Why does Austrian business cycle theory predict a spike in unemployment when boom turns to bust, but not during the boom itself? Because the former involves a demand shock and the latter doesn’t.
ABCT predicts that fiduciary overexpansion will cause a distortion of relative prices that will bias capital intensive changes to the structure of production. The changes are price, or profit, driven, and these profits are created by fiduciary overexpansion. When credit growth slows or ceases, these profits begin to dry up and non-random malinvestment is revealed. What is malinvestment? Imagine a capital structure prior to the boom, which we’ll call ‘A,’ and a new capital structure at the height of the boom, ‘B.’ This A→B transition represents a change in the type of capital goods produced, because changes in the distribution of profits also change the imputed values attached to each good. When the source of “false profits” (profits caused by credit expansion) dries, the various capital goods that gained value from these profits lose that value; this is a capital consumption shock, where a large chunk of the capital stock loses value, because of a demand shock.
To make the point clearer, let’s think about how something similar could occur if it were to be completely consumption-driven. That is, suppose we’re at structure of production ‘A’ — no boom —, where the values of the means of production are imputed from the demand for their final product (the consumer good). Suppose there is a sudden shock to consumption, because there is a dramatic society-wide change in time preference, in favor of an immediate doubling of savings. Let’s assume that this shock is large enough to make a noticeable downward impact on the prices of the different consumers’ goods. The prices of the relevant means of production will also change, because this scenario presents an immediate halving of revenue for the consumer good industry. The structure of production will have to change (becoming more capital intensive), but since these changes take time there will be a period of heightened unemployment. In any case, note that this is also a case of a capital consumption shock, since those capital goods once valued for their final product are no longer as highly valued (because demand for their final product falls).
There may be secondary features that differentiate the two cases (a credit induced boom–bust v. a dramatic change in time preference). For example, our hypothetical “consumption shock” will probably not lead to significant “debt deflation,” where bad loans are revealed and liquidated. The financial malinvestment that characterizes real world business cycles is a byproduct of the fact that banks, or financial intermediaries, are key players in the extension of credit — they finance the malinvestment. There may be some monetary disequilibrium, due to any increase in uncertainty, but this wouldn’t be the primary cause (or perhaps not even a major cause) of the unemployment spike. But, the two cases are similar where it matters: the relationship between imputed profits/prices, changes in these, and capital consumption.
Of course, these kinds of wild swings in consumption patterns aren’t likely to occur in the real world (it would be a black swan event). I only bring it up to show the relevance of the demand shock to the business cycle. Changes to a more capital intensive structure of production are gradual; the Austrian business cycle, at least where boom turns to bust, is not a theory of gradual change, but of the need for immediate transition — to avoid the spike in unemployment the capital structure has to immediately change. Because immediate change is unrealistic, we have a temporary period of increased idleness.
This discussion, by the way, reminds me of a paper by Hoffmann and Urbansky, “Order, Displacements and Recurring Financial Crises.” It’s been over a year since I read it, and I vaguely remember some areas of disagreement, but I think it more-or-less captures what I write here and presents it in the more sophisticated framework of order.