Raising the status of real business cycle theory

Garett Jones's guest post on El Salvador reminded me of a more general point: real business cycle theory explains 98 percent or more of the business cycles in human history, especially before the eighteenth century and the advent of modern financial markets.  Some parts of the world still are ruled by real business cycle theory.

And yet somehow the theory has low status or sometimes is considered outside the mainstream.  Admittedly, the theory can be misapplied or oversimplified by its proponents.  Yet…it explains 98 percent or more of the business cycles in human history.  It also helps explain the propagation mechanisms of monetary-based cycles.

I call that an important theory.

If only there were a textbook which taught RBC!


Where did you get the 98% from ? What does this mean - you have a recession and the RBC guys find a "productivity" shock to explain the recession ? More generally, how do you define RBC theory ? Most mainstream macroeconomists would hardly consider it to be outside the mainstream. Poorly thought through posting.

Like Mathusianism?

Can you just add Garrett Jones as a regular blogger already?

So in the present-day US is the marginal value of RBC close to zero?

Spring floods could explain 98% percent of all damage to unprotected houses or cities on the river.

Although Spring floods may explain the damage or potential damage, that doesn't explain why people build levies and do things to avoid or mitigate damage, such a build a levee or dam to control the spring floods.

The question should be: What can be done to prevent or avoid damage, or even better one's position by building dam or levee so you could have a city with really cool housing on the river. Or, applying it to the economy, avoiding or mitigating damage from a financial crisis.

We should know more than we did in the 18th century. Both in managing water or managing an economy, notwithstanding our knowledge that we will have spring rains or financial crises from time to time. We don't need to be so passive.

I think Barro's textbook is close to RBC. For better or worse.

RBC is mostly graduate student stuff.

The history of the IS-LM model of Keynes moving into undergraduate texts tells us something: moving too fast to model a theory and make it textbook is a bad idea.

Were there no financial panics or banking crises in the ancient world, in the medieval world?

I'm not sure what to make of this. First, how do we define "human history"? I'm sure that if we stretch back far enough, we can make RBC theory account for an arbitrarily high share of business cycles -- I don't think that monetary or financial issues applied to cavemen, even though the basic logic of RBC still held. But the fact that an issue has been prominent throughout human history doesn't automatically make it important today -- for one, we don't live in tribes.

I've always been bugged by claims that RBC "explains" some percentage of business cycles, because they seem to me to reflect a very weak standard for what constitutes an "explanation". If you take measured Total Factor Productivity as exogenous and stick in the right parameters, then yes, RBC does "explain" most recessions. (Though not this one -- we've seen countercyclical productivity! I suspect that this is due to composition effects wherein the less skilled workers were the ones who lost their jobs, not because productivity in the abstract has actually risen. But this just goes to show why it is important to interpret what economic statistics actually mean, and not naively assume that a number measures productivity because that's what the statistical agency calls it.)

I suspect that I could pick many alternative models from the infinite-dimensional space of macro models that would "explain" business fluctuations in the same shallow way. This is virtually a statistical inevitability, which explains why Ed Prescott hates statistics. Indeed, this is a deep problem in macroeconomics, and to address it we must impose further sanity checks on our models, making sure that they are consistent with microfounded intuition and data. But RBC clearly fails that test -- there's no matching movement in real wages, and a moment's introspection will tell you that most unemployment during recessions is not voluntary.

Isn't the vast majority of macro research these days in the vein of RBC and DSGE models? The idea that it is under-appreciated is somewhat surprising?

"It's not so much the advent of modern financial markets (financialization as we know it only dates to the late 70s) but the advent of modern borrowing as a significant percentage of GDP, which makes the money supply volatile.

Even a pure barter economy would exhibit monetarist business cycles if borrowing is permitted."

This post makes so little sense it is hard to even criticize it.

If there was a pure gold standard and no monetary tinkering would there still be business cycles? Why do the individual-perspective fluctuations of bearishness and bullishness reinforce each other rather than damping out? Where does the coherence or resonance arise from?

The funding of Columbus by Spain's monarch caused an economic boom, and inflation, and more.

Lincoln signing the bill that funded building the Transcontinental Railroad created a boom in California.

The British government pursuing empire through colonialism fueled a boom in global trade and in the economy of England.

Hoover pushing for the sped up construction of the Hoover Dam created a boom in Boulder, CO.

The GI Bill produced a boom in the US economy in housing and higher education after WWII demobilization.

How is a Keynesian not a believer in RBC?

Or are changes in government policy in response to events not real "shocks" to the economy? Is the end of a war not a real shock to the economy?

On smaller scale "shocks", was the British government investment in celestial navigation, like an accurate clock, which led to British domination not real?

The benefit of the accurate clock is spread thinly around the economy and requires other similarly small developments like accurate charts and education in the art of navigation. Without an agency that can benefit from all the dispersed benefits of many coordinated efficient allocations of resources, those allocations are not made, or not made as heavily as is globally optimal.

Lincoln didn't need to sign the law to ensure the railroad would be built to the West Coast, but the completion a decade or two later would not have provided the needed counter shock to the shock of the Civil War demobilization. And California might not have become a State.

Question: Would you say RBC theory did a decent job predicting the just recent (or to some current) recession? What would RBC theorists say caused it? I think the answers to those two questions suggest that RBC is not so good a theory.

It looks to me like RBC is a general theory, that explains a big picture. The point is to see the big picture, not to predict the future.

I will contrast it with vague fuzzy explanations of a couple of other big pictures.

Keynes said that the economy had some missing feedback loops, and as a result it would occasionally get out of balance in bad ways. Government could palliate the problem and get things running again.

Marx said that capitalist economies had some missing feedback loops that would inevitably result in a series of increasingly deep depressions. Presumably modern Marxists would say that government palliation itself has missing feedback loops so that the system will still inevitably fail repeatedly and deeper.

Both concepts say that there is something wrong with the economy, and they propose methods to fix it.

RBC says there's nothing wrong with the economy, that the economy is doing the right thing no matter what it does, and anybody who tries to fix it will only mess it up.

Here is a metaphor, from bacterial cellular physiology. Bacteria have evolved over the last 3 billion years or so, and in favorable conditions some modern bacteria can go through 50 generations a day. There's every reason to think that they lack no important feedback systems for their core functions. Bacteria collect energy sources and raw materials and use them to create everything that makes up bacteria -- structures, knowledge, capital equipment, etc.

Bacteria have feedback loops that result in them increasingly fine-tuning the amount produced of each kind of capital equipment. The capital equipment self-assembles into assembly lines that produce intermediate products and finished goods, and there are feedback loops at each place an assembly line branches, to regulate the amount of each product. They do a very good job of this, particularly for the things that have been done by large numbers of bacteria for a long time.

Now, what happens if a cell has only acetic acid for an energy source, and then you give it a lot of glucose? Glucose is a much better energy source so it switches. But it has to create the capital equipment to metabolize glucose. It has to transport glucose into the cell and break it into useful pieces. And it no longer needs the machinery that made the same things out of acetate.

So the cell stops growing for awhile and starts gearing up for the new energy source. After awhile it grows faster than before, but there's a lag.

The cell then multiplies. Two and four and eight and so on. One generation, the concentration of glucose will drop a little bit. The next generation twice as many cells will use up as much glucose as had been used in the whole history. The concentration drops more. Soon there is only half as much glucose as we started with. And one generation later there is none.

In perhaps 3 generations the bacteria must notice the problem and do something about it. They have probably been making some waste products, things that were not worth using when they had glucose. They can gear up to metabolize those. And the remaining acetate. Do they all start to use the best of them, and when that runs out then use the next best, and so on? No, they somehow pick different inferior energy sources so that everything gets used at once by different cells. This uses up the resources more efficiently.

When they run out of everything that actually allows growth, then they start burning up their own bodies for energy. A fast-growing cell needs lots of ribosomes to make proteins. A slow-growing cell needs a few. They can use the extras for food. But of course, when a new energy source shows up they can't adapt as fast because it will take them longer to retool.

So even with what is probably a completely optimal strategy, bacteria get periods of slow growth when they respond to environmental changes, and the slow growth tends to happen even when the change will let them grow faster than before.

When they "had reason to expect" the environment to be stable for awhile, they grew faster by keeping no reserves to help them adapt faster. Then when an unexpected change does happen they must somehow divert resources just to provide the capital to retool. Until the new assembly lines are working, it costs resources to make them which will only be paid back later, so they grow slow at first.

By analogy, we can expect slowdowns from response to external changes even if we get all the right feedback loops in an economy. It's probably unavoidable.

Does our economy have all the feedback loops that would be useful to it? I don't know. I've seen various persuasive arguments that it does not. But the fact that depressions happen does not prove that there is something wrong with the economy. Such things could possibly happen even if we could get the particular economy that grows the fastest possible on average.

If the economy was perfect, then we could point to an "external shock" for each recession. If the economy was not perfect, we could point to increasing imbalances that eventually tilt into recession.

So for example, we did not immediately get a bad recession from Enron, which would have been an obvious trigger if the recession had happened then. And when we found all the major accounting firms were giving crooked reports so that we could not trust the annual reports of any US public corporation, that would have been a fine shock to trigger a recession. And then when we had a housing bust and lots of people who had bought expensive houses hoping to sell them for a profit found out they couldn't sell without a big loss, that would have been a good time. And then when the financial tricks connected to the houses turned into a swamp where nobody knew whether their bank was about to go banrupt or not -- finally that did it.

If Enron had waited to happen until we were ripe for recession, would that have been enough to cause recession? I think so. I think when we're ripe for it, it will happen and we can pick some event as the cause.

What makes us really ripe for it? We've used up more than half the oil and there's no decent alternative energy source in sight. Coal gives us a limited amount of dirty energy at increasing cost. Nuclear gives us an unlimited amount of expensive energy with pollution from hell. Biofuels are currently a mirage -- starch to alcohol is worthless, but someday we might get somewhere with cellulose. Wind power is OK but like hydro it's limited, it needs the right site and we won't know the expense until we know how long the equipment lasts in nature. We don't know what to do, and if we choose wrong we'll have giant expenses for a bad system. So we wait. We eat our ribosomes while we wait for a better choice....

Dirk, you are not thinking in terms of feedback loops.

Exogenous causes result in responses. Sometimes even a wonderful event can result in delays and slowdowns while the economy gears up to respond.

The other side though is that feedback systems often result in limit cycles (think rabbit lynx trapper in Canada etc). Or they can wander erraticly, or oscillate increasingly.

It's only natural for people who have not studied control theory and who have not actually ever built any feedback systems to assume that economic control loops must function adequately or even optimally for some purpose. They don't know any better.

Comments for this post are closed