, at least right now) we could collect all the gains from trade and redistribute them in such a way as to make everybody better off. Seeing as this does not happen very much, we are left with winners and losers. Free-market economists, because of their values, may believe that the government should not have the right to prevent willing participants from engaging in mutually beneficial trade, and they may believe that the fact that trade yields a net gain is more important than the actual distribution of gains. Trade skeptics, on the other hand, may believe that the government should have the right to interfere with international transactions if doing so impacts the distribution of income in a way that they find desirable. This is simply a question of values.

Now, to briefly analyze and critique the ten propositions:

1. The market economy is the most efficient of all economic systems.

Actual economic theory does not support any such universal proclamation. It is true that in an economy with no information asymmetries, no externalities, no barriers to entry, and no industry concentration, the market allocation will be efficient (which means that there is no way to make anybody better off without making somebody else worse off). This is in fact the first fundamental theorem of welfare. This sounds like thin gruel compared to the statement that the market economy is the most efficient of all economic systems.

On the other hand, the Myerson-Satterthwaite theorem states that in a market with voluntary participation and asymmetric information about how much participants in the market value the good, there is no efficient mechanism of exchanging the goods. This means that even the market is inefficient here. Basically, there will be instances where mutually beneficial trade could occur between buyers but does not, precisely because of the information asymmetries.

Lastly, one issue that prevents the unfettered marketplace from being efficient is the issue of incomplete markets. In particular, for some goods (and bads) there are no markets in which they can be traded. For example, this can mean that individuals do not have sufficient ability to insure against various types of economic risk. In addition, this means that externalities arise, such as the externality arising from firms being able to pollute without taking into account the cost imposed on the rest of society.

Ironically, the article gives the example of cap-and-trade to support its statement that markets are efficient:

"Market mechanisms are so efficient that they can manage threats to long-term development, such as the exhaustion of natural resources, far better than states can. If global warming does become a real problem, for example, price mechanisms or a carbon tax would easily encourage a more efficient use of energy."

On the one hand, markets are so efficient that they can manage threats to long-term development. On the other hand, the government has to interfere with the economy in order to create a market to deal with the externality caused by pollution.

Modified principle: In most instances, markets, where they exist, tend to allocate resources in the most efficient manner. However, government can improve economic efficiency by creating markets where they do not exist, and by regulating markets to deal with the inefficiencies caused by information asymmetries.

2. Free trade helps economic development.

Mr. Sorman claims the following:

"In fact, economists have long understood the law of comparative advantage: whenever differences in the cost of producing goods exist between two countries, both will benefit from free trade, a mechanism that allocates their resources most effectively."

Mr. Sorman is correct that economists have long understood the law of comparative advantage, except that what he wrote down is not the law of comparative advantage. Instead,

"The principle of comparative advantage shows that even if a country has no absolute advantage in any product (ie. it is not the most efficient producer for any good), the disadvantaged country can still benefit from specializing in and exporting the product(s) for which it has the lowest opportunity cost of production."

In short, a country has a comparative advantage in a particular product if the trade-off within that country between producing that product and producing another product is better than that same trade-off within the other country. The example I like to think of is that of a lawyer and his secretary. Even if the lawyer is better than his secretary both at arguing cases and at typing documents on the computer (ie he has an absolute advantage in both areas), it would benefit the lawyer to hire the secretary to do his typing for him so that he can focus on the area where he has the comparative advantage, namely arguing cases. Since his comparative advantage is arguing cases, it must be that the secretary's comparative advantage is in typing documents on the computer, even though the lawyer has an absolute advantage in both areas. The key is that his trade-off between arguing cases and typing documents is much worse than hers-- he has to give up much valuable time arguing cases in order to type the documents, whereas since she cannot argue cases anyway, she does not have to give up time spent arguing cases.

My point here is not to quibble over economics definitions, but rather to distinguish between the actual source of the benefits of trade and that touted by pundits. The United States does not benefit from trade primarily because it can import cheaper products. Instead, the United States benefits from international trade because trade causes us to specialize in the production of goods and services that we are best at producing, while allowing us to import those products that other countries can better specialize in.

Aside from mentioning comparative advantage, the article also states that "Free trade not only generates the greatest possible growth; it tends to distribute it widely, both within nations and among them." This is not entirely true. Within a country, people working in the sectors in which the country has a comparative advantage will say their pay increase, whereas people in the other sectors will see their pay decrease (or they will lose their jobs). Since the United States has a high-tech economy, trade causes the pay of those with more skills (who tend already to be more well-off) to increase, while decreasing the pay of others. Therefore, in the United States, trade likely causes a (modest) increase in inequality. In less-advanced economies, the effect of trade is the exact opposite. Where Mr. Sorman is correct is in stating that free trade tends to distribute growth widely among nations. We have seen this as South Korea, Japan, Western Europe, and many other nations have seen their incomes converge to that of the United States.

Note: While most of what I have said is widely accepted among economists, I also recognize that there are more sophisticated models that may give rise to exceptions or even disagreements. Seeing as I am not an expert on these issues, I give you what I know and what economists generally agree with.

3. Good institutions help development.

I generally agree with what the article says on this topic, at least until it veers into the area of behavioral economics. I do not have much to add here, except to say that the economics of institutions is becoming an even more active field of research, and I look forward to learning more about it.

4. The best measure of a good economy is its growth.

The main problem I have with this proposition is not the content of the paragraphs that follow it, but the actual text of the proposition itself. There are many objective measures that we can use to assess the health of the economy. GDP is one of the most important measures, since it tells us the total value of all goods and services produced within the country. If GDP decreases, then that means we are producing less, which means there is less to go around. However, there is nothing in economics that dictates that growth is a better measure of a good economy than low inflation or a broad distribution of wealth. Nevertheless, I do agree as a matter of personal opinion that we should focus on having strong economic growth as a high priority, since without growth many economic debates would degenerate into zero-sum games where we fight over an unchanging pie.

5. Creative destruction is the engine of economic growth.

I agree.

6. Monetary stability, too, is necessary for growth; inflation is always harmful.

Economists are in wide agreement that inflation is harmful for the economy, and that one of the primary responsibilities of a central bank is to keep inflation low. Furthermore, high inflation typically means unpredictable inflation, whereas low inflation is usually less variable. High inflation rates discourage savings and investment because it makes the rate of return less certain and more likely to be lower than it would be if people expected the price level to be stable. In addition, inflation, by eroding the value of the dollar, makes the real value of nominal debt decrease. While this of course is of great help to debtors, high inflation expectations will make lenders less willing to lend, which hurts borrowers. Inflation also exerts a pernicious influence through the tax code, known as bracket creep. Since the income thresholds in the tax code are based on nominal dollars, inflation over time causes people to appear as if they are wealthier than they are, and therefore pushes them into higher tax brackets.

In a more abstract but equally valid sense, inflation is harmful because it drives a wedge between the social cost of money and the economic cost, whereas in economics we want social costs to equal economic costs. The social cost of money is the cost to actually produce money, which is effectively zero. The economic cost, however, is not zero, as inflation causes nominal interest rates to rise.

7. Unemployment among unskilled workers is largely determined by how much labor costs.

"So regulating the labor market (with a minimum wage, for example) adds to labor costs, economists acknowledge, and increases unemployment. No solution to excessive unemployment is conceivable without reducing such regulations."

Guy Sorman tries to generate a consensus that does not actually exist among economists. Even if labor markets are perfectly competitive, what Mr. Sorman says is not true. This is because if labor markets are perfectly competitive, wages will adjust so as to cause zero unemployment (or at least zero frictional unemployment). If firms have market power (or in the extreme, if a firm is a monopsony-- the analogy to a monopoly, except the firm being a single buyer instead of a single seller), then labor costs are not the only factor in determining unemployment. In such a setting, the minimum wage can actually increase efficiency in the labor market by bringing the market closer to perfect competition.

Even in labor markets that are not characterized by firms with a lot of market power, the minimum wage can have a more complicated impact on labor market outcomes. Some economic research, such as Daron Acemoglu's paper Good Jobs and Bad Jobs, suggests that minimum wages can induce firms to create higher paying jobs and invest more in training. This result is not universal, however, and there are many studies that point to the negative effects of minimum wages. Since minimum wages do increase the cost of labor, it is entirely reasonable to expect that firms will purchase less of that labor.

Minimum wages do not represent the only form of labor market regulations, however. I have no intention of discussing every bit of labor legislation, but I will say that Mr. Sorman hits the nail on the head when he talks about some of the labor market woes in Europe. Whenever the government makes it nearly impossible to fire a worker, such as in France, employers have a strong incentive not to hire. After all, if the economy worsens, or if the worker underperforms, the employer is already locked into the employment contract. This is one reason why in France unemployment is much higher than in the United States, and also why there is much more temporary employment there.

8. While the welfare state is necessary in some form, it isn’t always effective.

This statement, mostly by virtue of its vagueness, is true. Mr. Sorman's first statement says it all: "Economists recognize that government assistance always produces incentives that may affect, for good or ill, recipients’ behavior and well-being." Whether we are talking about unemployment insurance, welfare benefits, Medicare, or Social Security, these programs affect the marginal costs and benefits of saving, investing, going to school to get a more high-powered occupation, finding a job and working more. The reason for the success (at least partial success) of welfare reform in the 1990's was that it reduced the cost to earning more income and made it more costly to stay on welfare. Before welfare reform, it was actually possible to be in a situation where you were living on welfare, food stamps, and Medicaid, and if you found a job, you would lose enough eligibility so as to actually be worse off. Unsurprisingly, these kind of incentives lead to more people locked in dependency with little ability or motivation to get out of it.

None of this is to say that we should dismantle the welfare state. I will have to postpone fully delving into this issue until later, but needless to say, the key is to design a safety net that helps people get on their feet, and provides assistance and incentives to invest in themselves and move up the economic ladder.

9. The creation of complex financial markets has brought about economic progress.

I have already discussed this issue at more length in my posts about Fannie Mae, Freddie Mac, and government bail-outs. Nevertheless, in summary, the US benefits greatly from having sophisticated and fluid financial markets. At the same time, these markets, due to the extreme importance of transparency and information, and due to their strong interdependence with the rest of the US economy, are more fragile and in need of effective, efficient regulation.

10. Competition is usually desirable.

Without rehashing many of the points I have already made, I'll go ahead and simply agree.

Conclusion: Economics has evolved and developed substantially in the past few decades. Its methods have become more rigorous, mathematical, and objective, even while its conclusions in many areas remain debated and contested. None of this is to say that economists are completely devoid of consensus. On many of the big, philosophical issues of the day, economists will find that they have more in common with each other than not. However, economics is not at a point where it can dictate the best public policy. In fact, the field of economics itself will likely never be at such a point, since public policy choices rely on economic facts as well as values. What we can look forward to and aspire to is having economics more accurately and decisively weigh in on the workings of the economy and on the effects of different policy choices.

No comments: