Bryan Caplan  

Question for Recalculationists, Round 2

Book 1 and Book 2 Watch... Calomiris on the Financial Cri...
Arnold answers my question with another question:
[W]hy does nominal GDP suddenly fall by 5 percent?
Let's keep it simple: The public freaks out for no good reason and responds by trying to increase cash balances, and the Fed doesn't accomodate.

Comments and Sharing

COMMENTS (17 to date)
Doc Merlin writes:

1. You misunderstood his question. He is saying that recalculation is why the GDP falls by 5%.

2. "freaks out for no good reason" is a really bad normative statement to make as an economist. People's value for currency increases and their value of debt, goods and services decreased. The costs associated with a large scale shift in preferences is what recalculation is.

Arnold Kling writes:

In this example, I would have to say that a standard monetarist-Keynesian synthesis gives the correct answer.

Back in the real world, the "freak out" consisted of a big increase in the demand for T-bills, not currency. People no longer wanted to invest indirectly in risky mortgages. They sold bank shares and tried to escape from mortgage securities in order to get into T-bills.

Maybe that all works just like ordinary Keynesian liquidity preference, with T-bills playing the role of the money supply. But if it does, then the Fed swapping currency for T-bills is really an exchange of like assets.

Vasile writes:

Well, if The public freaks out for no good reason then mailing everybody a reassuring check can boost spending almost overnight. With either immediate price inflation or (less likely) shortages. And with a lot of windfall profits for holders of consumer goods' inventories.

Doing it in the usual roundabout way (via banks), will be a lot less effective.

Freaking out is a change in consumer preferences, namely the (subjective) value of present consumption goes down while the (subjective) value of future consumption goes up. Decreasing the marginal cost of present consumption via credit rate manipulation is may not necessarily offset the change in preferences.

Winton Bates writes:

I think it is useful to distinguish between the recalculation in the "real" economy such as the decline in investment in the housing sector and the panic following the financial crisis i.e. sudden changes in demand for cash balances/ liquidity preference/ velocity of circulation.

If V declines and M does not increase sufficiently to compensate, then PY (NGDP) must decline.

Vasile writes:

Back in the real world, the "freak out" consisted of a big increase in the demand for T-bills, not currency. People no longer wanted to invest indirectly in risky mortgages.

I'd say that this makes efforts of our brave reflationists even more surreal. For investors to leave T-bills and rush back into mortgages (a.k.a toxic assets) the return on T-bills not only should be negative, but prohibitively negative.

fundamentalist writes:

Dr. Kling is right. If the public freaks out for no good reason, then there was no recalc problem and monetary pumping will help restore confidence. But isn't it interesting that Caplan had to resort to a fictional scenario in order to help Sumner out?

Devin Finbarr writes:


But if it does, then the Fed swapping currency for T-bills is really an exchange of like assets.

Indeed. This is the real reason why the monetarists are wrong. QE doesn't work because the bank is just swapping one piece of government paper for another. In order to actually spur NGDP/aggregate demand you need to repair people's balance sheets. If private balance sheets do not change, people will not spend more, and there will be no NGDP effect.


Let's modify Brian's example a little bit. Imagine their is a glitch in the software running the financial system. This causes everyone's paper wealth to fall by 20%. I wake up tomorrow morning, log in to Bank America, and find that all my accounts and investments are 20% lower.

What happens? What is the proper response?

Vasile writes:

Devin, I like your question about the glitch in software, even if the answer is obvious in this case. Here the problem is purely nominal so fix the glitch. If that is impossible the prices will adapt to the decreased quantity of money. Yes it will be painful.

But now, I'd like to slightly modify your software glitch scenario. Suppose that every time you buy something 100 cents are transferred to your counterpart account, but only 90 cents are subtracted from your account. It looks good. Everybody's happy. For a while. Than the glitch is discovered. Now what?

Here the problem is not just nominal. The problem we are facing here is general overconsumption during a false prosperity period. Now what? How will you fix that?

fundamentalist writes:

Devin, I didn't notice in Caplan's scenario that balance sheets had been damaged. They suddenly and irrationally wanted to hold more cash and not buy anything. Of course, that never has happened. I don't see why giving them more cash than they want to hold would not spur spending and ngdp.

As for your glitch scenario, we could imagine all kinds of fictitious scenarios all day, but what's the point? Aren't we more interested in reality?

Of course, the Austrian view might be that a sudden, unexplainable drop in the stock of money would require nothing but a drop in prices and the economy would be up and running again. Of course, those in debt would suffer greatly, so a boost in the money supply would do little damage.

Doc Merlin writes:

In Caplan's scenario the answer is obvious, do nothing. People will unpanic. A good example of this was the panic caused by fear of the y2k bug. The bug was real, but was fixed. The panic still happened despite the bug being fixed, but it didn't really cause much problem.... although I know a lot of people who had far too many MRE's and gas stoves afterwards.

Bill Woolsey writes:

Who said that the Fed can only buy T-bills?

Matthew C. writes:

freak out. . . for no good reason

Debt, Brian, Debt!

Devin Finbarr writes:


To make the scenario more realistic, imagine the glitch accidentally doubles the amount of money that people get when taking out loans. Then after a number of years, the glitch mysteriously reveres and destroys the money that was previously created.

During the years of the glitch, investment and consumption funded by loans will have been overproduced. When the glitch reverses itself, there will be a "recalculation" employment affect, as people leave homebuilding and produce goods not funded by debt.

But there will also be unemployment caused simply because of the destruction of money and the corresponding deflation, and fall in aggregate demand.

The solution is fairly obvious to me. The central bank should renominate all balance sheets to restore the money supply to the level before the glitch destroyed the money it created. All money originally created by the glitch, remains created.

But other than that, the central bank needs to step back and allow the recalculation to happen. There will be unemployment due to people moving from building houses to providing healthcare or whatever else. But by renominating the money supply, the bank avoids the catastrophic, unnecessary cross-industry unemployment caused by deflation.


I added the bit about the balance sheets. The paragraph above to Vasile is basically exactly what happened. There was a glitch in the design of the financial system that first created excess cash for people taking out loans, and then uncreated the excess cash.

If either of you is interested, I discussed this issue fare more in depth on another forum:

fundamentalist writes:

Brian: "Suppose that the modern U.S. economy faced no out-of-the-ordinary recalculation problems. By what percentage do real GDP and employment fall if nominal GDP unexpectedly declines by 5%?"

Brian: “Let's keep it simple: The public freaks out for no good reason and responds by trying to increase cash balances, and the Fed doesn't accomodate.”

Given that consumer spending is about 70% of gdp, I would guess 70% of 5%, or a 3.5% in real gdp.

CJ Smith writes:

Freaks out for no good reason? What market have you been in for the last five years, Bryan?

I did title work and worked for one of the largest title insurance underwriters in Florida during this period, and everyone was irrational in the beginning. Lenders were offering 125% loan to value mortgages, no docs, to people who didn't have a credit score high enough to rent an apartment. Borrowers were signing up for mortgages with a 1-2% interest rate that would reset in year 2 or 3 to quarterly variable rates that ended up being 12% or higher, because they were going to flip the property before the reset. Title insurance companies ignored the potential future title problems in favor of getting the biggest slice of the pie. Investment firms packaged and sold tranches of these CDOs at Aaa ratings, and the investing public couldn't get enough of them. Three years later, the first round of resets and defaults started. People began to realize that, in the real world, not your irrational world, that housing inventory levels were so overblown that the market would take 5+ years to absorb all the new inventory, if at all. Here in Orlando 5 years ago, you couldn't swing a stick without hitting a new development; today you can't find a new development and many started developments halted barely past the road layout stage. The Aaa tranches were downgraded to Ba1, Ba2, Ba3 or worse, because irrationally exuberant mortgages were found to be significant portions of otherwise investment grade tranches. Worse, the echo effects began to effect otherwise solid firms, like construction, banking, mortgage and financing companies.

Investors and banks hoarded money because they couldn't accurately assess their default exposure on both the CDO, and needed time and a settling of the market prior to exposing themselves to risk again. Imagine being wheeled into surgery and being told that a recent staph outbreak at the hospital meant your original mortality risk had jumped, but no one could estimate how high. Wouldn't you want to wait until the staph issue was resolved?

No, the market hasn't freaked out. It has gotten a slap in the face, forcing it to face reality - Arnold's recalculation. The "freak out" occurred five or more years ago, under Greenspan, who though irrational exuberance was mere "froth in the markets."

CJ Smith writes:

I followed the link to your prior post Bryan.

"... is consistent with EMH." Do you really think that banks and investors had all the information and irrationally ignored the information? Or is it more possible that some PhD in Finance, Economics or Accounting developed a theoretical model and sold it to financial firms who chose to use the models as justification for extremely profitable short term lending programs to the greedy and ignorant? Seems like a rational short term decision on the part of lenders - just one that didn't accurately predict long term consequences versus short term rewards. Misconceptions, errors in predictions, and differnt weighting of short term versus long term don't necessarily equate to irrational behavior. And given the number of titles and mortgages I closed on market participants who didn't bother to read their mortgages, much less understand the possible consequences, I think EMH isn't nearly as fool-proof as we might like (because fools are so darn ingenious when it comes to doing stupid things...).

Vasile writes:


I see what are you trying to say and I agree that monetary disbalances had to be rebalanced. And yes increasing the quantity of fiat money is cheap. Just a few mouse clicks.

But that's the nominal part of the problem, and there's a real component to it.

Let's take the contrived but simple example of Robinson Crusoe, who due to some mental glitch in his one-man accounting of fish and cocos had, for a period of time, over-indulged in leisure time and fish eating/cocos drinking until, on day D, the sad truth was revealed.

To contrive the example even more, that Robinson Crusoe dude can "renominate" his mental accounting unit of wealth in such a way as to compensate for the losses, but that's just mental... Well something mental.

In order to compensate for the over-consumption in previous period Robinson will have to reduce his current consumption, and reduce it bellow the "normal" levels until his fish/cocos reserves will restore to a more satisfactory levels.

That being said, I'm reluctant to support the reflationist argument for a rather simple reason. The reduction in the quantity of money, will reduce current consumption in a brutal, but effective way. And that's one of the necessary conditions for recovery.

As for historical examples, there are (obviously!) all the banking panics before the Great Depression, and more specifically the 1919-1921 panic.

And to a lesser degree Chilean banking troubles in 1982-83. Where the government ultimately intervened but reluctantly and not exactly giving free money to those who have lost them.

And there's the example of Japan who during its post WWI depression and more lately in the eighties dis just the opposite.

Also you seem to consider money which to consider money created due to the existing glitch in the current financial real, while I consider them imaginary, a phantom, a Fata Morgana.

Comments for this entry have been closed
Return to top