Never reason from an interest rate change
I recently caught Carmen Reinhart taking a short-cut, and ending up with erroneous conclusions:
Reinhart is the toast of economic circles these days for speaking out about the newest way Western governments are using financial repression to liquidate their debts, particularly after a financial crisis. They’re doing this on the backs of savers, including pension funds, according to economists. In practice, financial repression can lead to “the rape and plunder of pension funds,” Reinhart tells Institutional Investor. Financial repression consists of very low nominal interest rates combined with captive lending by large banks or pension funds to a government. The low, stable interest rate facilitates the servicing costs of large public debts. Sometimes modest inflation is added to the mix. This results in zero to negative real interest rates that reduce government debt. Hence, broadly defined, financial repression is a wealth transfer from savers to debtors using negative real interest rates “” with the government as one of the key debtors.
I’m sure my readers will immediately spot the error, but just in case here’s Milton Friedman in 1997:
Low interest rates are generally a sign that money has been tight, as in Japan; high interest rates, that money has been easy.
. . .
After the U.S. experience during the Great Depression, and after inflation and rising interest rates in the 1970s and disinflation and falling interest rates in the 1980s, I thought the fallacy of identifying tight money with high interest rates and easy money with low interest rates was dead. Apparently, old fallacies never die.
Some might argue that it doesn’t matter whether the low rates were caused by tight money or easy money, they help debtors either way. But this isn’t true. That’s because the tight money policy of 2008 that led to the low interest rates today also drove NGDP more than 10% below trend. Indeed it would be fair to say rates are low today precisely because NGDP fell so sharply. And this fall in NGDP hurt debtors like the US government much more than the low rates helped them. That’s why the debt to GDP ratio has risen sharply in the US, and most other developed countries.
In fact, very low interest rates (aka “tight money”) hurt both lenders and borrowers. How can both be hurt? Simple, it’s not a zero sum game. The Fed’s policy reduced real income and real wealth, leaving America a poorer place. Both lenders and borrowers have shared in that loss.
Admittedly the debt service costs are temporarily depressed. But when the economy recovers rates will rise again. But meanwhile the debt/GDP ratio will still be much higher than before the recession. In the long run tight money (aka “low interest rates”) makes the government worse off.
PS. I’m now so busy that I won’t be able to answer all comments. I’ll continue answering comments from the most recent 5 posts. Because this is a new policy, I’ll answer older comments one more time on Wednesday, to give commenters one more chance to respond to anything I said. I’ve recently been getting comments from as many as 15 different posts on a given day–and this is just too much work with my teaching responsibilities and my attempt to revise my manuscript. And I’m also doing new posts and keeping up with news in the blogosphere. Plus I’ll be doing a lot of traveling and speaking in the next month.
Tags:
28. February 2012 at 08:32
Regarding your P.S. and the comments you have answered for so long: Now it’s hard for anyone to start an economics blog if in fact they don’t allow comments and what’s more, respond to them! People may not be completely convinced yet about NGDP targeting, but you seem to be making a difference in the way people approach blogs.
28. February 2012 at 09:02
Another excellent post by Scott Sumner.
Side note to Scott: Most popular bloggers only respond to comments on latest post or two.
28. February 2012 at 09:42
Ben,
Good point about the latest post or two. However, for any blogger who sets up a blog detailing a specific agenda that continually evolves over time those still-open previous posts can serve a productive purpose, if the blogger actually has time to maintain them. The setup Scott adopted was ideally suited for that purpose.
28. February 2012 at 09:47
Benjamin Cole,
We get it: you want to have Scott’s babies. Don’t we all. But you’re kind of beginning to freak me out a little.
28. February 2012 at 09:57
ssumner said,
“Low interest rates are generally a sign that money has been tight; high interest rates, that money has been easy.” [via Friedman]
[…]
“That’s because the tight money policy of 2008 that led to the low interest rates today also drove NGDP more than 10% below trend.”
I see it differently. IMHO, it has nothing to do with whether money policy is tight or easy.
And it has everything to do with real economy. If real economy is booming (i.e., providing the healthy returns) then holders of capital are requiring compatible healthy returns on money in the financial markets. If real economy is stagnating (i.e., providing the sluggish returns) then holders of capital are satisfied with compatible miserable returns on money in the financial markets.
PS Above I’ve excluded the Fed from the picture. Its inclusion will somewhat complicate the depiction but the real economy impact will still prevail in my view.
28. February 2012 at 10:00
Scott, maybe you could provide some clarifications around two points above:
1 – If we’re never supposed to reason from a change in interest rates, why are low interest rates a sign of tight money? (Please refer to the first half of that question before answering the second half.)
2 – If I understand your (and Friedman’s) reasoning correctly, low interest rates today indicate that there was tight money yesterday. Why do you spend the rest of your blog post calling tight money and low interest rates the same thing? Are you suggesting that if the Fed lowers one of the key interest rates, that is a tight money policy?
28. February 2012 at 10:02
Scott,
The fact that you don’t take your success for granted means a lot to readers. Telling us you have to scale back to a level just 10 time that of most humans is endearing.
Reinhart misses the the fact that those same “savers” were holding a lot of bad debt instruments that were cleaned up at above market price, by ZIRP. They did just fine.
28. February 2012 at 10:12
Indeed it would be fair to say rates are low today precisely because NGDP fell so sharply.
It would be even fairer to say that the low rates today are an effect of the same thing that caused NGDP to fall, which only appears as a positive correlation between the two.
Ben CTRL-P Bernanke has been printing money like gangbusters since around June 2011, and inflation of the money supply (M2) is now running at a 10% annual rate.
Price inflation is already starting to heat up, but like BTO song goes, “B-b-b-b-b-baby you ain’t seen nothing yet.”
Interest rates are going to soar, and it will also make NGDP soar. Of course we can expect that you will say “Interest rates are going up because NGDP is going up.”
28. February 2012 at 10:15
It’s analogous to shallow arguments about the disparate effects of inflation and deflation on debtors and creditors.
On the face of it, borrowers benefit by paying off debts in money that is worth less than expected when the loan was originated, while lenders benefit when the money is worth more than expected. All else being equal, this is, of course, uncontroversially true. However, if the inflation or deflation is caused by monetary disequilibrium, then the risks associated with loans also change.
Due to unexpected deflation, a lender may receive money that is worth more than what had been lent, but if this was caused by an excess demand for money, then the risk of default may have also risen as debtors suffer an unexpected tightening of their monetary income. The lender may get higher returns but at the cost of higher default-risk, but if that was the combination of risk and return the lender wanted, then they would have made different loans in the first place. It’s far from clear why a lender would benefit from deflation if it forces him to accept a combination of risk and return that he doesn’t want.
The same is true, in reverse, for debtors. When borrowers discover unexpectedly high nominal incomes caused by an excess supply of money, they will also have an unwanted combination of risk and return. Sure, they get to pay off the debt in devalued money, but they also find out that they missed an opportunity to borrow even more and still pay it off comfortably.
This is really just more proof that inflation and deflation are very weak concepts–they tell us very little about what is actually going on.
28. February 2012 at 10:33
According to Wikipedia, the term “financial repression” was coined by Edward Shaw and Ronald McKinnon in 1973. Both were at Stanford.
I view the phrase “financial repression” as a politically motivated term and I hate hate hate the way Reinhart, Rogoff and the WSJ editorial board have pushed it out into the mainstream media and financial community. It appears *everywhere* now on Bloomberg, CNBC, and various punditry (such as Bill Gross and El-Erian). I view use of the term “financial repression” as prima facie evidence that the speaker lacks the ability to think clearly about economic matters. It is “the rape and plunder” of economics.
28. February 2012 at 11:27
negative real rates are “bad” when they are combined with low nominal rates… but good otherwise?? huh?
pension funds have duration and funding targets including equity/fixed income asset allocation mixes depending on the profile of the underlying liability (future payments to retirees). A pension fund counting on 8% return is looking mostly to equities to provide that, not fixed income.
if ngdp declines so do equities, and the pension looks more underfunded.
the rape and pillage is being caused by the Fed (tight money). I can almost guarantee if the Fed steered ngdp 30% higher over the next several years pensions would look to be in much better shape…
28. February 2012 at 11:51
Lee Kelly-
Yes, I am a Scott Sumner fan. Given the competition in the bulk of the economics “profession”—knee-jerk tight moneyists, or Krugman spenders—I think my infatuation is justified.
28. February 2012 at 12:02
Given the competition in the bulk of the economics “profession” “” knee-jerk loose moneyists, or Krugman spenders “” I think my non-infatuation is justified.
28. February 2012 at 12:07
Had the fed been targeting NGDP for the past few years and been prepared to use non-banking channels and expectations settings to increase the money supply then we would have long since escaped the zero-bound and have a healthier economy with higher interest rates and this would have benefited both creditors and debtors alike.
But I am cynical enough to believe that the government may have seen enough short time advantage in maintaining its cheap borrowing options to have influenced the fed to stay at the zero-bound and avoid more serious consideration of more appropriate and aggressive policies. After all especially when it can borrow at close to zero why should the govt care, at least in the short-term , about debt/GDP ratios ?
28. February 2012 at 12:34
“After all especially when it can borrow at close to zero why should the govt care, at least in the short-term , about debt/GDP ratios ?”
Exactly. What is the textbook signal carried by high prices to a monopolist? Maybe the monopolist issuer of government debt should respond to the market signal by increasing supply.
Major’s head might explode, but that explosion will have only an imperceptible impact on the market.
28. February 2012 at 12:51
Major Freedom-
Milton Friedman, Alan Meltzer, John Taylor and Ben Bernanke all called for QE, hot and heavy, in Japan.
Sadly, MF has passed on, but Meltzer and Taylor now do not call for QE in the USA. Bernanke tried a feeble, small amount of QE, for a limited duration.
Inflation is at historic lows. We look like Japan.
Where are the “loose moneyists”? How do we get them onto the FOMC board?
Even a slutty, morally bankrupt, debauched loose moneyist would be better macroeconomist today than a Krugman or a John Taylor.
Better yet is a Scott Sumner.
28. February 2012 at 14:11
Scott,
If tight money leads to low interest rates and vice versa, how come interest rates have been historically low over the past 30 year period while the quantity of money (by any measure) has grown at historically unprecedented rates?
What justifies historically low rates for the past 10, 20, 25 years? It certainly isn’t the US savings rate.
The Austrian explanation is that an unconstrained central bank will try to constantly boost the supply of money in an effort to drive down interest rates with the goal of inducing a permanent quasi-boom (to paraphrase Keynes). What’s the monetarist answer for this stuff?
28. February 2012 at 14:12
Benjamin Cole,
How can you be fan of something unproven?
I’ve just presented (here and in the previous thread as well) a legitimate alternative view on the behavior of interest rate and am waiting patiently for the Scott’s intelligent response.
28. February 2012 at 16:22
pls tell me book comes out after Nov 2012
28. February 2012 at 16:27
@math
but whether the economy is booming or not is driven by monetary policy (whether the Fed is expanding the monetary base, or not).
so when the fed is expanding, the economy is booming, and real interest rates are higher. tight money means the fed is not expanding fast enough, the economy is not booming, and hence interest rates are low.
sounds to me your idea is same idea as Friedman – the idea of a boom is expanding nominal aggregates. the only difference is that the Fed can control whether the economy is booming by expanding money. hence low rates are the symptom of tight money.
28. February 2012 at 16:41
“Fed can control whether the economy is booming by expanding money. hence low rates are the symptom of tight money.”
As in, “the market for money is tight, as witnessed by the high price.”
As opposed to “the Fed is tightening the market for money, as witnessed by the high rates they are charging for money, which causes banks to reduce supply relative to demand.”
One must be careful to avoid confusing policy and outcome.
28. February 2012 at 16:50
dwb,
Thanks for your comment. Let me wait for Scott’s response first then I’ll answer you about the Fed’s role in economic expansion (and I suspect it’s gone).
28. February 2012 at 17:34
Thanks Becky and Ben, and . . . and Lee 🙂
Math, Yes, but it’s the Fed’s tight money policy that causes the real economy to be so weak.
I do agree that weakness for some other reason can have a modest impact on rates, but think NGDP growth is the dominant factor.
Ryan, The point of the Friedman quote was to show why you shouldn’t reason from a price change. Low interest rates could mean easy money, but as Friedman points out they usually mean money has been tight.
I probably should have made that clearer.
Window washer, Good point.
Major Freedom, You said;
“Interest rates are going to soar, and it will also make NGDP soar.”
You’ve reversed causality.
Lee Kelly, Good points.
Steve, In fairness, Reinhart approves of the policy, which is different from the WSJ’s view.
dwb, I agree.
Ron, I’m not quite that cynical–I think there was an element of ignorance involved.
John, NGDP growth has trended downward since 1980. That’s caused nominal interest rates to trend downward. When rates fall there is a rise in money demand, so the money growth rates will look unusually high relative to NGDP growth during the transition to lower NGDP growth. It’s a question of changes in trends vs. changes in levels.
It’s been well known for quite some time that a slowdown in money growth will often be followed by temporarily higher money growth a demand for money rises at lower nominal rates.
I seem to recall that in 1924 there was lots of growth in the German money supply, but almost no inflation, as they went back on gold (and lowered nominal rates.) Most people would say that was tight money.
Morgan, Yes, after Obama is re-elected.
29. February 2012 at 05:27
Scott,
The money supply (any way you wanna measure it) has blown up like a hockey stick, especially since the 1990s. But even before that it had really started growing since Nixon broke the tie to gold in 1971. So you’re saying this has to do with lower NGDP affecting money growth and not the Fed creating more money than ever before? Stop pulling my leg.
29. February 2012 at 06:04
John, NGDP depends on money supply and demand. Inflation has been falling steadily since 1980–you need to face facts that the old monetarist model focusing on money growth alone is obsolete.
29. February 2012 at 06:06
John,
Do you mean the money stock or the money supply? US broad money statistics from 2001-2011 are similar to the 1960s, with big swings from low growth or even contraction to big spikes.
The US money stock, laid out on a graph, will naturally have a hockey stick formation in any economy where the stock is increasing at anything other than a proportionately slowing growth rate. The exact same phenomenon can be observed in population statistics.
That said, apart from (what may yet be) a brief and peculiar period of stability from about 1983-2005, the key date for the age of monetary instability is indeed 1971. In every economy, to varying degrees, the end of the gold-link (direct in the case of the US, indirect for the rest of us) was followed by a period of outrageous monetary expansion, the consequences of which are still being felt today.
The lesson from that is that nominal anchors, like linking a currency to gold, are more stable than the alternatives. That’s why having an NGDP target is such a good idea: we can’t trust policymakers with anything more sophisticated.
29. February 2012 at 07:27
ssumner said,
“Math, Yes, but it’s the Fed’s tight money policy that causes the real economy to be so weak.”
That’s not a scientific but a religious answer. Guys, sorry but you have become a religious sect in my view like MMT-ers.
dwb,
I want to break my promise to discuss about Fed’s role. I don’t like talking to guys with idée fixe.
Have fun playing with your dogmas. I rather go back devising math models.
29. February 2012 at 08:51
@math
not dogma, empirics, plus occams razor. i have vast experience with math models, spent many late nights with proofs and code – they are at best cartoons of reality (in economics, anyway), and can be created to derive anything i want based on input assumptions. garbage in = garbage out. at worst they make counterfactual predictions yet people stick to them (thats dogma). I can recreate a nifty model that perfectly matches most planetary orbital data as observed from earth, yet with the sun revolving around the earth. occams razor and satellite pictures say otherwise and suggest the model is wrong. Creating a math model is no big trick. knowing how to test the assumptions, thats the real value.
reality is the thing which, when you stop believing in it, is still there.
29. February 2012 at 09:57
dwb,
You are completely wrong. You have no idea what math models are, how they are created and why they are used.
Nobody (except pseudo-scientific liars of course) creates wrong math models (like “the sun revolving around the earth”) on purpose. They are created because better ideas are not conceived yet. As soon as better ideas are devised, they are incorporated into the math models.
Math modelers as a rule have no prejudice. They develop models and make their conclusions based on the models and not on some kind of preconceived ideology.
From a philosophical point of view, true scientists understand that world around us is complex enough that can only be described through the quantitative equations and not through some vague qualitative observations as the ideologues and charlatans envisage.
As an example, try to create a complex mechanical or electrical construction without using the math models and quantitative equations. It’s rather impossible.
Ciao
29. February 2012 at 10:39
ssumer:
“Interest rates are going to soar, and it will also make NGDP soar.”
You’ve reversed causality.
*Sigh*, I wasn’t making a causality statement between interest rates and NGDP.
The “it” in “it will also” is NOT interest rates. It is the aggregate monetary inflation. I am not saying interest rates will cause NGDP to soar. I am saying the same thing that will cause interest rates to go up is the same thing that will cause NGDP to go up (that is, if the Fed isn’t continuing to lower interest rates as a byproduct of inflating bank reserve in an attempt to continue to boost NGDP).
And NGDP changes do not cause interest rate changes. What proximately affects true market interest rates is profitability, and what proximately affects nominal interest rates is nominal profitability, which itself is a function of aggregate monetary inflation, and aggregate monetary inflation in turn affects NGDP.
NGDP can go up by 5%, but if profitability doesn’t go up, then neither can interest rates. Only to the extent that NGDP growth accompanies profitability growth will interest rates rise, which is the same thing as saying it’s not NGDP that causes nominal interest rate changes. NGDP is an effect of the same cause of nominal interest rates.
29. February 2012 at 11:33
@math:
I am glad to see your faith is strong. i’ve been in your shoes I bet. Please hurry up and solve all the problems of monetary theory now while you know everything. while i still have time to enjoy it.
29. February 2012 at 18:37
If I win the bet, you dedicate the book to me, k?
1. March 2012 at 06:11
W. Peden,
You’re talking about what I like to call the gold standard vs. the PhD standard. At least you can admit that having an anchor to gold was stabilizing. I don’t trust the PhDs to be as stable as gold, or even smarter.
1. March 2012 at 16:38
John,
I think it depends entirely on what the doctors are doing with those PhDs. However, I do agree that- in the absence of a central bank- a gold standard system will considerably outperform any central bank targeting real variables (as occured in the 1970s) and a semi-gold standard system like the Bretton Woods system will likely outperform a real variable target.
And a currency board, a central bank targeting a sensible nominal variable like NGDP, the price of money (aka “inflation”) or even a money supply target will outperform all of the above. What good there is in a gold-based system comes out of the fact that it is, as Treasury Knights of the 1920s put it, “knave-proof”. Or at least, it is unless it’s a gold standard + central bank, since under those conditions central banks (notably the Bank of France and the Fed in the late 1920s/early 1930s) can make it exceptionally sub-optimal.
1. March 2012 at 17:51
Math, You said;
“Math modelers as a rule have no prejudice. They develop models and make their conclusions based on the models and not on some kind of preconceived ideology.”
That naivete is almost touching.
Major Freeman, I don’t doubt that there is some correlation between interest rates and profitability, but with all due respect is this really the best time to put forward that theory? Have you checked interest rates and corporate profits lately?
1. March 2012 at 21:59
ssumner:
Major Freeman, I don’t doubt that there is some correlation between interest rates and profitability, but with all due respect is this really the best time to put forward that theory? Have you checked interest rates and corporate profits lately?
[Facepalm]
Tap tap tap, is this thing on?
For the millionth time, interest rates are determined by profitability, but not when there is a central banking system continually injecting new money into the loan market, thus pulling interest rates down artificially.
In a free market, one without a central bank, interest rates will correlate even more closely with profitability (it won’t be perfect of course since it is possible that some of the newly mined money will be lent).
If the Fed stopped printing money tomorrow, and banks have no more fuel to expand loans ex nihilo, then interest rates will rise towards profitability.
2. March 2012 at 18:18
Major Freedom, The Fed was only created in 1913, banks created lots of loans before that. There’s no limit to loans that can be created. I loan my brother a trillion dollars and he simultaneously loans it back to me. Loans aren’t restricted by monetary policy.
4. March 2012 at 18:20
ssumner:
Major Freedom, The Fed was only created in 1913, banks created lots of loans before that. There’s no limit to loans that can be created. I loan my brother a trillion dollars and he simultaneously loans it back to me. Loans aren’t restricted by monetary policy.
You aren’t loaning $1 trillion to your brother and he isn’t simultaneously loaning $1 trillion to you. Your ability to loan money is restricted by how much money you own. You can’t lend $1 trillion that you don’t have. What kind of fantasyland nonsense is this?
Loans from the banking system are restricted by monetary policy. The more reserves increase by the Fed, the more loans the banks can create. If the Fed stops inflation tomorrow, then at some point, the banking system will not be able to lend any more on net, because the growing withdrawal and transfer requests will restrict their cash flows going to additional loans until there is no money left to add to loans.
9. August 2013 at 14:02
[…] were alive, Paul would not like him as chairman, since Friedman wouldn’t fall into the “low interest rates equals easy money” fallacy, and would probably want the Fed to do more than they currently are. Friedman would probably make a […]
3. July 2015 at 17:40
[…] low interest rates signify easy money. Scott is quite flummoxed how anyone can think that, since Milton Friedman himself made the point in the late 1990s. Scott repeatedly lectures us to never reason from an interest rate […]