I promised to write a post on why some economists disagree with current economic stimulus, but the economics bloggers are beating me to the punch again. Menzie Chinn has a nice introductory piece explaining their view at Econbrowser:

The Current Downturn: Labor-leisure tradeoff or technological regress

In that post, Chinn discusses the ideas of the Real Business Cycle (RBC) economists, who seem to be the majority of those who oppose the spending stimulus and want only tax cuts, and seem to have had the ears of those in the media opposing it.

I do not think Chinn gets to the conceptual heart of the matter. So there is an excuse for me to put in my two cents. Regardless of whether you agree with opponents or not, it is a good idea to understand those who have an opposing view. First, they may have some good points that you have not considered, even if you finally decide they are mostly wrong. And second, it helps you to respond to those who are opposing the stimulus in the public debate.

Ultra free market economics starts with the proposition that, left alone, there is a general equilibrium on all markets that balances supply and demand. Economic forces drive society towards this general equilibrium, and we are never far away from it. If government does more than protect property rights, ensure competition, and protects us from violence, it can only reduce our well being.

Economists like to point out concrete, specific, real world examples of how individual economic markets adjust and reach equilibrium very quickly always and everywhere, in order to convince us of the universal validity of economic thinking. The Freakonomics book and blog are examples of this, though I do not know whether the writers are ultra free market advocates or not.

I think it is true that we can see economic adjustment on individual markets all the time, they do drive towards some kind of equilibrium on that individual market rapidly, and these are very powerful forces. But, to advocate for an unregulated free market on a large social scale and for the economy as a whole, these examples of individual market adjustments to equilibrium miss the point. What also has to be true is that all these individual markets must fit together into a coherent general equilibrium of all markets, and also that can sustain itself over time. There might be equilibrium at one time or another on most individual markets, but not a general simultaneous equilibrium on all markets that can continue over time.

To use an analogy, you can imagine a car that works perfectly and runs with ruthless efficiency part by part, but the whole machine does not work when all the parts are put together. Suppose you had a car with the engine of NASCAR racer, the steering and braking of a Buick, and the wheels of a Soapbox cart. Each may work wonderfully for a particular purpose, but put them together and the whole machine does not work well at all.

A long standing dream of economists slowly came true between 1948 and 1970, during which a general equilibrium was shown to exist mathematically under a very wide variety of conditions. But there are several problems. The biggest problem is that no once has ever discovered a way to identify when a real economy is in a general equilibrium, using any data that we have, or could conceivably obtain.

Now economists need to have an equilibrium. Unlike other sciences, economics has no commonly accepted theory of how an economy behaves out of equilibrium. So they assume that the economy is always at or close to some kind of equilibrium where demand equals supply on all or most markets.

Keynesian economists believe in economic equilibrium –they are also helpless without one. But they believe that in certain circumstances the economy can get into an unemployment equilibrium where businesses would like to hire and invest, and workers would like to work, but for some reason they cannot reach a deal. So, there is an equilibrium where supply equals demand, but it is one where some people are left unemployed but who would like to be employed. They go to the labor market regularly and offer their work at the market price, but they don’t get work, so they stop going and drop out of the labor supply.

There are several theories for how this can happen. One theory explaining unemployment equilibrium is price and wage rigidities –a shock hits the economy and prices and wages cannot adjust to find a new equilibrium. Joseph Stiglitz and Bruce Greenwald showed that after such a shock firms that are risk averse will find it safer to reduce how much they produce rather than reduce their prices or wages. Another source of rigidity is that all prices (wages, products, rents) and the value of loans and debts, do not change all at once. Some change a lot and change quickly, but others don’t. Another is coordination failures and frictions–some shock destroys usual connections and information flows between business and workers, and economic activity falls apart faster than they can make new connections and contracts.

RBC economists have a very different view. They believe that often unemployment and business downturns are optimal responses to some shock to the economy. Decentralized individual decision making is the best way to get back to a welfare maximizing equilibrium, and government interference can only harm people.

So, RBC macroeconomists don’t believe in unemployment equilibrium. But they cannot directly show that a general full employment equilibrium exists. So, their strategy is to write down a model that assumes a full employment equilibrium with model businesses and consumers who maximize their profits and welfare over time. Their guiding assumption is that the market is always at, or close to, full employment equilibrium, and would continue at one forever, but outside shocks in technology, or resource shortages, etc, hit the economy, which then adjusts and recovers. That is they try to show that their concept of full employment equilibrium can explain that conception of the economic adjustment process, which is basically their interpretation of the business cycle.

They model businesses and consumers using what they consider reasonable parameters describing preferences for leisure/labor, discount rates for future vs. current consumption, levels of risk aversion, and other characteristics. Then they run these model economies and see if they can reproduce common features of the aggregate macroeconomy. Examples would be interest rates, investment, consumption, real wages –both their movements and comovements (or correlations) over time . If they can reproduce the behavior of the economy with their models then they take that as evidence that their theory of the economy is true.

Of course Keynesian economists can play this game too. Until recently, purely statistical models beat most economic theory driven models in reproducing important macroeconomic behavior. Unfortunately for statisticians, recently both Keynesian and RBC models have been about as good as purely statistical models (that are consistent with either theory).

I think one of the reasons that there are few estimates of what normal people would consider important variables, like multipliers, is that many economists have been spending time in this modeling race, trying to see who could produce models that reproduced the behavior of the aggregate economy better.

This might be considered a waste of time, but it is important to remember that different theories might predict the same observable responses to government policy, but have very different implications for the peoples’ welfare.

For example, normal people (say workers and small businesspeople) are obsessed with these multipliers, and they want them to be BIG (whether tax or spending). An RBC economist could see a big government spending multiplier effect, but that would not change their mind that government spending was a mistake. I will give a simple example, using a government spending stimulus program.

What the aggregate statistics would say is this: Economic activity, GDP, was declining. The government engaged in deficit $1 billion stimulus spending. Later it was found that GDP was $1.5 billion higher than it was expected to be. The multiplier is estimated to be 1.5. The government stimulus worked! Everyone is better off!

An RBC economist might see something different. There was a shock to the economy and workers became unemployed as labor and business began their optimal individual adjustments. Some workers choose not to work, some businesses choose not to produce as much at current market prices. But so what? Maybe they have better things to do during the adjustment process. The government spent $1 billion dollars in stimulus, which produced no productive opportunities for the economy that were worth the cost. Workers and business understand that all that had happened was that the government would have to raise taxes in the future to pay for this unproductive government expenditure. Everyone realizes that they will have to pay more taxes in the future, out of incomes that are lower because the government went and wasted this money. So, workers work harder and businesses produce more in order to make up for this loss in wealth. The GDP increases, and the estimated multiplier is 1.5, but the stimulus made everyone worse off.

The lesson here is that the observable data that most people would look at in order to evaluate the success of a policy, DO NOT distinguish between the rival theories, at least for economists. Recently, a lot of effort has gone into trying to estimate things that are usually different in the Keynesian and RBC theories. For example, in Keynesian theories, consumption usually rises after a government spending stimulus. In most RBC theories, consumption falls; this is because people are working harder after the stimulus only because their wealth has been reduced by wasteful government spending. So they earn more now than they otherwise would have, but consume less now so they will have more later when they have to pay for the wasteful government spending through higher taxes.

A lot of research has been looking at consumption behavior following government spending and tax fiscal policies to see if that can distinguish between the two theories. The results, I think, generally favor the Keynesians (consumption rises after fiscal spending shocks), but there are some results that support the RBC theory. And there are methodological problems. For example, many of the big government spending stimuli have been during wars, when government has no interest in increased consumption, and adopts policies to reduce it. Also the results are sensitive to details of the economic models and estimation techniques that should not make a big difference, but do.

I think the bottom line is that because there is no direct test of when the economy is in equilibrium, so economists must use odd and indirect routes to support their theories. And this leads to research that seems odd and indirect, and maybe silly, to most practical people. But if you have a better approach, let everyone know, and you will become famous.

It is kind of late early morning now, and also Friday the 13th, so I will end here. I will post a conclusion later, and will provide references for further reading.