Arnold Kling  

Demand Stimulus in a Jones-Minsky Economy

Sentences to Ponder... Spermsmanship...

In a Garett Jones economy, a worker represents an investment. A firm can vary its production without adding or subtracting workers. At the margin, what workers do is provide additional capabilities, what Jones calls organizational capital.

In a Hyman Minsky economy, investment is cyclical. Right after a crash, firms use "hedge finance," meaning that they invest out of profits. As their confidence recovers, they use "speculative finance," meaning that they are willing to borrow to invest, as long as they see believe that the profits from the investment will repay the borrowing. When they become euphoric, they engage in "Ponzi finance," where they borrow because asset prices are being marked up so rapidly that they can repay today's borrowing by borrowing even more tomorrow. When this becomes evidently unsustainable, you get the crash.

We are in the post-crash phase of the Minsky cycle. In a Garett Jones world, a demand stimulus that increases output does not generate a proportionate increase in employment. Hence we read of The Death of Okun's Law (thanks to Mark Thoma for the pointer).

In a Minsky economy, at this stage of the business cycle firms are looking for profits as their source of investment. And in a Jones economy, they look at workers as an investment. Hence, we can see that profits are the key to getting employment to rise.

Next, add--what else?--the Recalculation story. The Recalculation story says that the market is trying to figure out where to reallocate workers. Unless government technocrats are brilliant enough to know which workers belong where, demand stimulus will raise output without raising employment.

The good news is that with output rising faster than employment, profits should increase, and indeed they have. In a Jones-Minsky economy, this will eventually lead to higher employment, as firms invest those profits. Of course, they can sit on those profits for a while.

Putting all this together, I would be willing to bet on the optimistic side of the usual forecasts for employment, which seem to be based on extrapolating the recent past (when employment has been lower than the models forecast) into the future. Instead, I think that firms have restored profitability to the point where they are not going to be as ruthless about laying off workers and they are going to be a bit more relaxed about hiring new workers.

The unemployment rate is so high and the recalculation problem is so severe that it is bound to take many years to bring the unemployment rate down to even as low as 6 percent. But I think that the rate of job growth will soon pick up faster than most forecasters expect.

Finally, my view implies that stimulus does work. However, on a dollars-per-job basis, fiscal stimulus would work better if it were focused on increasing profits sooner. By throwing money at consumers and at state governments, the stimulus is taking a lot longer to have an effect.

Comments and Sharing

CATEGORIES: Macroeconomics

COMMENTS (9 to date)
JPIrving writes:

Do firms still expand employment on profits, when those profits depend on temporary government stimulus? What does recalculation conscious fiscal policy look like?

Philo writes:

You write, in a Minskian vein: "Right after a crash, firms use 'hedge finance', meaning that they invest out of profits. As their confidence recovers, they use 'speculative finance', meaning that they are willing to borrow to invest, as long as they see believe that the profits from the investment will repay the borrowing."

This doesn't make much sense. If, right after the crash, they lack confidence, why do they invest at all--out of profits or otherwise? (And are they making *profits* right after the crash?) If they have enough confidence to invest their own capital, why don't they have enough confidence to operate with "other people's money"--with borrowed funds?

If later--long after the crash--their confidence increases, their willingness to borrow will, of course, increase; but so will their willingness to invest their own capital. Why should one increase more than the other?

I haven't read Minsky; I am relying on you to expound his ideas. I can see that confidence would affect willingness or eagerness to invest--these are practically the same thing. I can't see why it should affect the decision whether to use equity- or debt-financing.

david writes:

Wouldn't this imply that monetary policy would be minimally effective right after a crash? Since firms won't be borrowing to invest anyway?

This doesn't seem to square with past experience.

Stephan writes:

So? You're thinking about a Minsky economy? Then you should have added, that the best stimulus would come from a government acting as a employer of last resort. At least according to Hyman Minsky. But you did not dare to raise that subject.

Norman writes:

I would think, if firms are hiring out of profits, that this is essentially the rationale for recalculation: it's going to take time to identify which firms can make sustained profits. That being the case, wouldn't government stimulus that focused on profits introduce noise, making it harder to identify which firms are going to be profitable and thus should be hiring? I don't see how stimulus could do anything but produce a double-dip recession (or at least drag out the current one) in this view. I'm not sure what you mean, then, by "my view implies that stimulus does work."

Chris Koresko writes:

We also just increased the minimum wage. Should that be expected to make the calculation of how best to reallocate labor more difficult, by rendering more of the previous cycle's results obsolete? Should it therefore make the Recalculation more time-consuming and delay the recovery?

Nathan Smith writes:

So stimulus works because individuals are irrational and don't offset it by saving, and because firms are irrational and channel profits into investment regardless of the impact of growth of government on future profitability, ignoring the possibility of debt financing so that it doesn't matter that government borrowing is soaking up all the capital? Too many different kinds of heterodoxy rolled into one argument, for my taste. Can you model it? Can you test it? At least one or the other would be nice.

Steve Roth writes:

Having owned and run quite a few businesses, I would suggest that business owners are smarter than you think.

When they look at current "profits" and consider investing in growth (workers, equipment, whatever) they will mentally discount any one-time government disbursement portion of those profits, and use their true sales-derived profits to predict the future, and hence to make their investment decisions.

Now you could argue that profits resulting from a wider dispersal would be discounted similarly, but I'm also here to suggest that business owners aren't *that* smart. If they have increased sales, they will be (justifiably) hard-pressed to discern what proportion comes from government stimulus.

And given that we're at the very peak of our thirty-year Reaganomics-driven binge of non-stop Keynesian stimulus (except under Clinton...), they *will* be smart enough to know that even if they can't discern which part of their sales/profits come from stimulus, the odds are pretty good that that stimulus will continue. They can see the projected deficit numbers.

So government donations to corporate profits

1. Are much more easily discernible as a discrete, identifiable portion of those profits,


2. Have a much lower likelihood of being continued and providing ongoing profits.

So helicopter money--which via distributed choice will more naturally flow to sectors where recalculation requires it, in the grand free-market tradition--should do a much better job of reducing unemployment and increasing sustainable corporate profits.

And in the short term, people who are really suffering will have that suffering relieved.

Jeff writes:

[Comment removed for supplying false email address. Email the to request restoring this comment and your comment privileges. A valid email address is required to post comments on EconLog.--Econlib Ed.]

Comments for this entry have been closed
Return to top