Arnold Kling  

Is Zero a Lower Bound for Interest Rates?

Incentive Ceiling... Morning Commentary...

Suppose that the Federal Funds rate falls to zero. Does that mean that we are in the infamous liquidity trap, in which the Fed is powerless to use open market operations to affect the economy?

I think not, and I'm pretty sure that only Paul Krugman or someone too awed by Krugman to think for himself would suggest otherwise.

The Fed Funds rate competes with the rate on short-term Treasuries. The rate on short-term Treasuries these days is being determined by their value as collateral. Every financial institution is saying "asset-backed securities bad collateral, Treasuries good collateral." So the demand for Treasuries is through the roof, and when the price of a bond is high, its interest rate is low. So you get ridiculously low short-term interest rates on Treasuries, and also on Fed Funds.

If this safe-collateral mania keeps up, zero might not be the lower bound. Somebody might be so desperate to put up a Treasury security as collateral that the would be willing to pay interest to the Treasury for the privilege. That would mean a negative interest rate.

Now, I don't really believe that's very likely. The main reason I don't believe in the liquidity trap is that I recognize that there are multiple interest rates. If it gets to the point where the Fed cannot trade money for Treasuries, then it can trade money for Freddie Mac debt, which is still trading at a high risk premium over Treasuries.

So if anybody tries telling you that we are in a liquidity trap, tell him to shut his trap.

I don't think it's as complicated as Greg Mankiw makes it sound.

Also, I think that the spread between inflation-indexed Treasuries and regular Treasuries right now reflects the collateral squeeze more than it reflects expected deflation.

Comments and Sharing

COMMENTS (12 to date)
guyson writes:

Arnold, why would anyone ever lend at a negative interest rate? At that point, wouldn't it make a lot more sense to just use cash as collateral, which can always be held at a 0% nominal rate?

El Presidente writes:

Think for a moment about holding real assets instead of instruments (either bonds or cash). If financial markets cannot hold value because the Fed is competing for control of the money supply, then individuals might be tempted to make a further flight to quality to brace for expected inflation; the Fed's last resort.

I know, I know. I'm supposed to shut my trap. This is a problem of definitions though. Yes, the classical definition of a liquidity trap is subject to evasion by unique circumstances, but the concept, the inability of the Fed to move real interest rates and stimulate AD because of consolidation of private lending power and insufficient incentive to lend, remains extremely relevant to our present condition. I am not a member of any Krugman cult. I am simply looking at the landscape and asking myself one question: "Does our collective incentive to grow AD through lending outweigh our individual incentives to protect profits (or minimize losses) through consolidation?" At the national level, the answer is probably a qualified "yes" based on expectations of changes in the tax code and government spending. But, are the incentives aligned for individual lenders and firms? If lenders are not inclined to lend in sufficient amounts to restore employment without strong inflation, we can call it a trap or whatever else we prefer; we are still stuck between a rock and a hard place, especially since that loathed inflation will impact the purchasing power of pensions and retirement accounts. Consumers will not restore spending until they have a sense of security, for which they presently have no logical basis. There is no absence of money; there's plenty out there. It just isn't flowing well enough to touch all the bases before heading back to home plate. We're playing cricket instead of baseball. If you don't want to call that a liquidity trap, fine. What else should we call it, a cricket match?

I get the sense that we will refuse to recognize the danger until we have have been snared. Then we'll probably work our way through denial, anger, bargaining, . . . Why is it that we won't prevent the grief instead of tempting fate to visit misery upon us; are we masochists? Oh, that's right: we're libertarians, so the market is never wrong. Here comes the pain.

NCSU_Student writes:


What you're saying is that when interest rates hit 0, traditional monetary policy loses its effectiveness and we have to find alternative methods.

Sure, that sounds fine. But do you really WANT the Fed throwing money at other assets? "Hey! Let's have the Fed buy mortgage-backed securities! This certainly wont have a significant impact on the allocation of resources, will it?"

See my point? The reason a situation like this is worrisome is because there isn't any good solutions.

Brian Shelley writes:

Could it be a currency bubble? Where people earn positive rates of return by holding cash? Speculators could sell physical assets to get cash, contributing to even more deflation and increasing the rate of return on holding cash?

Bill Woolsey writes:

Why isn't currency (Federal Reserve notes) adequate collateral?

The interest rate on currency is zero.

Assets with negative nominal interest rates can be sold for money. Deposit money can be redeemed for currency, which will have a better yeild than these negative-yeild assets we are supposing.

Of course, currency must be stored. If the issuer offers whatever high denominations people want, that isn't a problem.

And other financial instruments may some cost of storage too.

But, roughly, it is the cost of storing zero-interest currency that puts a lower bound on the nominal interest rate.

Think about a cashless payments system (as Black.)
No liquidity trap there. Interest rates can be as negative as you want.

Also, think about currency suspensions. What is used to quote prices? Deposit money (where deposits have negative interest rates) with currency trading at an explicit varible premium?

I think so.

Gu Si Fang writes:

Where can I find a correct definition of a liquidity trap? Thanks

Lord writes:

I would view this more as a question of will than way, but the Fed has been very willing of late.

Barkley Rosserr writes:

While they have not lasted over sustained periods of time, we have in fact seen negative nominal interest rates. It happened for Treasuries during the day of the ultra-meltdown, September 17, when all hell was breaking loose. It also happened several times in Japan in the late 1990s. I have it also from a primary source that on the last day of the old tax code, December 31, 1986, the actual federal funds rate oscillated between about negative one half percent and 18 percent.

So, not only can it happen, it has.

D. F. Linton writes:

Couldn't the federal government always just:

Cancel all taxes for one year.

Order the Fed to credit the equivalent of one year's tax revenues to the appropriate government accounts.

Spend as usual.

MattYoung writes:

There should be some negative interest rates if a competitive monetary system had some bankers falling out and others falling in.

In a fiat system, we use inflation to get negative interest rates.

But negative interest rates, or investment failures, need to happen once in a while to set the risk premiums. We don;t know the price of risk unless risk happens now and then.

Patrik writes:

Gu Si Fang,

you can find an excellent overview about liquidity at

Nathan Smith writes:

I second guyson's question.

Also a distinction between two claims:
(a) Liquidity traps are impossible regardless of Fed policy.
(b) Liquidity traps can always be prevented by monetary policy.

My casual opinion is that (a) is false, because cash sets a lower bound on nominal interest rates. If the Fed allows steep deflation, AD will collapse as people try to hoard scarce cash.

My casual opinion is that (b) is true, because the Fed can always "print money" to prevent deflation, subject of course to the usual lags.

I think Krugman is denying (b), because he wants to force a fiscal stimulus in order to engineer a government that eats more of the economy. I don't think his arguments make sense:

I think Arnold is accepting not only (b), but (a). I find that confusing, as well.

Comments for this entry have been closed
Return to top