This will be a confusing post, because I'm writing it to try to get my head clear on something. And it's still not clear. There are no answers here: only questions, and strange thoughts.
1. Lots of people believe the Inflation Fallacy: "Inflation makes us worse off because a 1% rise in prices means we can afford to buy 1% less stuff, duh!". It is hard work to explain to people why this is wrong, and each new generation has to be taught it anew. Ours the Task Eternal.
But if the central bank targeted NGDP, the Inflation Fallacy would be true. With the central bank holding the path of nominal spending constant, every extra 1% of inflation really would mean that real spending falls by 1%. Ordinary people would understand macroeconomics better, without us having to teach them anything. Would that be a Good Thing?
It might be a good thing politically, because it is hard politically for an inflation-targeting central bank to try to increase inflation, when inflation is below target. But then it might be equally hard politically for an NGDP-targeting central bank to try to reduce NGDP, when NGDP is above target. "Why is the central bank trying to make us poorer?" would be heard in both cases.
Or maybe the economy would work better if people's macroeconomic understanding were more in accordance with macroeconomic reality? Macroeconomic coordination failures might be less likely, if ordinary people's expectations were closer to rational expectations.
2. There's a second macroeconomic fallacy. And it's even harder to explain why it's wrong. It doesn't have a name, so I will call it "the Interest Rate Fallacy": "If you want higher interest rates, then the central bank should just set a higher interest rate, duh!". I think many economists believe this fallacy too. After all, it does sound rather obviously true. We saw this fallacy at work in Sweden recently. The Riksbank wanted higher interest rates, to reduce the dangers of financial instability, so it set a higher interest rate. But the only result was that the Riksbank needed to set an even lower interest rate a little while later. Lars Svensson understood the Interest Rate Fallacy, but then ordinary people are not Lars Svensson.
Would it be better to change the world to make the Interest Rate Fallacy true? It is possible to do this in principle, but would it work in practice? And would the world be a better place if we made it work in practice?
I have already explained (one example of) how it could work in principle. The central bank pays out new (base) money as interest on old (base) money. So if the central bank sets a (nominal) interest rate of 5%, this means that the (base) money supply growth rate is also 5%. Higher money growth causes higher NGDP growth, which causes higher equilibrium nominal interest rates. Doing this converts an unstable system into a stable system. It converts an inverted pendulum, where you have to move the pivot point of the pendulum the "wrong" way initially, into a regular pendulum, where you move the pivot point of the pendulum in the same direction you want it to go.
But I'm not sure this would work well in practice. Most base money is currency, and it's hard to pay interest on currency. Plus, if some exogenous shock caused the demand for currency to increase by 100% over the course of one year, the central bank would need to pay 100% interest on currency for that year, just to prevent deflation. I don't think that would work very well.
But if something can work in principle, there must be some way to make it work in practice. That's why God created engineers.
Maybe some variant of Irving Fisher's Compensated Dollar Plan could make it work in practice. Set the dollar price of gold as a crawling peg, and change the rate at which the peg crawls. The central bank would loosen monetary policy by announcing that the dollar price of gold would be rising at a faster rate until further notice, and loan markets would respond by raising (nominal) interest rates. It doesn't have to be gold, of course.
But I don't understand this very well.
(And don't say I didn't warn you, if you have read this far and are disappointed.)
Worthwhile Canadian Initiative 
Majromax: “That would be a good thing, but we’d have to give up the idea that “inflation is controlled by how the Central Bank sets the money supply.””
Hmmm. Yes. We would have to replace it with “NGDP is controlled with how the central bank sets the money supply, and inflation depends on NGDP and on supply shocks.” Hmm. Actually, I think I like that better. It handles the supply side better.
“How about the “Jackknife Fallacy?” If you’re reversing a trailer and you want it to turn one way, you must first turn the driving vehicle slightly in the opposite direction.”
Good metaphor, but a lot of people don’t know how to reverse a trailer either! [Actually, reversing a trailer is like a double inverted pendulum. Reversing a truck with no trailer is like an inverted pendulum, since you turn the front of the truck North to make the truck go South. When I was a young man I could sometimes reverse a 4 wheel trailer (front axle swivels) into a barn. That was very hard. Treble inverted pendulum.]
“The Bank of Sweden wasn’t trying to raise inflation with its rate hike, after all, it was trying to prevent (future) inflation.”
I thought it was trying to prevent financial instability, and was willing to accept the (illusory) trade-off of lower than target future inflation.
I don’t want it to target both inflation and NGDP; that can’t be done. But could we figure out a way it could use an instrument to target NGDP (or inflation) where if NGDP were below target, interest rates would rise? So that the inverted pendulum becomes a regular pendulum.
That would have an additional advantage at the ZLB. The only cost of hitting the ZLB would be that NGDP (or inflation) would be a bit above target.
[ Inflation Fallacy: Inflation makes us worse off because a 1% rise in prices means we can afford to buy 1% less stuff. ]
That isn’t a fallacy. When prices rise 1% overall, cash and bond savings buy 1% less.
Wages lag inflation, requiring people to renegotiate. In the interim, their wages buy less. How does one view inflation as neutral and even magically beneficial?
As people come to expect inflation, they attempt to get ahead of it with salary demands, higher interest expected on loans, and uncertainty about future payments in long-term contracts. I don’t see the benefit.
You don’t state directly how engineered inflation is going to make us rich, and there is no mention of any possible negatives. It seems that people dislike inflation because they are utterly stupid, and not because inflation brings to mind Germany, Zimbabwe, and the US in the late 70’s.
Please describe how to do inflation the right way, unlike in those examples.
In principle, you could arrange every aspect of the peasant’s lives to make them better off. It only remains for philosopher-engineers to take care of the details. There must be a way to do anything which is possible in principle.
“but a lot of people don’t know how to reverse a trailer either” Used to have to do that with the business, then I got a station wagon.
“When I was a young man I could sometimes reverse a 4 wheel trailer (front axle swivels) into a barn. That was very hard. Treble inverted pendulum.” Yikes. But that is presumably where you learnt to speak with slow rural dignity 🙂
Is that the interest rate fallacy? That’s not how I interpret what they’re saying. Maybe I don’t get it but, i think it means that the interest rate is a function of (government) deficit spending and that if you want rates on (government) liabilities to be higher, the supply of (government) liabilities relative to demand for (government) liabilities needs to be high enough to justify higher rates. Government as borrower of last resort. Pulling as opposed to pushing on a string.
I’m not smart enough to understand how your paying interest on old money isn’t fiscal policy. How isn’t it paid by government? The central bank doesn’t run a deficit (or have a negative net worth) does it? It pays interest out of interest earned by lending? Is the central bank allowed to be a net payer of interest? If it isn’t who is paying interest to money holders?
I hope that made sense and doesn’t make me sound too foolish.
The inflation idea may be a fallacy in aggregate, but it is certainly possible to be true for an individual.
Jim: it may be true in aggregate as well, that inflation makes us worse off. Indeed it is true in aggregate, at least when inflation gets above some level. An invalid argument may have a true consequent. There are costs to inflation (and some benefits too). But the inflation fallacy is still a fallacy, because it is an invalid argument about the costs of inflation.
Majromax: yes, there are two effects: it increases the demand for money and also increases the growth rate of the supply of money. With perfectly flexible prices and rational expectations it is easy to see the net result– increased inflation but no jump in the price level. But if there were a jump up in money demand, we would need a jump up in money supply to offset it.
Andrew: it is a fallacy. For every apple bought there is an apple sold. If the price of that apple rises by $1, the buyer is $1 poorer and the seller $1 richer. We can always afford to buy the goods we ourselves produce. That is the valid kernel of truth in the correct version of Say’s Law.
But you are right about the money-wealth effect. And maybe right about the bond-wealth effect too, in an overlapping generations model.
Lorenzo. The old farm workers could do it. Last month a truck driver was lost on Carleton Campus. So I joined him in the cab to show him where to go. And he reversed that big artic perfectly through narrow bends. He didn’t think it was anything special. There is a spatial sense that some people have, and can develop with practice. But others don’t have a clue.
Miami: makes sense, and not foolish, but it is wrong. Compared to a baseline with no money printing, if the central bank prints money and gives it away (or pays it as interest) there is no effect on the government budget constraint. But compared to a baseline where the central bank prints money and gives it to the government, or uses it to buy bonds and pays the interest on those bonds to the government, there is an effect on the government budget constraint.
Paying interest to people with money is a redistribution to people with money
Really would make us hold cash eh?
Keynes mentioned stamped money which is the opposite
You can keep interest rates high and redistribute money to the rentiers and shrink the economy if you don’t care about people in general
In fact Keynes thought this the usual… keeping interest rates too high
When I started Reading the post, I expected an explanation why inflation does not make every single individual worse off. Notice that this is different from an entire economy better (or worse) off, or government finaces better (or worse) off. But I could not find any. Maybe the author is tired of explaining. But considering that his considers his task “Eternal”, please, do it one more time.
Shorter everyone: At the aggregate level, inflation might be benign, but it is re-distributive at the individual level and so there will be unhappy losers and happy gainers who will lobby for their interests. Behavioral economics will assert that, in an equal transfer, the losers will be made far more unhappy than the gainers were made happier. This is why the “fallacy” persists.
“Lots of people believe the Inflation Fallacy: “Inflation makes us worse off because a 1% rise in prices means we can afford to buy 1% less stuff, duh!”. It is hard work to explain to people why this is wrong, and each new generation has to be taught it anew.”
Isn’t that a question of definition? For an economist, if prices rise but incomes do not, that is not inflation, right? But a price rise, whether wages rise or not, is inflation in common parlance.
However, mean income can rise with prices while median income does not. In that case, for most people a rise in inflation does mean that they can afford to buy less stuff.
“It might be a good thing politically, because it is hard politically for an inflation-targeting central bank to try to increase inflation, when inflation is below target.”
Really? Do you think that politics is what has prevented CBs from increasing inflation? (Aside from the ideology of certain central bankers.) Kaletsky ( http://blogs.reuters.com/anatole-kaletsky/2014/10/31/the-takeaway-from-six-years-of-economic-troubles-keynes-was-right/ ) thinks that the evidence of the past few years indicates that differences in fiscal policy is the key to differences in outcomes since the financial crisis.
“Macroeconomic coordination failures might be less likely, if ordinary people’s expectations were closer to rational expectations.”
Theories based upon so-called ‘rational expectations’ might work better if they accorded with ordinary people’s expectations.
1) If you work for a living, inflation is not a problem. Your salary is one of the rising costs. Most people did just fine in the 1970s, because they got regular cost of living increases. Even better, their home mortgage payments shrank radically. Inflation hurts those who do not work for a living since they rely on nominal values to purchase the goods and services they desire. Ending inflation to satisfy them required a massive recession that we have still not recovered from over 30 years later. (Sure, we “recovered” from each recession, but wages have been stagnant or falling, and each “recovery” takes longer than the previous one.)
2) It is possible to set higher interest rates by just setting higher interest rates. This happened in the late 1970s, and 20% mortgages were suddenly no longer a rarity. Why it can’t be done today is an interesting research topic. My guess is it is because banking operations are so completely divorced from the actual economy that 99+% of the population relies on.
In your interest on old money model at what rate does government income and expenditure change so that it remains the same in real terms? Isn’t it the same as the rate paid on old money? If central bank liabilities grow at rate m, government liabilities must also. In terms of real resources and because the accounting requires it, i think.
@djb:
Jose: the Inflation Fallacy is a fallacy because it implicitly assumes that whatever caused the inflation had no effect on dollar incomes. And since aggregate dollar income from the sale of newly-produced goods is exactly the same thing as aggregate dollar expenditure on newly-produced goods, that is a very dodgy assumption. You need a theory to tell you whether whatever caused the inflation will also cause an equal increase in dollar incomes. And that depends on what causes the inflation, and on the economic theory. For example, it is easy to build a theory where a 1% increase in the money supply causes a 1% increase in all prices and a 1% increase in all dollar incomes, so the only thing that happens is that each dollar is worth less, but nothing real changes. A theory like that is definitely not the last word on the subject, but it does create a counterexample, to demonstrate that the Inflation Fallacy is a fallacy, because it assumes the conclusion.
Shorter, simpler, example: If I earn my income from selling apples to you, and you earn your income from selling bananas to me, then if we both double our prices we can both double our incomes and can afford to buy exactly what we were buying before. (Whether we will choose to continue buying exactly what we were buying before is a separate question. But if the money supply doubles, and all prices double, we might choose to do this.)
Min: “Isn’t that a question of definition? For an economist, if prices rise but incomes do not, that is not inflation, right? But a price rise, whether wages rise or not, is inflation in common parlance.”
That is a different fallacy: the identification of “wages” with “income”. Capitalists are people too.
And if that’s what Kaletski said, he is wrong. Compare the Eurozone with the US. US tightened fiscal more, but looser monetary policy offset it, and the US recovered faster. But that’s off-topic. Yes, politics matters. See Germany.
Miami: “If central bank liabilities grow at rate m, government liabilities must also. In terms of real resources and because the accounting requires it, i think.”
Compare it to a 2 for 1 stock split. Each stock is only worth half as much as before, and the firm’s assets and liabilities are unchanged in real terms. It’s just we are talking about a continuous 1+m for 1 stock split, and all other prices are measured in those stocks.
Kaleberg: “2) It is possible to set higher interest rates by just setting higher interest rates. This happened in the late 1970s, and 20% mortgages were suddenly no longer a rarity.”
That happened in Canada in 1982(?). And the result was that interest rates subsequently fell, to a lower level than they had previously been. Canada in 1982 was exactly like Sweden recently, only moreso, and the Bank of Canada understood the Interest Rate Fallacy, and knew that by raising interest rates it would cause interest rates subsequently to fall.
Majromax doing sterling work.
My old post on the Inflation Fallacy
MOI: “Isn’t that a question of definition? For an economist, if prices rise but incomes do not, that is not inflation, right? But a price rise, whether wages rise or not, is inflation in common parlance.”
Nick Rowe: That is a different fallacy: the identification of “wages” with “income”. Capitalists are people too.
Sorry if I gave the impression that that’s what I think. Besides, I am not the one who came up with terms like “wage-price spiral”. 😉
Pedagogically, I think that it is easier to tell students that what they think inflation is is not what it means as a technical term. Easier for the students, that is. It is not that they are thinking incorrectly, it is that they are not thinking systemically at all. What they have grown up hearing is inflation in terms of some price index or other, usually the CPI. Nobody has told them any different. Labeling their ignorance a fallacy needlessly engenders resistance.
I was able to understand your model when I thought about the central bank as a repository for money (a real thing not unlike gold). Account holders are paid interest in real (gold like) money on deposit at the central bank. whatever is on the asset can be ignored since it didn’t change.
The curious thing to me as a non economist bystander is it seems like monetarists saying MV=C+G+I+(X-M) while the other side(FTPL?) is saying PQ=C+G+I+(X-M)
One side says increase M the other, increase Q; rinse, repeat.
One describing it in PQ the other in MV and both assuming the starting point (catalyst?) while the truth seems to be yes, both, continuously.
Each side has a preferred authority with a lever.
One side has nominal growth as a goal, the other real growth that’s potentially unsustainable and/or of questionable quality.
Is that the gist, sort of. Each side incorporating more or less similar assumptions and getting the same answers one describing it the ways M and V change and the other side P and Q.
Each side starting from the opposite side of the equation.
Monetarists arguing they have money and a machine that show the future and there’s 5% inflation in the future so people rush to buy things before they lose out to inflation causing the inflation they were hoping to get in front of.
FTPL people arguing for directly changing income flows through fiscal policy; actively managing the distribution and some provisioning of output.
Each side starting at the opposite side of MV=PQ, if that wasn’t clear in my previous comment.
Was my accounting correct (at least), given current institutional arrangements?
Nick, Suppose you had a country (Japan) with 0% trend inflation and 0.5% long term rates, and another country (USA) with 2% trend inflation and 2.5% long term rates. How do the Japanese peg the inflation rate at 2%? They could fix their currency to the dollar, and then (according to PPP) inflation expectations in the long run should rise to 2%. And according to interest parity the long term interest rate in Japan should rise to 2.5%. Immediately. John Cochrane would approve.
Your stock splitting analogy doesn’t describe your model. For a stock analogy to describe your model you would need to be talking about changes to the dividend, not stock splitting. That’s because you’re giving it to existing equity holders in proportion to their existing holdings. Nothing material changes other than the number of pieces you can divide your identical equity position.
@Scott Sumner
What assumptions are you making about the success of US monetary policy? Presumably, what you suggest for Japan would be an obstacle for US monetary policy.
Sorry, my mistake, stock splitting makes sense as an analogy. Prices in stock change but not in terms of the fraction of assets required to make the same purchases. duh (not sure why i was thinking in terms of purchasing power). Thanks.
I was confused since, it seems, the analogy breaks down if the central bank buys assets with stock it issues.
The stock analogy combined with the fact that the central bank is, usually, limited to buying government securities if it wants to issue more stock seems to validate the FTPL. For your analogy to make sense, given current institutional arrangements, it would seem like the central bank would need to promise to make crappy investments. If it can only invest in government securities they basically need to hope (like Bernanke did in his testimony) the fiscal authority spends enough to make it true. I’m sure you interpret things differently.
Min: OK. That might work. Or maybe we should start out by saying: it depends on what causes the inflation.
Scott: I think that would work, except for changes in equilibrium real exchange rates. But maybe a crawling peg could handle that, where the BoJ adjusts the rate at which it crawls. I think it’s similar to my gold example, except you have the US dollar replace gold. (Which has historical echoes!). All variants on Irving Fisher’s compensated dollar plan.
Miami: we normally assume that stocks trade in competitive markets. Two stocks with the same risk must trade at the same rate of return, and there are lots of stocks. Each individual stock faces a perfectly elastic demand curve at that market-determined rate of return. Zimbabwe shows that assumption does not work for money. The issuer of alpha money has de facto or de jure monopoly power, because that money is not a perfect substitute for anything else, and faces a downward-sloping demand curve. People demanded a smaller, but still positive stock of Zim dollars even when the rate of return on Zim dollars fell to a very negative number. The rate of return needed to fall a ridiculous amount before the Zim dollar lost all demand.
@Scott Sumner:
Nick Rowe: “Or maybe we should start out by saying: it depends on what causes the inflation.”
When I was a kid I heard about the Weimar hyperinflation. The tale was that people were paid their wages twice a day, which they put into wheelbarrows and rushed to the stores to buy stuff before the prices went up again. Obviously incomes had to go up with prices or they would quickly not be able to buy anything. That story primed me not to think that inflation was only about prices, even though, later on, I heard people complain about inflation when prices rose but their incomes did not. Yet. OC, my parents heard about the Weimar hyperinflation when they were kids. 🙂 I doubt if many kids hear that story anymore. Maybe they hear about Zimbabwe.
Anyway, as to where to start, as one of my philosophy profs tautologically said, we start from where we are. (I have told the story about that class. :)) I reread your original post about the inflation fallacy. You ask the students why inflation is bad. But where are the students starting from? What do they think inflation is? Do they think that it is bad?
Zimbabwe, seriously?
never mind the civil war, the land reforms, drought, collapsed output, sanctions, capital controls, required use of foreign exchange to buy food and raw materials (output had collapsed), massive corruption. (asset side)
Zimbabwe was the result of unsustainable unfunded government spending due to the things listed above. How in your mind did the money printed Zimbabwe make it into the economy or was simply printing it enough?
Stock splits don’t change the rate of return. How does owning 1/1000 or 2/2000 of Apple inc change the rate of return?
Prices of in terms of apple stock would double, sure. The real rate of return is zero in your model, always. It isn’t negative BECAUSE its paid on old money.
A stock split is nothing more than a change on paper to facilitate certain advantages. Each current stockholder’s holdings of stock double so that the proportion of the firm owed by the investor is the same both before and after the stock split. No loss of value is borne by the current stockholders.
I’m not sure from where you conjure up the negative return due to stock splitting. can you please explain?
Miami: Zimbabwe didn’t stock split. They printed new stock, and gave/lent it to Mugabe.
That was my point. I was talking past you, thinking you were saying something different than you were. My original point which I lost while ranting is, that Cochrane is saying the same thing as you. If the rate of interest paid on old money rises from zero to 5% and government spending on real resources falls by 5% (remains the same nominaly) you have what Cochrane is saying. Government saving makes raising rates a successful tool to combat inflation. Mugabe didn’t respond to rising rates the way a rational agent (borrower not to mention his lender) would if they were trying to avoid insolvency. Therefore, hyperinflation. The reason why Cochrane’s model makes sense is that its the same as yours except he’s explicitly modelling the behavior of the largest money user/issuer in the economy. That’s how i interpret it.
Other than attributing the cause to the quantity of money or the cause to spending decisions your models are the same. Same moving parts same outcome different order.
Miami: no. You have missed what I said about the demand curve sloping down.
If you don’t mind, can you be more explicit as to how I’ve missed that the demand curve slopes down?
of course no one will pay 2 apple shares for 1 apple share or $6 for $5. Uncompensated risks (I listed some above, which includes Mugabe paying himself and for stuff without the income to support it) drove the Zimbabwe $ value down.
Zimbabwe money continued to have liquidity value even as its price was falling. Probably less so, as use of foreign currency became more widespread.
So risk is held constant in your model? it doesn’t vary because of the real things I listed above?
Miami: to keep it simple, ignore government bonds. All government deficits are money financed.
FTPL says M/P = PV(fiscal surpluses). This treats money just like a stock.
But there is a rate of interest r(t) in that PV calculation. Suppose that r(t) is not exogenous wrt M(t)/P(t), but is a positive function of M(t)/P(t). Because Md/P is a negative function of the opportunity cost of holding money, which is the difference between the rate of return on other assets minus the rate of return r(t) on money.
Nick, I agree.
Why isn’t L(R-p) the same thing as the present value of the fiscal surplus?
sorry disregard
Plus, the real rate of return on holding money can be, and usually is, negative. (Currency pays 0% nominal interst and inflation is often positive). So the PV of fiscal surpluses doesn’t even add up.
M/P = P.L(R-p)/P
M/P = PV(fiscal surplus)
i’m sorry i keep bugging you but, those aren’t the same?
typo should be (fiscal surpluses)
off topic but i’m curious
Do you view expectations management by the central bank like this?
Expansion: for a limited time only we’re lending money at real discount prices, sale ends when we hit our target
contraction: bring us your money, we’ve raised the real price we pay you to lend us money, for a limited time only
because that’s expectations as i understand it; how it works and its inherent limitations or am i crazy?
Miami: not crazy. Sounds roughly right to me. That’s roughly what the Bank of Canada does in normal times.
So when you were wondering if Cochrane’s model was a monetarist model,I was saying, I think, they’re the same model (yours expressed in terms of money demand). So I wasn’t going mental? Phew.