Does it “Take a Model to Beat a Model”?

Imagine being a chemist in the Middle Ages. Your colleagues are all working on fanciful tasks like trying to develop a philosopher’s stone to grant immortality or turning lead into gold. You continually point out to them that all of their attempts have pretty much failed completely. In fact, based on your own research, you have strong reason to believe that the path they are taking is going completely the wrong direction. They aren’t just failing because they haven’t found the right formula to transform metal into gold, but because the task they have set out to do cannot be done. You urge them to abandon their efforts and focus on other areas, but they don’t seem to listen. Instead, their response: “well sure we haven’t been able to turn lead into gold yet, but can you do any better? I don’t see any gold in your hands either.”

You can probably already see where I’m going with this metaphor. Replace chemist with economist and turning lead into gold with DSGE macroeconomics and you’ll have a good sense of what criticizing macro feels like. I don’t think it’s an exaggeration to say that every critique of macro is invariably going to be met by some form of the same counterargument. Of course the model isn’t perfect, but we’re doing our best. We’re continually adding the features to the benchmark model that you claim we are missing. Heterogeneity, financial markets, even behavioral assumptions. They’re coming. And if you’re so smart, come up with something better. It takes a model to beat a model.

I won’t deny there is some truth to this argument. Just because a model is unrealistic, just because it’s missing some feature of reality, doesn’t mean it isn’t useful. It doesn’t mean it isn’t a reasonable first step on the path to something better. The financial crisis didn’t prove that the methods of macroeconomics are wrong. Claims that nobody in the macro profession is asking interesting questions or trying to implement interesting ideas are demonstrably false. We certainly don’t want criticisms of macro to lead to less study of the topic. The questions are too important. The potential gains from solving problems like business cycles or economic growth too great. And if we don’t have anything better, why not keep pushing forward? Why not take the DSGE apparatus as far as we possibly can?

But what I, and many others, have tried to show, is that the methods of macroeconomics are severely constrained by assumptions with questionable theoretical or empirical backing. The foundation that modern macroeconomics is building on is too shaky to support the kinds of improvements that we hope it will eventually make. Now, you can certainly argue that I am wrong and that DSGE models are perfectly capable of answering the questions we ask. That’s quite possible. But if I am not wrong about the flaws in the method, then we shouldn’t need a new model to think about giving up on the current one. When somebody points out a flaw in the foundation of a building, the proper response is not to keep building until they come up with a better solution. It’s to knock the building down and focus all effort on finding that solution.

I can confidently say that if a better alternative to DSGE exists, I will not be the one to develop it. I am nowhere near smart or creative enough to do that. I don’t think any one person is. What I have been trying to do is convince others that it’s worth devoting a little bit more of our research efforts to exploring other methods and to challenge the fundamental assumptions that have held a near monopoly on macroeconomic research for the last 40 years. The more people that focus on finding a model to beat the current model, the better chance we have to actually find one. As economists, we should at the very least be open to the idea that competition is good.

Where do we start? I think agent based simulation models offer one potential path. The key benefit of moving toward a simulation model over a mathematical one is that concerns about tractability are much less pressing. Many of the most concerning assumptions of standard macro models are made because without them the model becomes unsolvable. With an agent based model, it is much easier to incorporate features like heterogeneity and diverse behavioral assumptions and just let the simulation sort out what happens. Equilibrium in an agent based model is not an assumption, but an emergent result. The downside is that an ABM cannot produce nice closed form analytical solutions. But in a world as complex as ours I think restricting ourselves to only being able to answer questions that allow for a closed form solution is a pretty bad idea – it’s looking for your keys under the streetlight because that’s where the light is.

Maybe even more important than developing specific models to challenge the DSGE benchmark is to try to introduce a little more humility into the modeling process. As I’ve discussed before, there isn’t much of a reason to put any stock in quantitative predictions of models that we know bear little resemblance to reality. Estimating the effects of policy to a decimal point is just not something we are capable of doing right now. Let’s stop pretending we can.

Keynesian Economics Part 2 Investment and Output

In my last post on Keynesian economics I outlined a simple example that I think captures the core of Keynes’s economics. It will help to understand this post if you read that one first.

Keynes’s key insight was that an attempt to save by an individual does not always lead to an increase in aggregate saving. I showed how using a simple example in the last post, but we can also generalize the problem. Imagine that each consumer consumes only a fraction of their income (it does not have to be the same across individuals, but I will assume it is for simplicity). Then total consumption spending is given by

    \[C = bY\]

Where C is consumption, Y is income (and total output), and b is the fraction of income spent on consumption (the marginal propensity to consume).

Let’s say that the only spending in the economy is consumption spending. You might already be able to see that we have a problem. Total spending must always equal total income in the economy so that

    \[Y = C = bY\]

Which can only be true if Y=0, so the economy breaks down. Perhaps this scenario is easiest to see if we imagine the case where there is one worker and one firm. The worker works for the firm and gets paid Y. He then decides to buy bY of the output he just produced. The firm realizes he made too much stuff, so he cuts back on production. But this means he reduces his demand for the worker’s labor and cuts his hours. But now the worker makes less so he spends even less and the process continues until no production is carried out at all. The only way we could sustain production through consumption alone would be if nobody wanted to save at all.

If consumption spending isn’t enough to keep firm production positive, we need some demand from another source. One source could be other firms in the form of investment. If we fix income at Y and assume again households only want to consume bY, it is still possible that firms can make up the additional spending by investing (1-b)Y. Keynes argued that there is no reason to expect that investment would always exactly fill gap. If desired investment by firms is less than the difference between consumption and income, they won’t be able to sell all of their product and will cut back on production. We can see that if we write out our equation again, now with investment, it becomes

    \[Y = C + I = bY + I\]

And solving for Y gives

    \[Y = \frac{I}{1-b}\]

So the level of investment determines the level of income. It was through this logic that Keynes concluded that it was the “animal spirits” of firms that determined the state of the economy. It’s possible that the level of investment exactly corresponds to the full employment level of output of an economy, but there is nothing that guarantees that it will.

There are still a few subtleties we need to consider. The first is the role of interest rates. In the classical view of the economy, when people try to save more, they increase the supply of loanable funds, which pushes down interest rates (think of banks having excess money to lend and the only way they can get rid of it is by lowering the interest rate). That lower interest rate then makes previously unprofitable investment projects become profitable and investment rises. If the interest rate falls enough, it’s possible that the increase in investment would be enough to offset the decrease in consumption.

Keynes didn’t deny this possibility. However, he argued (I think correctly), that interest rates are certainly not the only, and likely not even the primary, factor that goes into a firms investment decision. If a firm expects demand to be low due to a recession, there is no interest rate where it will be profitable for them to make that investment. And, as we saw in the last post, by failing to make those investments, firms’ expectations become self-fulfilling and their pessimism is proven correct. Interest rate adjustments alone therefore cannot save us from a Keynesian recession.

Another potential question comes from the assumptions of the Keynesian consumption function. It is obviously unrealistic to assume that each household wants to consume the same constant fraction of their income. People like Milton Friedman have argued that what people really care about when making consumption decisions is their permanent income. If my income falls today, but I expect it to return to its previous level tomorrow, I will borrow in the bad times to keep a constant level of consumption. I think this criticism is valid, but I don’t think it stops Keynes’s story. As long as aggregate consumption is less than total output (which it almost certainly will be), we still need investment to fill the gap. We still rely on expectations of firms to be correct regarding their future demand.

By focusing on the case where investment was exactly enough to move the economy to full employment, Keynes argued that “classical” economists implicitly restricted the economy to a special case. Keynes set out to correct that theory by proposing a “general theory” where investment fluctuated unpredictably and could (and often is) less than the level that would sustain full employment. I think this contribution is extremely valuable and unfortunately often overlooked. Even modern “New Keynesian” models bear little resemblance to the economy Keynes described. Models with money at all are rare and ones that allow the type of monetary disequilibrium in Keynes’s theory are all but nonexistent.

 

What’s Wrong With Modern Macro? Part 14 A Pretense of Knowledge in Macroeconomics

Part 14 in a series of posts on modern macroeconomics. This post concludes my criticisms of the current state of macroeconomic research by tying all of my previous posts to Hayek’s warning about the pretense of knowledge in economics


Russ Roberts, host of the excellent EconTalk podcast, likes to say that you know macroeconomists have a sense of humor because they use decimal points. His implication is that for an economist to believe that they can predict growth or inflation or any other variable down to a tenth of a percentage point is absolutely ridiculous. And yet economists do it all the time. Whether it’s the CBO analysis of a policy change or the counterfactuals of an academic macroeconomics paper, quantitative analysis has become an almost necessary component of economic research. There’s nothing inherently wrong with this philosophy. Making accurate quantitative predictions would be an incredibly useful role for economics.

Except for one little problem. We’re absolutely terrible at it. Prediction in economics is really hard and we are nowhere near the level of understanding that would allow for our models to deliver accurate quantitative results. Everybody in the profession seems to realize this fact, which means nobody believes many of the results their theories produce.

Let me give two examples. The first, which I have already mentioned in previous posts, is Lucas’s result that the cost of business cycles in theoretical models is tiny – around one tenth of one percent of welfare loss compared to an economy with no fluctuations. Another is the well known puzzle in international economics that trade models have great difficulty producing large gains from trade. A seminal paper by Arkolakis, Costinot, and Rodriguez-Clare evaluates how our understanding of the gains from trade have improved from the theoretical advances of the last 10 years. Their conclusion: “so far, not much.”

How large are the costs of business cycles? Essentially zero. How big are the gains from trade? Too small to matter much. That’s what our theories tell us. Anybody who has lived in this world the last 30 years should be able to immediately take away some useful information from these results: the theories stink. We just lived through the devastation of a large recession. We’ve seen globalization lift millions out of poverty. Nobody believes for a second that business cycles and trade don’t matter. In fact, I’m not sure anybody takes the quantitative results of any macro paper seriously. For some reason we keep writing them anyway.

In the speech that provided the inspiration for the title of this blog, Hayek warned economists of the dangers of the pretense of knowledge. Unlike the physical sciences, where controlled laboratory experiments are possible and therefore quantitative predictions might be feasible, economics is a complex science. He argues,

The difficulties which we encounter in [complex sciences] are not, as one might at first suspect, difficulties about formulating theories for the explanation of the observed events – although they cause also special difficulties about testing proposed explanations and therefore about eliminating bad theories. They are due to the chief problem which arises when we apply our theories to any particular situation in the real world. A theory of essentially complex phenomena must refer to a large number of particular facts; and to derive a prediction from it, or to test it, we have to ascertain all these particular facts. Once we succeeded in this there should be no particular difficulty about deriving testable predictions – with the help of modern computers it should be easy enough to insert these data into the appropriate blanks of the theoretical formulae and to derive a prediction. The real difficulty, to the solution of which science has little to contribute, and which is sometimes indeed insoluble, consists in the ascertainment of the particular facts
Hayek (1989) – The Pretence of Knowledge

In other words, it’s not that developing models to explain economic phenomena is especially challenging, but rather that there is no way to collect sufficient information to apply any theory quantitatively. Preferences, expectations, technology. Any good macroeconomic theory would need to include each of these features, but each is almost impossible to measure.

Instead of admitting ignorance, we make assumptions. Preferences are all the same. Expectations are all rational. Production technologies take only a few inputs and outputs. Fluctuations are driven by a single abstract technology shock. Everyone recognizes that any realistic representation of these features would require knowledge far beyond what is available to any single mind. Hayek saw that this difficulty placed clear restrictions on what economists could do. We can admit that every model needs simplification while also remembering that those simplifications constrain the ability of the model to connect to reality. The default position of modern macroeconomics instead seems to be to pretend the constraints don’t exist.

Many economists would probably agree with many of the points I have made in this series, but it seems that most believe the issues have already been solved. There are a lot of models out there and some of them do attempt to deal directly with some of the problems I have identified. There are models that try to reduce the importance of TFP as a driver of business cycles. There are models that don’t use the HP-Filter, models that have heterogeneous agents, models that introduce financial frictions and other realistic features absent from the baseline models. For any flaw in one model, there is almost certainly another model that attempts to solve it. But in solving that single problem they likely introduce about ten more. Other papers will deal with those problems, but maybe they forget about the original problem. For each problem that arises, we just introduce a new model. And then we take those issues as solved even though they are solved by a set of models that potentially produce conflicting results and with no real way to differentiate which is more useful.

Almost every criticism I have written about in the last 13 posts of this series can be traced back to the same source. Macroeconomists try to do too much. They haven’t heeded Hayek’s plea for humility. Despite incredible simplifying assumptions, they take their models to real data and attempt to make predictions about a world that bears only a superficial resemblance to the model used to represent it. Trying to answer the big questions about macroeconomics with such a limited toolset is like trying to build a skyscraper with only a hammer.

Free Will, Morality, and Libertarianism How can you be held responsible for something that wasn't your fault?

An astute reader sees an apparent contradiction between my last post and my post on free will, asking “how can a libertarian reconcile no real choice with the importance of being free to choose?”

It’s a great question. If free will doesn’t exist, if every single action is pre-determined, can we even have a consistent concept of morality? I made a distinction between actions made freely and those forced by the state, but isn’t that distinction meaningless in a deterministic world? When all actions are at some level outside an individual’s control, is there any difference between the direct coercion of government and more indirect factors influencing the decision (like genetics, education, religion, etc.) that are also completely removed from the realm of free choice? The answers to these questions are far from obvious and I certainly don’t pretend to have a perfect response, but I hope this post will clarify the way I think about the issue. My ideas here are heavily influenced by (who else) Hayek’s discussion of similar topics in chapter 5 of The Constitution of Liberty.

First, it is important to understand my earlier defense of determinism. The key point is that every action can be traced to a chain of previous events. Going back far enough in each person’s chain, some link will be the result of an event that is outside of their direct control. And if I know all of the links in this causal chain, if I know everything that has ever influenced an individual, I can predict with absolute certainty their next move. Their “choice” was determined long before they are required to make it.

If we accept this argument, then at the moment of a decision nothing can be done. There is no way that a given individual would have made any decision different than the one they made. We could replay the same history a million times and get the same result in every single trial.

But here’s the problem. Go back to the example I gave in the last post. A person is trying to decide whether to give money to the poor (let’s call him Bob). Note that I will consider giving to the poor to be a good thing. If for some reason you disagree with that assessment, replace “giving to the poor” with any action you consider moral and the argument should still go through.

Now assume Bob exists in two universes (A and B). In each universe, Bob has had almost the exact same experience. He has the same parents, the same teachers, read all the same books. As a result, in each universe he has developed a system of values which teaches him to care for his fellow human beings. Now introduce one difference between the two universes. Universe A has a government which forces Bob to give to the poor through taxation. In Universe B, Bob is free to do as he pleases. Of course, we know  that Bob is not really free to choose. If he chooses to give to the poor, it is only because he grew up in a society that taught him that that was the right thing to do, and only because of his upbringing that he has any desire to do the right thing at all. Universe B Bob doesn’t choose to give to the poor any more than Universe A Bob does. Both only give due to the influence of others. How can we say one is more moral?

Consider George. George also exists in both universes, but he has had a different experience than Bob. Where Bob was taught to live a life of compassion, George only cares about his own material well being. Help the poor? How does that help George? In Universe A, George still has to give to the poor. The government forces him to give against his wishes. And it makes him angry. He works hard to earn his money, why should he give to those who don’t? He sees Bob gives to the poor as well, but he believes it is only because the government forces him to do so.

In Universe B, George doesn’t give to the poor. His values tell him that you get exactly what you deserve in life and he acts on those values. There is no government to force him to do otherwise. And yet he sees Bob give to the poor anyway. Maybe he just dismisses Bob as too stupid to realize that his money won’t help them, that the poor need to help themselves. But maybe Bob’s actions give him pause. Maybe they form a new link in George’s causal chain. Maybe he questions his decision, and even though he could never have changed his choice at that moment, he might think about the situation differently next time. Maybe his system of values begins to change.

Phrased in this way, we begin to see a real distinction between the decision to give in each universe. It is true that in both universes Bob would have given to the poor. Our reason for calling Bob’s actions moral cannot be that he himself could have made another choice at the moment of his decision – without free will, he really couldn’t have. But we can compare Bob’s choices to those of another individual. If we replace Bob with George in Universe A, the result is the same – both give to the poor, and we can’t judge morality because nobody could have acted differently. In Universe B, however, Bob and George are allowed to act differently in the same situation. Their decision tree has two branches. Neither will choose any branch other than the one already pre-determined by their life experiences, but the existence of the branches matters because somebody else could have.

So now we can give an answer to the original question. If nobody can truly make choices of their own volition, why does choice matter for morality?

Because if a choice is available, even if each individual will always make the same choice, another might have acted differently.

But even if you buy the argument above, a question still remains. We might agree that Bob made a moral decision in the example above even though it wasn’t truly his choice, but does that mean that George is responsible for his actions? Can we blame George for not giving to the poor? After all, it’s not his fault that he didn’t have Bob’s life. Here I defer to Hayek:

Strictly speaking, it is nonsense to say, as is so often said, that “it is not a man’s fault that he is as he is,” for the aim of assigning responsibility is to make him different from what he is or might be. If we say that a person is responsible for the consequences of an action, this is not a statement of fact or an assertion about causation. The statement would, of course, not be justifiable if nothing he “might” have done or omitted could have altered the result. But when we use words like “might” or “could” in this connection, we do not mean that at the moment of his decision something in him acted otherwise than was the necessary effect of causal laws in the given circumstances. Rather, the statement that a person is responsible for what he does aims at making his actions different from what they would be if he did not believe it to be true.

We assign responsibility to a man, not in order to say that as he was he might have acted differently, but in order to make him different.The Constitution of Liberty, p. 137-138

What do we want from our society? Do we want to move from Universe B to Universe A, from a world where people are free to make their own value judgements to one where they are not given any choice, where we claim to know what is best for them? Or do we want to convert people from Georges to Bobs, to convince them to buy into the system, convince them that the ideals we aim for are ones worth striving to achieve? In my view, a free society gives us the best chance of achieving the latter goal. Only a free society allows us a choice, and even if our choice is set long before we make it, knowing that other individuals could have made another choice remains important.

Social Cooperation Is the free market argument in need of rebranding?

The standard free market analysis places the individual at its center. As Adam Smith famously noted in 1776, although they act in their own self interest, an individual in a free market is “led by an invisible hand to promote an end which was no part of his intention.” And it is generally argued that competition is the driving force behind the benefits of the market process. Entrepreneurs constantly search for new opportunities to make profit, and as a result they find more efficient ways to provide goods to consumers.

Phrased in this way, the free market argument tends to evoke images of Social Darwinism – the best rise to the top, and the weak are left behind. Competition implies a constant struggle between market participants to seek their own benefit at the expense of others. This vision often leads critics to argue that the free market ideal generates an uncaring society. If everybody acts only in their own self interest, there is no room for cooperative behavior that is essential for human interaction. Morality, emotion, personal connections – none of it matters. The free market places “profit over people.”

This critique stems from a wildly incorrect reading of the free market argument.

Go back to Adam Smith. At the center of his work is the idea of division of labor. A market economy thrives not because individuals work in isolation. Instead, it depends entirely on the relationships between individuals, focusing each person’s talents on an activity where they possess a comparative advantage.

Perhaps the best illustration of the role of cooperation in a market economy is Leonard Read’s famous essay I, Pencil (there is also an excellent video inspired by the essay). Read points out that no individual on their own knows how to make even something as simple as a pencil. The production process requires dozens of firms and hundreds of workers each performing specialized tasks with little knowledge of the final product. There is no planner describing how to make a pencil and yet through the actions of individuals as well as the interactions between individuals, the production process arises spontaneously. An individual acting alone would quickly fail in a market economy.

Ludwig von Mises’s famous treatise Human Action, a comprehensive analysis of the working of the free market system, was almost given a different titleSocial Cooperation. Although Mises dropped this alternate title, the theme that markets depend as much on cooperation between individuals as they do on individual action itself runs throughout the book. Mises notes:

Within the frame of social cooperation there can emerge between members of society feelings of sympathy and friendship and a sense of belonging together. These feelings are the source of man’s most delightful and most sublime experiences. They are the most precious adornment of life; they lift the animal species man to the heights of a really human existence.
Human Action p. 144

A free market, in Mises’s view, doesn’t destroy relationships between individuals, but instead fosters these feelings. Even if we take the idea of “Social Darwinism” seriously, even if we admit that all individuals are driven by the desire to fight for their own survival, that doesn’t lead us to a world of selfishness (in a narrow sense) because “the most adequate means of improving his  condition is social cooperation and the division of labor” (Human Action, p. 176).

But the argument that markets and morals are inconsistent faces an even deeper flaw. In a market economy we have a choice. Of course we can choose to think only of ourselves, to put money over family, to value material goods over relationships. But that has absolutely nothing to do with the free market itself. Nothing in the market argument says that I should only care about wealth. If you want to put other priorities first, nobody in a free market has any right to stop you (the catch is that you also don’t have any right to make other people pay you).

A free market doesn’t place any moral judgement on the actions of individuals. It is perfectly consistent with both a savage society where everybody fights for their narrow self interest and ignores others as well as a responsible one where we care for our fellow humans. It is up to each of us as individuals to choose to live our lives morally (but of course, this choice is only an illusion).

What is the alternative? The only clear alternative I can see is to use the state to try to impose your morals on others. By enacting laws that force people to behave morally, maybe we can create a more caring society.

Such a system seems doomed to have the opposite effect. Let’s say you believe that redistribution of wealth is important. Poor people aren’t poor because they didn’t work hard. They just had bad luck. It’s the responsibility of the rich to help these people out. I am sympathetic to this reasoning. However, by forcing people to give up their wealth through taxation, we change the equation from one of responsibility to one of coercion. Rather than giving to the poor out of some sense of moral duty, I give because I don’t want to go to jail. Is attempting to legislate morality in this way more likely to generate a caring society or a resentful one? Respect between classes, or class warfare?

The free market argument should not marry itself to the individual. It is true that all actions must at their core come from individual decisions, but the market only works through the relationships between individuals. Human Action is only half of the story. Social Cooperation is equally important. By obscuring this fact, defenders of markets concede too much. Emphasizing efficiency and the incredible material progress society has made since adopting a market system is fine, but we can’t ignore the moral argument. Morality can’t be imposed. It has to be a choice. And only a free society offers that choice. “Liberty is an opportunity for doing good, but this is so only when it is also an opportunity for doing wrong” (Hayek, The Constitution of Liberty, p. 142).

What’s Wrong With Modern Macro? Part 9 Carrying on the Torch of the Market Socialists

Part 9 in a series of posts on modern macroeconomics. This post lays out a more philosophical critique of common macroeconomic models by drawing a parallel between the standard neoclassical model and the idea of “market socialism” developed in the early 20th century. 


If an economist had access to all of the data in the economy, macroeconomics would be easy. Given an exact knowledge of every individual’s preferences, every resource available in the economy, and every available technology that could ever be invented to turn those resources into goods, the largest problem that would remain for macroeconomics is waiting for a fast enough computer to plug all of this information into. As Hayek pointed out so brilliantly 60 years ago, the reason we need markets at all is precisely because so much of this information is unknown.

Read any macroeconomics paper written in the last 30 years, and you’d be lucky to find any acknowledgement of this crucial problem. Take, for example, Kydland and Prescott’s 1982 paper that is widely seen as the beginning of the DSGE framework. The headings in the model section are Technology, Preferences, Information Structure, and Equilibrium. Since then, almost every paper has followed a similar structure. Define the exact environment that defines a market, and then equilibrium prices and allocations simply pop out as a result.

What’s wrong with this method of doing economics? To understand the issue, we need to take a step back to an earlier debate.

Mises’s Critique of Socialism

In 1922, Ludwig von Mises published a book called Socialism that remains one of the most comprehensive and effective critiques of socialism ever written. In it, he developed his famous “calculation” argument. Importantly, Mises’s argument did not depend on morality, as he freely admitted that “all rights derive from violence” (42). Neither did his argument depend on the incentives of social planners. “Even angels,” claims Mises, “could not form a socialist community” (451). Instead, Mises makes a far more powerful argument: socialism is practically impossible.

I can’t explain the argument any better than Mises himself, so here is a quote that makes his main point

Let us try to imagine the position of a socialist community. There will be hundreds and thousands of establishments in which work is going on. A minority of these will produce goods ready for use. The majority will produce capital goods and semi-manufactures. All these establishments will be closely connected. Each commodity produced will pass through a whole series of such establishments before it is ready for consumption. Yet in the incessant press of all these processes the economic administration will have no real sense of direction. It will have no means of ascertaining whether a given piece of work is really necessary, whether labour and material are not being wasted in completing it. How would it discover which of two processes was the more satisfactory? At best, it could compare the quantity of ultimate products. But only rarely could it compare the expenditure incurred in their production. It would know exactly—or it would imagine it knew—what it wanted to produce. It ought therefore to set about obtaining the desired results with the smallest possible expenditure. But to do this it would have to be able to make calculations. And such calculations must be calculations of value. They could not be merely “technical,” they could not be calculations of the objective use-value of goods and services; this is so obvious that it needs no further demonstration
Mises (1922) – Socialism p. 120

Prices are what allow calculation in a market economy. In a socialist economy, market prices cannot exist. Socialism therefore is doomed to fail regardless of the intentions or morality of the planners. Even if they want what is best for society, they will never be able to achieve it.

The Response to Mises: Market Socialism

Although Mises argument was effective, the socialists weren’t prepared to give up so easily. Instead, a new idea was offered that attempted to provide a method of implementing socialist planning without falling into the problems of calculation outlined by Mises. Led by Oskar Lange and Abba Lerner among others, the seemingly contradictory idea of “market socialism” was born.

Lange’s argument begins by conceding that Mises was right. Calculation is impossible without prices. However, he argues there is no reason why those prices have to come from a market. Economic theory has already demonstrated the process through which efficient markets work. In particular, prices are set equal to marginal cost. In a market, this condition arises naturally from competition. In a market socialist society, it would be imposed by a planner. In Lange’s view, not only would such a rule match a market economy in terms of efficiency, but it could even offer improvements by dealing with problems like monopoly where competition is unable to drive down prices.

We are left with one final problem, which is the pricing of higher order capital goods. Lange admits that these goods pose a more difficult problem, but insists that the problem could be solved in the same way the market solves it: trial and error. In other words, just as entrepreneurs adjust prices in response to supply and demand, so could social planners. When demand is greater than supply, increase prices and when supply is greater than demand, reduce them. At worst, Lange argues, this system is at least as good as a free market and at best it is far better since “the Central Planning Board has a much wider knowledge of what is going on in the whole economic system than any private entrepreneur can ever have” (Lange (1936) – On the Economic Theory of Socialism, Part One p. 67)

The Market Socialist Misunderstanding: Why do Markets Work?

Lange’s argument should be appealing to anybody that takes standard economic theory seriously. A Walrasian General Equilibrium gives us specific conditions under which an economy operates most efficiently. We talk about whether decentralized competition can lead to these conditions, but why do we even need to bother? We know the solution, why not just jump there directly?

But by taking the model seriously, we lose sight of the process that it is trying to represent. For example, in the model the task of a firm is simple. Perfect competition has already driven prices down to their efficient level and any deviation from this price will immediately fail. As Mises emphasized, however, the market forces that bring about this price have to come from the constant searching of an entrepreneur for new profit opportunities. He concedes that “it is quite easy to postulate a socialist economic order under stationary conditions (Socialism, 163).” Conversely, the real world is characterized by constantly shifting equilibrium conditions. The market socialist answer assumes that we know the equations that characterize a market. Mises argues that these equations can only come through the market process.

Hayek also made an essential contribution to the market socialism debate through his work on the role of the price system in coordinating market activity. For Hayek, prices are a tool used to gather pieces of knowledge dispersed among millions of individuals. An entrepreneur does not need to know the relative scarcities of various goods when they attempt to choose the most efficient production process. They only need to observe the price. In this way, Lange’s pricing strategy cannot hope to replicate the process of a dynamic economy. When prices are no longer set by market participants looking to achieve the best allocation of resources they lose almost all of their information content.

The final piece missing from Lange’s analysis is perhaps also the most important: profit and loss. In Lange’s depiction of a market economy, he seems to imagine one that is already close to an equilibrium. In particular, he assumes that the production structures in place are already the most efficient. Without this assumption, there is no way to determine the marginal cost and no way to use trial and error pricing effectively. Mises and Hayek instead view an economy as constantly moving towards an equilibrium but never reaching it. Entrepreneurs constantly search for both new products to sell and new methods to produce existing products more efficiently. Good ideas are rewarded with profits and bad ones driven out by losses. Without this mechanism, what incentive is there for anybody to challenge the existing economic structure? Lange never provides an answer.

For a longer discussion on the socialism debate, see a paper I wrote here.

Repeating the Same Mistakes

So what does any of this have to do with modern macroeconomics? Look back to the example I gave at the beginning of this post of Kydland and Prescott’s famous paper. Like the market socialists, their paper begins from the premise that we know all of the relevant information about the economic environment. We know the technologies available, we know people’s preferences, and we assume that the agents in the model know these features as well. The parallels between the arguments of Lange and modern macroeconomics are perhaps most clear when we consider the discussion of the “social planner” in many macroeconomic papers. A common exercise performed in many of these studies is to compare the solution of a “planner’s problem” to that of the outcome of a decentralized competitive market. And in these setups, the planner can always do at least as good as the market and usually better, so the door is opened for policy to improve market outcomes.

But because most macro models outline the economic environment so explicitly, competition in the sense described by Mises and Hayek has once again disappeared. The economy in a neoclassical model finds itself perpetually at its equilibrium state. Prices are found such that the market clears. Profits are eliminated by competition. Everybody’s plans are fulfilled (except for some exogenous shocks). No thought is given to process that led to that state.

Is there a problem with ignoring this process? It depends on the question we want to answer. If we believe that economies are quick to adjust to a new environment, then the process of adjusting to a new equilibrium becomes trivial and we need only compare results in different equilibria. If, however, we believe that the economic environment is constantly changing, then the adjustment process becomes the primary economic problem that we want to explain. Modern macro has heavily invested in answering the former question. The latter appears to me to be far more interesting and far more relevant. The current macroeconomic toolbox offers little room to allow us to explore these dynamics.

 

 

Competition and Market Power Why a "Well-Regulated" Market is an Impossible Ideal

Standard accounts of basic economics usually begin by outlining the features of “perfect competition.” For example, Mankiw’s popular Principles of Economics defines a perfectly competitive market as one that satisfies the following properties

  1. The goods offered for sale are all exactly the same
  2. the buyers and sellers are so numerous that no single buyer or seller has any influence over the market price.

    The implication of these two properties is that all firms in a perfectly competitive market are “price takers.” If any firm tried to set a price higher than the current market price, their sales would immediately drop to nothing as consumers shift to other firms offering the exact same good for a cheaper price. As long as firms can enter and exit a market freely, perfect competition also implies zero profits. Any market experiencing positive profits would quickly see entry as firms try to take advantage of the new opportunity. The entry of new firms increases the supply of the good, which reduces the price and therefore pushes profits down.

After defining the perfectly competitive market, the standard account begins to extol its virtues. In particular, a formal result called the First Welfare Theorem shows that in a perfectly competitive equilibrium, the allocation of goods is Pareto efficient (which just means that no other allocation could make somebody better off without making somebody else worse off). So markets are great. Without any planner or government oversight of any kind, they arrive at an efficient outcome on their own.

But soon after developing the idea, we begin to poke holes in perfect competition. How many markets can really be said to have a completely homogeneous good? How many markets have completely free entry and no room for firms to set their own price? It’s pretty hard to answer anything other than zero. Other issues also arise when we begin to think about the way markets work in reality. The presence of externalities (costs to society that are not entirely paid by the individuals making a decision – pollution is the classic example), causes the first welfare theorem to break down. And so we open the door for government intervention. If the perfectly competitive market is so good, and reality differs from this ideal, doesn’t it make sense for governments to correct these market failures, to break up monopolies, to deal with externalities?

Maybe, but it’s not that simple. The perfect competition model, despite being a cool mathematical tool that is sometimes useful in deriving economic results, is also an unrealistic benchmark. As Hayek points out in his essay, “The Meaning of Competition,” the concept of “perfect competition” necessarily requires that “not only will each producer by his experience learn the same facts as every other but also he will thus come to know what his fellows know and in consequence the elasticity of the demand for his own product.” When held to this standard, nobody can deny that markets constantly fail.

The power of the free market, however, has little to do with its ability to achieve the conditions of perfect competition. In fact, that model leaves out many of the factors that would be considered essential to a competitive market. Harold Demsetz points out this problem in an analysis of antitrust legislation.

[The perfect competition model] is not very useful in a debate about the efficacy of antitrust precedent. It ignores technological competition by taking technology as given. It neglects competition by size of firm by assuming that the atomistically sized firm is the efficiently sized firm. It offers no productive role for reputational competition because it assumes full knowledge of prices and goods, and it ignores competition to change demands by taking tastes as given and fully known. Its informational and homogeneity assumptions leave no room for firms to compete by being different from other firms. Within its narrow confines, the model examines the consequences of only one type of competition, price competition between known, identical goods produced with full awareness of all technologies. This is an important conceptual form of competition, and when focusing on it alone we may speak sensibly about maximizing the intensity of competition. Yet, this narrowness makes the model a poor source of standards for antitrust policy.
Demsetz (1992) – How Many Cheers For Antitrust’s 100 Years?

Although the types of competition outlined by Demsetz are a sign of market power by firms, they are not necessarily a sign that the market has failed or that governments can improve the situation. Let me tell a simple story to illustrate this point. Assume a firm develops a new technology that they are able to prevent other firms from immediately replicating (either because of a patent, secrecy, a high fixed cost of entry, etc.). This firm is now a monopoly producer of that product and can therefore set a price much higher than its cost and make a large profit. The government sees this development and orders the firm to release its plans so that others can replicate the technology and produce their own version. Prices fall as new firms enter and profits go to zero. Consumers are better off since prices are lower and they have a larger choice of products. (A similar story could also be told if the government simply mandated a lower price by monopoly firms).

But the story isn’t over. If I’m another entrepreneur (or even an existing firm) watching this sequence of events, I’m a bit worried. That new idea I was thinking about is going to cost a lot. If I had the possibility to make a large profit, maybe I would be willing to take the risk and go for it anyway. If, on the other hand, I knew for sure that even when I achieve success the government immediately reduces my profits to zero, am I still going to undertake that project? Not a chance. The potential for future profits is an incredibly important incentive for innovation.

Here’s another example from the real world that illustrates the opposite case. In the late 1990s, Microsoft tried to bundle Internet Explorer with their Windows operating system (essentially giving away Explorer for free). This move made it difficult for independent internet browsers to compete (Netscape was the market leader at the time). An antitrust lawsuit was brought against Microsoft and they were initially ordered to break up (which never actually occurred in the end as far as I know, but that doesn’t matter for the story). In the EU, they were required to provide a browser choice page when installing Windows.

In each of the two examples above, there is a clear tradeoff. In the first, consumers are better off in the short run (lower prices), but potentially worse off in the long run (less innovation). The second case is exactly the opposite. Consumers are worse off in the short run (they don’t get a browser for free) but potentially better off in the long run (more browser competition). Can we say for sure whether regulation helps or hurts in either case? Can we even say whether the regulation would push the market to be more competitive or less? I don’t see how (but which browser you are using right now despite the relatively lenient restrictions on Microsoft might give some indication).

I’m not saying regulation is never a good idea in theory. But in practice, it turns out to be really hard. Even in the cases above where it is obvious that a firm is trying to take advantage of monopoly power, it remains unclear whether a move closer to “perfect competition” will result in an increase in actual competition. You can of course pick apart the stories above and come up with some regulatory scheme that balances present and future costs and benefits. But doing so in general would require governments to have even more information than the already ridiculous knowledge assumptions implicit in the perfect competition model. It’s easy to point out imperfections in markets. It’s much harder to figure out what to do about them.

Notice that I haven’t necessarily made an argument against regulation. The takeaway from this post should not be that markets always work or that regulation always fails (I’ll leave that for future posts!). My point is simply that pointing out a flaw in the free market does not automatically imply an opportunity for a regulatory solution. The question is much more complicated than that.

But having said that let me leave you with one final thought. Markets are incredibly dynamic. Whenever the market “fails,” all it takes is one clever entrepreneur to come up with a better method and correct the failure. When government fails? Well, maybe we can come back to that in ten years when they get around to discussing it.