Arnold Kling  

John Cochrane Muses on Monetary Theory

Forum Today on Financial Crisi... Brad DeLong and Kevin Murphy, ...

John Cochrane has posted an essay on his take on the economy, stimulus, and so on. I think he gets muddled in several places.

He starts out with this.

Every dollar of increased government spending must correspond to one less dollar of private spending. Jobs created by stimulus spending are offset by jobs lost from the decline in private spending. We can build roads instead of factories, but fiscal stimulus can't help us to build more of both.[1] This is just accounting, and does not need a complex argument about "crowding out."

The footnote is to Fama. But Fama is not right.

But then he starts to say some things that make more sense.

People are trying to shift their portfolios out of stocks and especially out of anything with a whiff of credit risk, and into cash or Treasuries. We see this desire in the dramatically high interest rates, and correspondingly low prices, for any debt - jumbo mortgages, corporate bonds, municipal bonds, securitized debt -- that has even a small chance of default and no government guarantee, and the low interest rates and very high prices for government bonds and government-guaranteed debt. People are also trying (finally) to save more, but they want to do so in the form of safe, government debt or government guaranteed debt.


If we just had a credit crunch, we would expect to see stagflation - lower quantities sold, but upward pressure on prices. A credit crunch, like a broken refinery is a "supply shock." Since we are seeing lower quantities sold and easing inflation, we must also be seeing a "demand shock," and we need to understand its source.

I cannot evaluate the above statement without knowing the definition of a credit crunch. In my mind (see lecture no. 10), a credit crunch means that butcher cannot borrow to buy cake, the baker cannot borrow to buy candles, and the candlestick maker cannot borrow to buy meat, so you get deflation, not stagflation.

He goes on,

The bottom line, then, is that people want to hold more of both money and government debt - and don't particularly care which. Trying to get it, we are trying to buy less of both consumption and investment goods. Again, this is a deflationary pressure.

At this point, I generate an internal warning light that blinks about stocks and flows. Consumption and investment are flows, meaning that they are purchases per unit of time. Money and debt are stocks, meaning that they are outstanding quantities. It's not a good idea to add flows to stocks, or vice-versa. In the national income accounts, we say that (with no government or foreign sector) income equals consumption plus saving. We don't say that income equals consumption plus saving plus money plus debt.

I'm ok with a Tobin story, in which a shift in portfolio preference lowers the market price of capital below the replacement cost of capital, causing investment to plummet. I'm ok with an Austrian story, in which a shift in demand from more roundabout production to less roundabout production causes the economy to undergo a major adjustment, which in the short run is contractionary. I am not ok with talking about shifting from consumption and investment into money. That is what I mean by muddled. Let's be generous and assume that he is telling the Tobin story, but being careless.

Finally, he writes,

Fiscal stimulus can be great politics, at least in the short run. The beneficiaries of government largesse know who wrote them a check. The businesses and consumers who end up getting less credit, and the businesses that can't sell them products, can only blame "the crisis," and call up their congressmen to get their own stimulus.

This is important to keep in mind. Whatever the effect the policy has on total spending and employment, it will make people come to government to support whatever economic activity they want to undertake.

Comments and Sharing

COMMENTS (4 to date)
Bill Woolsey writes:

Suppose I hold $1000 of moeny I want to hold $1100 of money. My income is $45,000 per year. I spend $45,000 on consumer goods each year. However, this year, I spend $44,900 on consumer. Over the course of the year, my money holdings will increase from $1000 to $1100. I have adjusted my stock to meet my demand by reducing my "outflow."

I have moved from consumption to money.

Of course, in aggregate, it is impossible for everyone to increase their money holdings by spending less.

But that doesn't mean that a shortage in the stock of money can't disrupt the flow of income and expenditure.

those who were selling me consumer goods, are selling me less. If they continue to spend money at their same rate, their money holdings will drop. They need to reduce their spending as well.

MV = Py.

M is a stock. Py is a flow. You really think this is impossible because of stocks and flows?

V = 1/k k of the fraction of income held as money balances. Perhaps this is wrong, but is it incoherent for the demand for a stock to be a function of the size of a flow?

Philippe B. writes:

Indeed, Cochrane is not mixing flows and stocks, he is mixing flows and changes in stock.

Furthermore he considers what you put in the first citation as a fallacie and he explains it later (section "A monetary argument for fiscal stimulus.."). You can't blame him on that.

I consider this as a really great article, even one minor detail is ommitted : what if US government can borrow a large part of its debt increase from other countries? (that is, if people are not ricardian and if they want to keep constant their money stock, a worldwide fiscal stimulus will not work but a US fiscal stimulus will theoretically work if the rest of world agrees to increase its net total lending toward the US).

Bill Woolsey writes:

Cochrane's article has some good insight.

I am troubled by his claim that "inflationary" finance of the fiscal stimulous must result in higher prices and the new money can't be pulled out of cirulation in the future or else it will be ineffective. I took that to mean that it will only motivate an increase in spending now if people expect permanently higher prices in the

This appears contradicted by what he wrote a bit later about what happens when there is an increase in money demand. If people want to hold more months of income in the form of money (higher k, exactly) then spending and output falls until the price level falls enough so that the current money stock is two months of income.

Well, suppose the nominal money stock rises until it is two months of income at current prices? The price level is higher than it would have been, (not falling) so I suppose you can say that people have a reason to maintain spending relative to the fall that would have occured because of the deflation that didin't happen. But it seems strained to me.

Further, if k rises and then falls, I don't see how reducing the money supply when k falls makes the increase when k rises ineffective.

It seems to me that Cochrane has trouble getting out of the market clearing always mindset. If k has risen, then the price level must already be lower. We must be somehow trying to disturb this equilibrium that already has real aggregate demand equal to producive capacity.

I think conflating money and government bonds is a mistake. Individual money holdings can be incerased by spending less and reduced by spending more. That the government issues base money is not essential to this characteristic. It would still be true if gold were used or private money. Government bonds don't have this charateristic.

Now, his point about the current situation is almost certainly correct. Open market operations that reduce the supply of T-bills remaining on the market will cause an increase in money demand. If the interest rate is at zero and can't fall, the reduced supply of T-bills leaves a shortage. What will people do with the money they would have spent on bonds? They hold it, because money is pretty much a prefect substitute for zero interest bonds as a store of wealth.

While the operation increases the supply of money, the increased demand and supply match and so there is no correction of the monetary disquilibrium. (It is easy.. k rises more as the supply of T-bills shrinks.)

Corchrane's point about the Fed purchasing other sorts of assets is correct. But calling it all government bonds is confusing--I think.

Personally, I think the Fed should buy up all of the near zero interest government bonds (and sure as hell should stop paying interest on reserves.) It should just realize that marginal increases in the money supply will not lead to marginal increases in nominal expenditures under such conditions. They need to buy longer term and higher risk assets. But still, I would suggest buying up all the near zero stuff more or less at once and then going forward from there.

I also agree with Cochrane's point about a credit cruch reducing aggregate supply. If spending is maintained, but finance breaks down, then the productivity generated by an effective financial system disappears. Productive capacity is reduced.

In reality, credit crunches nearly always impact aggregate demand and I think this is the most important impact. But, they do so because they impact either the supply or demand for money. While the borrowers don't spend in the crunch the lenders still have the money. What do they do with it? If one imagines that they spend it rather than hold it, so there is no monetary disequilibrium, what is left? Those who would have lent spend rather than lend. Finance is no longer moving funds about in value enhancing ways. Productivity suffers.

Kling's butcher, baker and candlestick maker might is a bit different. They were making payments without money and using trade credit. If sellers will not sell to willing buyers because they no longer trust them--there is actually demand (the buyers want to buy.) The
sellers don't have money that they are refraining
from lending and we must ask, what do they do with
it. They just don't sell. The buyers need money to make the payment now.

Realistically, when trade credit contracts, the
retailer willing to buy but only on credit has no
goods to sell to his clients. They had money. What do they do with it? They show up at the retailers shop to buy, and there are no goods.

Shortages. If it werent for this disruption, the retailer would have collected the money from them and used it to pay the wholeseller later.

But that doesn't happen because there are no goods. What do the frustrated buyers do with the money?

There is a problem with goods supply. Goods are not moving.

But if we imagine using trade credit and offsetting debts or something, it is different.
IF that stops, there is an increase in the demand for money. The supply of money must increase. For this sort of analysis, it doesn't matter where. If excess supplies of money are created elsewhere, spending may occur there. But then, we see the problem is with supply. These goods aren't moving because those who were borrowing to purchase can't borrow. The lenders (perhaps somone other than the sellers who won't provide trade credit,) are spending the money rather than lending it to these credit starved buyers.

Finally, I find it hard to believe that Cochrane expects us to take seriously his models of fiat currency being like shares of stock in the government enterprise. Maybe there is some insight to be found from models that treat the demand for fiat currency like holding stock in the government. But, other people are supposed to take this idea and imagine it relates to the real world and current problems?

Perhaps part of the problem we face in macroeconomics today is that a substantial part of the "macro" wing of free market economists really think that new classical macroeconomics is "true" because simple and formalistically complete models fit their notion of what is scientific.

It is like Ricardian equivalence. Because the model people (person) rationally saves to pay future taxes, we are supposed to assume this has a connection to reality?

I am sure that the fiscal stimulous package implies higher future taxes. But when I look at my check account balance and make spending decisions-- this has no impact on me. Who are the real people saving up to pay the higher taxes?

Then we have the wing of microeconomists who don't remember Macro.

Then we have the Rothbardians--in praise of deflation and everything must always be liquidiation of malinvestments caused by the market interest rate having been below the natural interst rate. Mises' one scenario is the round peg that must fit every hole, no matter how square.

No wonder those with Sweden-envy carry the day.

Robert Waldmann writes:

As in the post on Fama to which you link, your willingness to criticize Cochrane who agrees with you is very admirable. Such intellectual rigor and integrity isn't as common as it should be among economists (I'm an economist). You almost give me hope for the profession.

Also excellent comments Bill Woolsey. I think you diagnosed the problem with fresh water macro perfectly. I'm going to steal your explanation.

Now there are free market economists who aren't blinded to reality because of their mathophilia and physics envy. Greg Mankiw, for example, is very very polite when discussing Fama and Cochrane (I don't approve as I wrote above) but he doesn't apprach macro the way they do. Robert Hall is a right wing Keynesian and very smart. On right wing, he loved the Reagan tax cuts and advocated a flat tax. Martin Feldstein never got over being CEA chair, but he's smart, in touch with reality and conservative. And you know you can't expect a Milton Friedman to be born every single century.

By the way I personally am a Sweden envier.

Comments for this entry have been closed
Return to top