Arnold Kling  

Still Jackasses

Say It Loud, Say It Proud... Help Me Understand...

Stephen Williamson writes,

From a New Monetarist point of view, a key element of the financial crisis relates to the scarcity of liquid assets. There is one type of liquid asset, which is outside money. Currency and bank reserves play their own unique roles as media of exchange in retail and large-scale financial transactions. A second important set of liquid assets are (in the US) Treasury securities...The financial crisis also increased the scarcity of the second type of liquidity. For example, some mortgage backed securities, which had been widely traded in financial markets, and had served as collateral in various credit arrangements, dropped in value and were no longer traded...The key point is that an important phenomenon in a financial crisis is a shortage of "type 2" liquid assets, reflected in a low real interest rate, rather than a real interest rate that is too high, as in New Keynesian theory.

Pointer from Mark Thoma.
Scott Sumner writes,

If the Fed arbitrarily exchanges dimes for nickels it will not be inconsequential, as Kling seems to suggest. There are two possible outcomes, either the value of nickels will rise above par value, or more likely there will be shortage of nickels. People will go to the bank, ask for a roll of nickels, and find that the bank is out of them. Then they will waste time checking out other banks (the equivalent of queuing costs.) Alternatively, banks might charge $3 for a role of nickels that normally costs $2. Either way, Kling's analogy doesn't hold.

Oy. Monetary theory gives me a headache. In 1968, Paul Samuelson wrote an article for the Canadian Journal of Economics called "What Classical and Neoclassical Monetary Theory Really Was." The article drips with scorn. He starts out by talking about a farmer, who, having lost his jackass, tries to find the animal by asking himself, "If I were a jackass, where would I go?" Samuelson uses that farmer as a metaphor for the modern economist trying to understand classical and neoclassical monetary theory.

I am having an equally hard time understanding modern monetary theory. I can only get it if I start from the assumption that The Fed is Very, Very, Important and then proceed to the conclusion that when the Fed goes into the market and exchanges one type of asset for another this is a Very Big Deal. So, if the Fed were to scoop up nickels by exchanging them for dimes, that has to create a dire shortage of nickels, causing all sorts of havoc. Or, if the Fed exchanges Treasury bills for mortgage-backed securities, that does all sorts of magical things.

And that is mainstream, standard thinking. I am the oddball here. Instead of assuming that the Fed is all-powerful and adapting my view of these asset exchanges to fit that assumption, I start from the other direction. How much should these asset swaps matter? I say, to a first approximation, "not much." (Incidentally, I do not limit my view to the case of low nominal interest rates and a "liquidity trap." Even if the interest rate is 5 percent, I do not see the Fed's purchases and sales as mattering very much.) And if we concede that the exchanges that the Fed makes might not matter much, then the inference is that the Fed is not all powerful, and we need to look elsewhere to explain movements in most market interest rates (I will concede that the Fed can move the Fed Funds rate, an overnight interest rate that I don't think matters all that much, in the grand scheme of things), employment, and inflation. I think that conclusion is too unsettling for people to bear.

Comments and Sharing

CATEGORIES: Macroeconomics , Money

COMMENTS (15 to date)
Lord writes:

I tend to agree these don't matter very much, unless and until they create inflation or deflation, then they matter very much as these create and destroy money on their own.

The Unbeliever writes:

Maybe the Fed doesn't matter so much "mechanically", but it's probably uncontroversial to say they have a huge effect on expectations and marginal uncertainty. Not to mention the effect of the individual Fed governor's speeches, comments, and Congressional testimony.

Huxley writes:

Interesting article

Jim writes:

Get a free copy of the article "What Classical and Neoclassical Monetary Theory Really Was." through the EBSCO Business Source Premier database search.

Hennepin County, MN public library provides this search free to card holders.

jsalvatier writes:

It's not so much that the Fed is special; it's that money is special. The Fed can have a large impact on money because they can print it. If you don't understand the monetary perspective on the role of money in the economy then I suggest reading old posts from Sumner's blog and Nick Rowe's old posts on Worthwhile Canadian Initiative.

Philo writes:

"I am having an equally hard time understanding modern monetary theory." Somebody is a jackass; the only question is: Is it you, or the modern monetary theorist?

Money must be important, economically, or else a barter economy would work almost as well as a monetary one. So the Fed's actions in controlling the amount of money in circulation (principally by buying or selling non-money assets) have to be important, at least when they reach a certain volume. Perhaps you think the Fed would have to do a lot more buying or selling than it normally does in order to have a significant economic effect; but why would you think that?

Jeff writes:

jsalvatier is right. The Fed doesn't just exchange one asset for another, it has the unique ability to acquire assets by printing money, and no one can refuse to take Federal Reserve notes in payment for anything.

Imagine that the Fed stops paying interest on excess reserves and buys up all existing Treasury debt. Do you really think this would have no effect on anything? How about if the Congress cuts all taxes to zero and just pays all it's bills with checks drawn on the Fed? And let's leave bales of currency on every street corner, just for fun.

I don't understand how you don't understand that the Fed can have huge effects. In a fiat money system, a determined central bank can always pump up nominal aggregate demand as much as it wants to.

Rebecca Burlingame writes:

What I get from this is that fiat money and the Fed is as psychological as anything else, in nature. Apparently the Fed is all important because the government not only wants it to be, but gets the majority of economists in line with the prevailing dogma.

The primacy of the Fed evolved in large part, because of the unspoken contract that government made with people when it stopped barter everywhere in the process of setting up Social Security. The contract was essentially, if you agree to do most of your transactions with our money, we promise to take care of you in your old age. In a recent book, Ron Paul stated, "We could permit alternative currencies. Those who want out of the monetary system should be protected by law."

Presently this would seemingly never happen. But it could happen if people created alternative currencies in which a certain portion goes back to the system itself, on the people's terms. (This would also create direct democracies) The best focus of what people could give back is in the very areas that governments in all developed countries now increasingly struggle to provide. In other words, people could take back the social contract that is now bankrupting governments. In the meantime, the Fed is supposed to be in control of our economic lives, for what that is worth.

J writes:


Are you arguing that the Fed has not affected the relative price of MBS? If so, I would greatly appreciate if you could explain why; clearly one cannot demonstrate causality, but casual empiricism suggests the Fed has indeed affected relative prices of MBS vs. Treasuries and/or swaps.

Regards, J.

Troy Camplin writes:

I find the note on the nickels a bit bizarre. First, it requires a prettty bizarre obsession with nickels. So long as there is a reasonable replacement -- 5 pennies -- nobody's going to be going around searching for nickels. And nobody's going to pay 50% more for nickels -- unless they have been taken out of circulation, and no more will be used, meaning they are no longer money, but collectors' items. SO then we are no longer talking about money, and monetarism or any other theory of money qua money no longer applies.

endorendil writes:

Stephen, it's not that hard.

Changing nickels for dimes trades one liquid asset for another liquid asset. There are minor practical consequences, but there's no theoretical issue with the change.

Changing MBS for treasuries is not the same thing. As long as we believe that the US government will be willing to raise taxes and cut services rather than default on its debt, treasuries remain very liquid. On the other hand, MBS stopped trading before, still trade with difficulty and may never really recover. Changing between MBS and treasuries is more like exchanging easy-to-forge or short-lived 1000 dollar bills for well-protected, durable 100 dollar bills. It improves the overall liquidity in the economy.

But of course, there's not been an exchange between MBS and treasuries. There's only been a couple of trillion dollars worth of treasuries (extra US government debt) unleashed on the market. That's not going to replace the 7 trillion MBS that were around, not to mention all the derivatives from it (which technically had a face value of around 40 to 50 trillion). It's the loss of liquidity for those tens of trillions worth of financial instruments that is damaging the financial system. We're not even past the point where we figured out solvency of the main participants in this giant money shuffle, but when we will be, we'll still have to deal with a financial system that has a lot less liquid assets to move around - probably less than a quarter of what was there, and a lot of it not really at play but in storage at foreign governments. That permanently reduces profitability and increases overall friction in the economy.

Gregor writes:

So if the Fed rasied the Fed funds rate by 100bps tomorrow this would have little to no impact on the markets or the real economy? Is that really your position?

fundamentalist writes:

Mainstream economists take the Fed too seriously because they take the quantity theory of money too literally, against which Hayek and Mises warned. To paraphrase, it's a serious mistake to ignore the quantity theory of money, but it's just as bad to take it woodenly literal. The Fed is not all-powerful and all-knowing. As Friedman wrote many times, the lags between Fed policy and effect are long and variable. In depressions, the Fed is virtually impotent.

And for the purists, yes that sounds like Keynesian "liquidity trap." Just because Keynes wrong about most things doesn't mean he was wrong about everything. I think most Austrians would agree that something like a liquidity trap does happen at the bottom of depressions. But the liquidity trap doesn't cause the depression; it is an effect of the depression.

azmyth writes:

The higher the real interest rate, the more demand there will be to hold an asset. Bonds typically have higher interest rates than currency, thus increasing the number of bonds and decreasing the amount of low interest currency will be contractionary.

Bonds affect expectations of the future. Bonds are destroyed when they are paid off. A large debt creates the expectation of higher future taxation, whereas large amounts of currency does not. Thus, adding bonds to the economy does not have as much of an inflationary impact as adding currency.

I agree that the Federal Reserve cannot impact long term real interest rates very much and that talk of interest rates is a red herring.

Alan Reynolds writes:

I sense a surplus of theory and shortage of fact.

The Fed "moves the fed funds rate" by buying or (rarely) selling Treasury bills, so the T-bill rate tracks the funds rate very closely. Longer maturities do too, unless the funds rate is perceived to be unsustainable (e.g., below the anticipated inflation rate). Long rates do not move nearly as much as short rates, but they very rarely move in the opposite direction.

The assumption that low interest rates are inherently "stimulative" seems more debatable. Very low nominal interest rates are observed only in stagnant economies - like Japan in recent years.

Some cyclcial events would be very hard to explain if the Fed was impotent -- notably the recession of 1937-38 when reserve requirements were doubled, and the recession of 1981-82 after the Fed raised the fed funds rate from 9% to 19% at the end of 1980.

Comments for this entry have been closed
Return to top