Arnold Kling  

Why Are Interest Rates Low?

The First Amendment and the Pr... Matching Narrative to Policy...

Raghu Rajan writes,

My sense is that we do not know enough about the effect of ultra-low interest rates to state categorically that they are an unmitigated good for reviving the economy. But perhaps the most important cost of low rates is its effect on risk taking and illiquidity seeking. Remember that the United States Fed under Greenspan helped precipitate the recent crisis by keeping rates too low too long. That suggests we cannot be sanguine about the risks that are being taken now. Indeed, many of those who urged Greenspan to keep rates ultra low then are urging the Fed to keep rates ultra low now.

Read the whole thing.

My sense is that Scott Sumner believes that interest rates are low because expected nominal GDP growth is low. Therefore, he would argue that monetary policy should be more expansionary.

My sense is that Krugman and DeLong believe that interest rates are low on U.S. Treasuries because there is a lot of demand for safe assets. Therefore, the U.S. should create more safe assets by running larger deficits.

Rajan is implicitly suggesting that interest rates are low because the Fed is already pursuing a very expansionary monetary policy and is forcing rates to be unnaturally low.

Of the three explanations, I find Rajan's the least plausible. That is because I come to this discussion with a belief that the Fed is much less of a factor than just about anyone else presumes. I believe in a Fischer Black view that capital markets are rich and complex, and interest rates are set by these markets. The Fed at most affects relative supplies in a tiny segment of these markets. In this view, monetary policy does not matter much. So from a policy perspective, I do not have a horse in this particular race.

I lean to a Sumnerian explanation for the recent drop in long-term rates. Economic growth and employment have been weaker than many of us expected, so long rates have come down.

I do not think that we have seen a dramatic increase in the spreads between corporate and government debt in the last month or so (I do not have time to check this, so I could be wrong). If the increase in risk spreads has been modest, then the DeLong and Krugman view has only modest plausibility.

My view of interest rates in the past two years is that markets have been trying to tell the financial sector to shrink. Investors do not trust banks. They would like to invest directly in stocks, government debt, and non-bank corporate debt. Government policy has been an all-out effort to help banks and to fight against the market. These efforts include the Fed's balance sheet moves, TARP, and the recent actions of the Eurocrats.

Sumner, Krugman, DeLong, and Rajan are acting as if macroeconomic stability is the main policy issue. My view is that the policy makers in fact are focused on bank stability first and foremost. Policy makers tell us, and tell themselves, that the key to macroeconomic stability is bank stability. That is one story. My story is that they are less able and less motivated to bring down the unemployment rate than they are to prop up banks.

Comments and Sharing

CATEGORIES: Monetary Policy

COMMENTS (17 to date)
Brian Clendinen writes:

Not sure if I have been living in a remote region in the land of economist when it comes to the Central Bank interest rate theories. This is the first time I actually have heard someone agree with my position on the Fed and interest rates other than my dad.

When I was introduced to the idea at 17 in a college Macroeconomics class I found the idea a little dubious. Over ten years later I find it just out right silly to assume the Fed has even a portion of control over interest rates thru interest rate targets as your average economist seems to believe. I can understand why the Keynesians to Communist side believe the position but I do not get why so many other schools seems to believe it also. I mean look at the Greenspan Fed before 9/11, it pretty much just followed the market rates with the bench mark. If I remember correctly, the Volker fed did pretty much the same thing.

As far as I am concerned unless the rates go wacky due to a drastic reduction of volume(like they did at the beginning of the Financial Crisis) that should be the feds policy. Then again, unless there is strong good evidence my defacto position is always to assume the government can’t control it.

Rebecca Burlingame writes:

I envy anyone who could figure this out at seventeen! I did not get serious about understanding and working with economics until after many years of trying to remain either employed or self employed.

Richard A. writes:

Over the decades, I have observed central banks crash their economies in order to preserve the fixed exchanged rate.

By allowing or causing nominal GDP growth rate to decline will cause the inflation premium on interests to be lower. Those who have lent long and borrowed short will make out like bandits. This of coarse will destabilize the economy.

azmyth writes:

I have come to believe that the main way the Fed affects interest rates is through expected inflation in the Fisher equation. Sure they can affect the fed funds rate, because the market is small relative to the money injected, but the Fed couldn't use expansionary policy to reduce the 5 year treasury rate or anything like that. Sumner, Krugman and DeLong's explanations are all consistent with this as the Krugman story is demand side. Deirdre McCloskey wrote a piece on this awhile back:

Tom Grey writes:

I think there are few differences between 0.5 and 1.5% Fed interest rates, and also that when you or Raghu or Scott (or Brad or Paul) talk about "low" interest rates, a rate or range is specified.

Raghu's key insight is the punishment of the little saver. When rates are below 3% or so, it's silly to expect small savers to save much. And, since for 30+ years, such savers were "saving" in their houses, low interest rates won't be tempting them to save more in banks. Of course, the macro economists claim to want to increase current AD, so they don't want low-wealth folk to be saving, they want everybody to spending.

Raghu's point about low rates helping banks (hurting savers) is very stong.

The monetarist equation PQ = MV can be modified for housing (h) to be Ph*Qh=Mh*Vh. We know the Prices of houses actually sold last year, and the total Quantity of houses built and owned/paying taxes (mortgages). ("we know" doesn't mean I know tho!)

The usual case has the value of V be whatever it takes to make the equation. In the case of housing. Vh is some function of the %total houses sold in the prior year; fewer house sales means lower Vh. For housing 2000-2006, the Mh became the driving variable, and it was the fact that so many investor/speculators wanted to get the higher interest rate by investing in housing that led to Mh being so high and thus the Price to lead an increase in house construction until there was overbuilding. (I'm sure most speculators thought the "worst case" would be 0 growth, not price drops. )

Anyway, Mh is not the same as M, and until the housing market is stable / not dropping, there won't be a real recovery.

In a related issue, there was no need for a huge construction worker bailout when house construction tanked in 2006. Lots of workers lost jobs, owners lost companies. Too many builders. Now, there are too many bankers. The Big Banks should have been re-regulated thru the market and the needed gov't -- bankruptcy court.

The market is based on organizations making promises and keeping those promises, or else going bankrupt and letting the gov't judge/process decide how much of which promises are kept. The US doesn't need so many bankers.

Mike Rulle writes:

The Federal Reserve continues to increase its Fannie/Freddie mortgage portfolio. They own $1.3 trillion of their mortgages and holding company paper. This has the odor of monetization of government debt---which itself is $750 billion on Fed balance sheet for a total of $2 tril or so.

We are propping up the mortgage market. In addition, the positive yield curve injects profits into the financial system like a fire hose (analogous to early 90s)---from savers into bank profits---40% of which goes to compensation on top of it all.

One analogy is to physical therapy. A man with a broken hip needs physical therapy---hard work and pain---to walk on his own again. Coddling him in bed means he will never walk again and will continue to be a ward of the system.

The Fed is one of the chief enablers of this process. Just because the patient is not screaming in pain does not mean the actions are good for them. They surely are not good for us.

Boonton writes:

Over ten years later I find it just out right silly to assume the Fed has even a portion of control over interest rates thru interest rate targets as your average economist seems to believe. I can understand why the Keynesians to Communist side believe the position but I do not get why so many other schools seems to believe it also.

I don't get this confusion at all. The Fed is able to control the 3 month rate because it is able to print an infinite amount of money to buy up the necessary T-bills to force the rate down to what it wants. If it wants to raise the rate it can presumably sell as many T-bills as required from its inventory to raise it.

Likewise the Fed can (but usually doesn't) do this with longer duration rates. But for the rate it is not targetting, the market can and does shift the price up and down.

Sure they can affect the fed funds rate, because the market is small relative to the money injected, but the Fed couldn't use expansionary policy to reduce the 5 year treasury rate or anything like that. Sumner, Krugman and DeLong's explanations are all consistent with this as the Krugman story is demand side.

Why not? Are you saying the Fed couldn't just start printing money and buying 5 yr bills until the rate fell to what it wants? Why not?

The only downside to such action is what tools is the Fed left to fight inflation? If inflation kicks up interest rates rise and the price of long term bonds falls. If the Fed wants to drain cash from the economy it may discover it can't sell those 5 year bonds for enough cash to drain what is needed. When you're dealing with 3 month bills, if nothing else, you just wait 3 months and the economy itself has to redeem the bills by providing the cash value of their principle.

azmyth writes:

I suppose I misspoke. I meant to say, "The Fed won't..." not the Fed can't. They could buy 5 year treasurys until the rate fell that low, but they'd have to deal with far higher inflation than they have expressed preference for in the past. The Fed could peg the nominal interest rate that low, but then they'd be the only lender to the Treasury since the real return on those bonds would be negative.

azmyth writes:

Now the flip side to that is the Fed could easily hit a 1% target by pursuing contractionary policy. If they liquidated their portfolio and burned the cash, the ensuing deflation would hit the bond markets so hard they'd be a 0-1% for a long long time.

Boonton writes:

The flip side, though, is that when you liquidate you have to hope the stuff you're selling commands enough cash to meet your goal. Like I said, for 3 month bills this isn't an issue because they won't normally sell at much of a discount and if you just wait 3 months they automatically turn into cash. 5 year bills are a bit different and 30 years even more so. Hence the reluctance for quantitative easing.

Let's also keep in mind that it targeting interest rates is relatively new. In the 70's the Fed was targeting money supply and letting all rates float. This was Milton Friedman's preferred policy from what I understand. But rate targeting came into fashion after velocity became erratic.

Troy Camplin writes:

I think there is quite a bit of evidence that the Fed is directly involved in having driven and in keeping the interest rates artificially low. For one, the Fed sets the interest rate at which banks lend each other money, and that has no small effect on the overall interest rate. Also, the money the Fed lent out to the banks (to the tune of $1 trillion) has also kept interest rates low.

I wonder, though, to what extent high interest rate businesses like credit cards and payday loans (which, bizarrely, nobody seems to count in determing the interest rate) help to keep bank interest rates low.

mulp writes:

I argue interest rates are low because there are no investments. And realistically haven't been since 2001.

Since 2001, banksters pitched real estate as a high rate of return safe play, but now they are shown to be crooks and liars. M&A artists took solid firms and with investor cash stripped them to bankruptcy, so Wall Street showed themselves crooks and liars. CEOs at firms like ENRON and others including Merrill and Bear and their ilk including Goldman showed themselves to be crooks and liars.

While the 90s were filled with new ventures that made money and grew, or at least grew and made money, the sectors that were poised for that at the turn of the century were inhibited by public policy, especially the policy to subsidize old mature dirty energy. That has kept the innovation out of the US where it was pushed by the policies of the 80s. While communities elsewhere are becoming close to being free of oil and cheaper, the US depends on imports to install the technology first developed in the US and then taken to competitiveness elsewhere. Investments take a few years, but the laws to bring parity don't last that long.

But, you say, oil is cheaper. Yeah, because it is heavily subsidized - in the 90s, Gulf leases were written with royalties of zero in attempts to drive drilling in expectations of the looming shortages. The other Gulf subsidy has been two decades of offering far more leases for auction than can be drilled so few leases have more than one bidder. The other subsidy has been lax regulation and royalty enforcement. And US firms shed exploration capacity because the price was too low and haven't been in a rush to invest in new rigs, leaving that to foreign firms like BP, while is the largest lease holder in the Gulf - as profitable as Exxon is, it doesn't see exploration to be worth the investment.

The big tech products of the past decade have come from abroad, with some important product specing, software, and marketing, like Apple or government mandates like DTV to drive them - neither created real jobs in the US.

So, what is worth putting those hard earned dollars flowing into 401Ks and IRAs? Bidding up the NASDAQ again only to see it fall again. Buying real estate securities? Corporate bonds funding LBOs and other M&A asset stripping.

At times like these, US government debt is the safe bet. After all, the Revolution was funded by printing money, but Hamilton insisted the US must make sure its debt was never repudiated, and in never has been in two centuries. History might not be the best predictor, but the last decade has destroyed the perception of safety for all the others but warren buffett, google, and apple, but those stock have already been bid up to the point one wonders if they are too high.

The past decade has been one of malaise n real innovation and in something-for-nothing financial "innovation" which was really just high tech pump and dump.

What innovation is there worth investment, investment that builds something that is productive like 50 years of electricity generation. Like what people bought stocks and bonds in when I was a kid half a century ago.

azmyth writes:

Boonton: I think I agree with you, I just worded my first post poorly. I don't know if interest rate targeting is the best way to go, especially when supply and demand for investments is even more erratic than V. I have been mostly convinced that inflation or NGDP futures would be better.

Troy: They try, but inflation pushes real interest rates back toward the natural rate. Credit card and payday loans are not factored into the interest rate because the reason they are high is because of their extremely high risk. Once the default risk gets to high enough, the interest rate on bank loans really doesn't make much of a difference.

Boonton writes:


Payday loans risky? I'd like to know if that's really the case. Usually payday lenders make you have your check direct deposited with them. Aside from leaving your job the day you get the loan, are there that many ways to default on a payday loan? Is there any actual documentation on the default rate payday lenders actually incur or do we just take them at their word since they tend to prey on desperate people and it 'fits' as a narrative that they would have a high default rate?

Targets: I think the word is probably not the best. A target is something you aim at and want to hit. Maybe 'scope' would be a better word here. In theory inflation and/or NGDP IS the Fed's target. The 3 month T-bill rate, Fed funds rate etc. are the tool the Fed uses to aim at that target.

azmyth writes:

Payday loan default rates are quite high, and because the amounts they are for are small, there are high transactions costs to making them. In a competitive market, excess profits should be competed away. Payday loans suck, but so do mafia loan sharks.

Brian Clendinen writes:


Maybe I should of been more precise, I was referring to real interest rates not absolute. However, I stand slightly corrected, in the short term the Fed can and does influence rates. I was thinking in general and ignored short-term. The case can be made that the Fed can cause some funky yield curves in the short term.

However, the policy is not sustainable over a long period. Just as one major player can temporary effect the price of a stock but in the long term their purchasing behavior can only go so far without a super majority ownership. The fed is no different, take AGI and GM. The government propped up the stock price but can anyone honestly say it will really make the stock sell at a higher price in the long run?

Granted some of it can be off-set thru the purchase of long term securities which is what they have done with Mortgage securities. However, this is just making the long-turn problem worse and will back fire. The Fed is about to loss a lot of money when it sells its portfolio off. Granted it will add liquidity to the market but it is no different the buying AGI and GM. There will be significant real economic loss to U.S. taxpayers.

Now I am just referring to the U.S., if the government was the primary source of supply and demand of capital then obviously they would have a major influence on real interest rates. However, like any other product the concept of the government regulating price is highly dubious, real interest rates are no different.

Elvin writes:

Interest rates are set by the IS-LM curve. So, the interaction between investment opportunities (equilibrium of investors and savers) as well as liquidity preferences of financial markets influence the level of interest rates.

I think that government bond yields in the long run have to be below the nominal growth rate of the economy--otherwise in the long run, the government can't finance its debt. However, in the short run government bond yields can be well below or well above this. For default free (we hope) return, government bond investors will accept a discount to the more volatile return stream that is related to nominal economic growth. Higher risk investors (equity holders) demand a return above the growth rate of the economy.

Sumner is right; the nominal growth rate of the economy in the short run is certainly lower than the 6.0% and higher that we have seen at times in the past. It's probably around 4.0% right now (1.0% inflation + 3% real growth). A 4.3% long bond yield might drop another 50 bps before hitting fair value.

This sort of a world is still tough for debt-holders, as fixed payments are harder to service.

Comments for this entry have been closed
Return to top