Keynesian and "Austrian" Speculation on the Nature of Recessions

Keynesian economist Nouriel Roubini writes at Project Syndicate:

Unlike the 2008 global financial crisis, which was mostly a large negative aggregate demand shock, the next recession is likely to be caused by permanent negative supply shocks from the Sino-American trade and technology war. And trying to undo the damage through never-ending monetary and fiscal stimulus will not be an option.

This, of course, is a type of Keynesian take on how a recession might occur.

It is an analysis almost as bad as some new "Austrian" commentary I am seeing that claims a recession could be started because of a downturn in one industry.

Both are nonsense.

They both violate Murray Rothbard's explanation of a recession: 

It is important, first, to distinguish between business cycles and ordinary business fluctuations. We live necessarily in a society of continual and unending change, change that can never be precisely charted in advance. People try to forecast and anticipate changes as best they can, but such forecasting can never be reduced to an exact science. Entrepreneurs are in the business of forecasting changes on the market, both for conditions of demand and of supply. The more successful ones make profits pari passus with their accuracy of judgment, while the unsuccessful forecasters fall by the wayside. As a result, the successful entrepreneurs on the free market will be the ones most adept at anticipating future business conditions. Yet, the forecasting can never be perfect, and entrepreneurs will continue to differ in the success of their judgments. If this were not so, no profits or losses would ever be made in business.

Changes, then, take place continually in all spheres of the economy. Consumer tastes shift; time preferences and consequent proportions of investment and consumption change; the labor force changes in quantity, quality, and location; natural resources are discovered and others are used up; technological changes alter production possibilities; vagaries of climate alter crops, etc. All these changes are typical features of any economic system. In fact, we could not truly conceive of a changeless society, in which everyone did exactly the same things day after day, and no economic data ever changed. And even if we could conceive of such a society, it is doubtful whether many people would wish to bring it about. 

It is, therefore, absurd to expect every business activity to be “stabilized” as if these changes were not taking place. To stabilize and “iron out” these fluctuations would, in effect, eradicate any rational productive activity. To take a simple, hypothetical case,
suppose that a community is visited every seven years by the sevenyear locust. Every seven years, therefore, many people launch preparations to deal with the locusts: produce anti-locust equipment, hire trained locust specialists, etc. Obviously, every seven
years there is a “boom” in the locust-fighting industry, which, happily, is “depressed” the other six years. Would it help or harm matters if everyone decided to “stabilize” the locust-fighting industry by insisting on producing the machinery evenly every year, only to have it rust and become obsolete? Must people be forced to build machines before they want them; or to hire people before they are needed; or, conversely, to delay building machines they want—all in the name of “stabilization”? If people desire more autos and fewer houses than formerly, should they be forced to keep buying houses and be prevented from buying the autos, all for the sake of stabilization? As Dr. F.A. Harper has stated:

This sort of business fluctuation runs all through our
daily lives. There is a violent fluctuation, for instance, in
the harvest of strawberries at different times during the
year. Should we grow enough strawberries in greenhouses so as to stabilize that part of our economy throughout the year. 

We may, therefore, expect specific business fluctuations all the time. There is no need for any special “cycle theory” to account for them. They are simply the results of changes in economic data and are fully explained by economic theory. Many economists, however, attribute general business depression to “weaknesses” caused by a “depression in building” or a “farm depression.” But declines in specific industries can never ignite a general depression. Shifts in data will cause increases in activity in one field, declines in another. There is nothing here to account for a general business depression—a phenomenon of the true “business cycle.” Suppose, for example, that a shift in consumer tastes, and technologies, causes a shift in demand from farm products to other goods. It is pointless to say, as many people do, that a farm depression will ignite a general depression, because farmers will buy less goods, the people in industries selling to farmers will buy less, etc. This ignores the fact that people producing the other goods now favored by consumers will prosper; their demands will increase.

The problem of the business cycle is one of general boom and depression; it is not a problem of exploring specific industries and wondering what factors make each one of them relatively prosperous or depressed...

The explanation of depressions, then, will not be found by referring to specific or even general business fluctuations per se. The main problem that a theory of depression must explain is: why is there a sudden general cluster of business errors? This is the first question for any cycle theory. Business activity moves along nicely with most business firms making handsome profits.

Suddenly, without warning, conditions change and the bulk of business firms
are experiencing losses; they are suddenly revealed to have made grievous errors in forecasting. 

So Roubini gets things wrong, but that is to be expected because he is not an Austrian school economist. But he is at least talking about a macro event.

On the other hand, it is shocking to see "Austrians," now writing of "classic free market driven depressions" when discussing a specific industry downturn. Rothbard makes clear in his important essay on Austrian school business cycle theory that such industry-specific downturns should not be confused with recessions. They are not recessions and they don't launch business cycle activity.

-RW