More reasons to ignore inflation
I frequently argue that inflation is a highly misleading variable, and should be dropped from macroeconomic analysis. To replace it, we’d be better off looking at variables such as NGDP growth and nominal hourly wage rates. Tyler Cowen found a paper by Coibion and Gorodnichenko that illustrates the problems with using inflation (although they reach very different conclusions.) The paper starts off with a quote from Bob Hall:
“Prior to the recent deep worldwide recession, macroeconomists of all schools took a negative relation between slack and declining inflation as an axiom. Few seem to have awakened to the recent experience as a contradiction to the axiom.” Bob Hall (2013)
Not my school!! In my book on the Great Depression (which my publisher seems determined shall never see the light of day) I argue that the standard model is wrong, slack does not cause disinflation. For instance, prices rose sharply after March 1933, despite the greatest level of slack in US history. Rather falling NGDP causes slack, and is often associated with falling inflation.
Here’s what they discovered about inflation expectations:
Specifically, we show that an expectations-augmented Phillips curve, using household inflation expectations as measured by the Michigan Survey of Consumers, can account for the absence of strong disinflationary pressures since 2009. The primary reason for the success of a household inflation expectation-augmented Phillips curve is that household inflation expectations experienced a sharp rise starting in 2009, going from a low of 2.5% to around 4% in 2013, whereas other measures of inflation expectations such as those from financial markets or professional forecasters have hovered in the close neighborhood of 2% over the same period.
During the housing price collapse I had great fun talking to average people about inflation. When they would point to rising gasoline and food prices, I pointed out that housing prices were plummeting, making houses much more affordable. They always gave me a strange look; “but aren’t falling houses prices bad?” That’s when it hit me; most people don’t know what inflation is. They confused “cost of living” with “standard of living.” The phrases do sound similar. I confirmed this by asking each of my classes “If all prices rise by 10%, and all incomes also rise by 10%, has the cost of living actually increased?” Almost everyone gets it wrong. And these are good students. Try it with your class.
Just as I expected, Coibion and Gorodnichenko found that the public’s confusion was due to the fact that they focused on price increases from the supply-side, i.e. those that reduce living standards, not the demand-side:
We document that more than half of the historical differences in inflation forecasts between households and professionals can be accounted for by the level of oil prices. With oil prices having risen sharply since 2009, this provides a quantitatively successful explanation for the rise in household inflation expectations.
Here’s another problem with using inflation:
And to the extent that this rise in inflationary expectations may have prevented the onset of pernicious deflationary dynamics, the rise in oil prices should perhaps be interpreted as a lucky break for policymakers, generating the very rise in inflationary expectations which policymakers have only recently begun to push aggressively toward in the form of forward guidance.
This is a beautiful example of “never reason from a price change.” Increases in inflation can be expansionary, but only if caused by more aggregate demand. That’s why you want to use NGDP growth, not inflation. Oil shocks don’t tend to raise NGDP, and hence aren’t expansionary.
Phillips did the original Phillips curve with wage inflation, and Paul Samuelson made a big mistake in switching over to price inflation. The Coibion and Gorodnichenko study shows that price disinflation was far less than expected during the recent period of high unemployment, but the wage version of the Phillips curve works just fine:
As with the two other measures of wage inflation, we find that the slope of the wage Phillips curve appears to have been stable across the pre-Great Recession sample and that wage outcomes during the Great Recession period are fully in line with what the earlier historical experience would have led one to expect.
Macroeconomists need to start ignoring inflation (unless they are trying to compare living standards across time–a pretty hopeless enterprise.) And they need to start focusing on the nominal aggregates that really matter, NGDP and nominal hourly wages. It’s the W/NGDP ratio, stupid. (And always has been.)
Tags:
28. October 2013 at 08:15
You could argue that if prices of widely used commodities are going up because of unexpected scarcity, these commodities have become intrinsically more valuable and their higher prices are justified. In these cases rising prices are not really caused by monetary phenomenons but by real world constraints. It would be physically impossible to continue on a fast rising consumption path for these commodities.
It doesn’t make sense to snuff out this kind of inflation. You will end up underpricing everything relative to “intrinsic” value.
I’m not sure if I understand things correctly but it seems to me like NGDPLT targeting would allow for these “justified” bouts of inflation yet be better at preventing unwanted inflation the rest of the time.
28. October 2013 at 09:10
I think the one of the nightmares of the doves at the Fed is this:
The Fed loosens substantially, and most of the money flows into commodities, not NGDP growth
28. October 2013 at 09:19
Speaking of Inflation was looking at an old publication from the late 1970s a while back..and there was an ad to lease a Mercedes for $700 a month…..that would be around $2500/month in today’s money…I couldn’t believe it…
28. October 2013 at 09:25
One of the problems of the eurozone is that when talking to almost everyone, if you complain that the problem is with monetary policy, they might agree and tell you that inflation has been too high.
The switch to the euro provides and anchor point and so repeatedly you’ll hear: “well, before the euro this used to cost X and now costs much more.”
If you try to say that inflation has been kept too low at too high of a cost, they’ll look at you as if you’re crazy (or, more likely, dismiss you as a hopelessly clueless neoliberal anglo-saxon; at least that’s what happens to me :).
Thus, while we may think that the ECB is a bunch of crazy hard money lunatics who prefer 20+% unemployment in the South to 2% inflation, the average upper-middle class European will tell you inflation is too high. The ECB is actually pretty much following the opinion of the broad elite, perhaps with a heavier weight on the opinion of older people.
28. October 2013 at 09:31
Edward, I think monetary easing affects the economy more broadly than that but why would it be a nightmare even if money went into commodities?
How is putting people to work mining copper and iron instead of staying idle considered a problem? Now is an efficient time to do more mining because of excess human resources keeping wages low.
It could be costlier to do extraction later when the baby boomers are retired and labor is tight and expensive.
28. October 2013 at 09:40
Scott,
Inflation is a useful term when properly defined. Inflation obviously comes from the verb to inflate meaning to make larger. It means making the supply of money larger regardless of any effect on prices. Obviously with more money and prices will tend to be higher other things equal, but rising prices are merely one of many consequences of increasing (inflating) the money supply.
You are right that for macro NGDP is more useful than inflation since money supply is hard to measure (depends on the measurement used) and all the arguments you made above apply.
When you say, “I confirmed this by asking each of my classes “If all prices rise by 10%, and all incomes also rise by 10%, has the cost of living actually increased?” Clearly the answer is that the cost of living has increased but the standard of living has remained constant. However, in the real world a 10% increase in prices will have a wide range of distributional effects that generally hurt the worst off in society. Some people’s income may go up 20% others 0%. You can’t ignore the distributional effects. It tends to destroy the middle class and divide society into those that benefit and those that are hurt. Over time society becomes more and more composed of haves and have nots with extended inflations. Read the histories of any period of hyperinflation to see this illustrated.
28. October 2013 at 09:44
@Benoit
Like you I’m not sure if Edward’s right that doves fear money going into commodities, but if they are, I’d bet that it would be some version of anti-blue collar (or at least pro-white collar) prejudice. Central bankers are all highly educated, upper middle class to upper class white collar workers. They might associate mining with situations like poor West Virginia coal miners or the kind of danger you might see in a Chinese mine. The latter of which might bring in anti-foreign bias with the idea that the money would just be going overseas. In any event, they associate the costs (higher prices) with their fellow educated white collar workers, and the benefits (higher wages) with blue collar foreigners.
But as I said, I’m not sure that’s something the doves do fear so I’m not convinced this kind of thinking is actually happening.
28. October 2013 at 10:07
Edward, Maybe, but there is no way for monetary stimulus to raise real commodity prices without raising RGDP. So I’d be happy to see higher commodity prices, it would indicate rising living standards.
Luis, I really wish someone would give me a link showing that monetary stimulus is the eurozone is viewed as a neoliberal idea. It would make my day.
John, You said;
“Inflation obviously comes from the verb to inflate meaning to make larger. It means making the supply of money larger regardless of any effect on prices.”
This is a common misconception. All the economics textbooks define inflation as rising prices, not rising money supply. That’s also how the media defines it. A very few people have alternative definitions, but they are not at all widely accepted. If you use the money supply, then the so-called Japanese deflation never happened.
You said;
“However, in the real world a 10% increase in prices will have a wide range of distributional effects that generally hurt the worst off in society.”
This is not accurate. When America had 10% inflation it hurt the rich much more than the poor. When inflation was brought down (after 1981) it helped the rich a lot.
28. October 2013 at 10:25
“NGDP causes slack, and is often associated with falling inflation.”
However, I would say that it is falling RGDP that creates slack. Or, more precisely, RGDP growth below productivity growth creates slack. Certainly, if NGDP grows less than RGDP then there is disinflation by definition. But, that doesn’t necessarily mean that there was an increase in “slack.”
“Phillips did the original Phillips curve with wage inflation, and Paul Samuelson made a big mistake in switching over to price inflation.”
Perhaps rather than abandoning inflation, we should be looking for the right inflation measure, and for many aplications wage inflation may be a lot better than CPI, PCE, PCE x food and energy, etc.
28. October 2013 at 12:43
The fact that homes are a chimera of consumption and investment causes such a problem. I have never heard anyone say that inflation was really high in 2003 because bonds had gotten so expensive. If only home prices could be listed in terms of implicit yield….
28. October 2013 at 13:21
Scott and Benoit essambre, hopefully what you say is true. But for people on fixed incomes…
As long as they have the option of buying gold then maybe yes, it’s alright
28. October 2013 at 14:25
“During the housing price collapse I had great fun talking to average people about inflation. When they would point to rising gasoline and food prices, I pointed out that housing prices were plummeting, making houses much more affordable. They always gave me a strange look;”
You’re lucky. The “average people” I know would have (rightly) slapped you for being such an insufferable boob.
Q: What percentage of people actually faced increasing prices for the commodities they bought on the market like gasoline and food?
A: Close to 100%.
Q: What percentage of people actually faced falling prices for housing
A: Maybe 10%? Maybe that’s optimistic. The observed market price for something that people rarely buy is irrelevant to most people. The overwhelming majority of people were tied to long-term contracts (mortgages or leases) for housing that don’t or couldn’t change based on current pricing. So unless they happened to get extremely lucky (and be extremely brave) in 2009, they enjoyed no to little decline in the price they were paying for housing.
That is, their cost of living was, in fact, rising.
28. October 2013 at 14:36
Scott you realize that there’s a huge difference between saying:
#1 Ignore Inflation
and
#2 We don’t need much inflation
You of course believe both of these, but you don’t do routine posts saying #2
28. October 2013 at 14:49
“In my book on the Great Depression (which my publisher seems determined shall never see the light of day) I argue that the standard model is wrong, slack does not cause disinflation. For instance, prices rose sharply after March 1933, despite the greatest level of slack in US history.”
Some other extreme examples:
Canada, 1934
Prices rose 1.36% after falling 4.70% the previous year. The unemployment rate was 20.6%.
Denmark, 1933
Prices rose 3.37% after falling 1.11% the previous year. The unemployment rate was 28.8%.
Germany, 1934
Prices rose 2.53% after falling 2.46% the previous year. The unemployment rate was 20.5%
Netherlands, 1937
Prices rose 5.66% after falling 4.5% the previous year. The unemployment rate was 26.9%.
Norway, 1934
prices rose 1.10% after falling 2.15% the previous year. The unemployment rate was 30.7%.
In each case there was deflation when NGDP was falling and inflation when NGDP rose.
These examples are also illustrations of where what Robert Gordon calls the “rate of change effect” dominates the level effect of unemployment with respect to inflation in the Phillips Curve. There are other less extreme historical examples of “rate of change effect” dominance, namely the UK from 1870 to 1913, the US from 1871 to 1891, interwar Australia, interwar Belgium, interwar Sweden, the interwar US, and, perhaps surprisingly, the US since about 1993-98 (depending on the measure of inflation).
This also relates to whether we should expect monotonic or oscillatory convergence for inflation, something I blogged about this summer:
http://andolfatto.blogspot.com/2013/06/sadowski-on-bullard-guest-post.html
There’s so much more I could say about this issue it could make the subject for a few blog posts.
P.S. All inflation data comes from Michael Bordo, and all unemployment data comes from Galenson and Zellner (1957).
28. October 2013 at 15:04
Scott I hope you were just joking. I’m really looking forward to your book.
28. October 2013 at 15:20
Your hypothrtical depends on your income. For example if your income is 10,000 and your yearly grocery bill is 1,000 (nice easy numbers) then with a 10% across the board increase your new wages are 11,000 vs 1,100 in groceries. So your net disposable income here increased by about 900 dollars. This is why I prefer the term purchasing power since most people think of inflation as purchasing power anyways. If we had wages that increased it would fix our problems. Can you convince people that if we inflate prices and wages will rise by a uniform percentage?
Or am I off base here?
28. October 2013 at 16:42
The economics profession became peevishly fixated on inflation sometime in the 1980s…it also was an easy stance to take—no one is for too-high inflation…tough to break decades of ossifying social norms…norms that have been enshrined, exalted and institutionalized at the Fed…can we get the Fed to genuflect to NGDP?
I would bet against it…Yellin has rhapsodized about 1 percent inflation…many central bankers contend that “fighting inflation” is their only “mission and duty.”
ZLB-America for a couple decades? Maybe…
28. October 2013 at 16:55
Doug, Yes, wage inflation.
kebko, Yes, but think how much cheaper houses are today for people just out of college, getting their first house. Lower prices and lower rates.
But yes, the cost of a capital good is tricky.
Edward, Hardly anyone is on a completely fixed income. Bank accounts earn varying interest. Social security is indexed to inflation. Stocks. In any case, those on fixed incomes have recently gained at the expense of the rest of the population, it’s only fair that they give a bit back.
MikeDC, So if only 10% of people buy a car in a given year, I guess that means car prices shouldn’t be in the CPI either. I buy apples every week, so I guess apples should be a bigger share of the CPI than cars. Okaaaay . . .
But thanks for confirming my point, it’s silly to ask average people about inflation, they have no idea what it is. They think inflation is something that hurts them.
Morgan, Just a day or two ago I did a post complaining that we don’t need 6% inflation, even 2% would help a lot. I recall I criticized Rogoff.
Mark, I wish they’d think in terms of nominal shocks having real effects, rather than real shocks having nominal effects.
Benny, Yes, but in real terms your disposable income did not increase at all.
28. October 2013 at 18:43
Scott, private equity firms are buying the houses with cash and renting to those graduates with a hefty ROI. If the banks were issuing mortgages to those graduates, home prices would be back to where they were in 2004. But, I’m afraid that if they were, everyone would yell “bubble”, and the Fed would beat us down again.
28. October 2013 at 19:12
NGDP is a misleading variable.
The only non-misleading variables are the outcomes of a particular process of human interaction, which of course is private property rights, including private property rights in money.
Whatever NGDP transpires in a free market, that is what it should be, because only then would every individual have a real world active input in the production of monetary statistics. As of now, the small group of people at the Fed decide what these statistics are. They are fake and not representative of society.
28. October 2013 at 20:46
So if only 10% of people buy a car in a given year, I guess that means car prices shouldn’t be in the CPI either. I buy apples every week, so I guess apples should be a bigger share of the CPI than cars. Okaaaay . . .
No, that’s sophistry. Housing, cars, and apples should get whatever weight the market basket survey says they should get.
The current price of apples is a good approximation of what people have actually paid for their apples. The current market rental price is a much less good approximation of what people have actually paid for their housing.
28. October 2013 at 20:53
In my book on the Great Depression (which my publisher seems determined shall never see the light of day) Think positive–central banks can remain incompetent longer than publishers dither. So your book will remain topical.
And yes, some of us are really looking forward to reading it.
28. October 2013 at 20:57
Speaking of Inflation was looking at an old publication from the late 1970s a while back…and there was an ad to lease a Mercedes for $700 a month…that would be around $2500/month in today’s money…I couldn’t believe it…
Back in 1980 I bought my first computer, a TRS-80 Model III, for $2,000, and paid another $2,000 for a daisy wheel printer (like a typewriter for the kids reading this). That $4,000 was over $11,000 in today’s money — for a computing system with 48k memory, 32k disk drive and a 2mhz processor. Testament to the march of both inflation and technology.
And it was a great buy too, my second-best investment ever. I made a lot of money by being well ahead of my business peers on the computer tech curve. (Today I can just barely operate my smart phone, a little bit — my day has sadly passed.)
28. October 2013 at 21:05
The average wage and CPI inflation have had a 99+% correlation over the last 50 years, FRED tells us. Yet people constantly see the prices they pay going up *without* seeing their wages going up.
In fact, I’d say that with inflation being a general increase in the level of prices, and the price of labor being the single biggest price there is, it is definitionally impossible to have significant inflation without corresponding wage increases. Impossible in practical terms as well, as a significant general increase in prices all across the economy without any corresponding increase in wages would be a really serious price shock (with very counter-inflationary effects soon to follow.)
@John
“in the real world a 10% increase in prices will have a wide range of distributional effects that generally hurt the worst off in society.”
That’s backwards. Inflation has always been the policy of populists because it helps the masses of poor who borrow, and hurts the bankers, bond owners and other and rich people who lend. Remember right here in the US, one of the major reasons for adopting the US Constitution was to stop the state legislatures from relentlessly inflating away local debts … William Jennings Bryan campaigning to free the farmers from the Cross of Gold … during the 1970s inflation, banks and bond owners getting killed while homeowners with 6% mortgages from 1972 or earlier hit the jackpot.
@ssumner
I frequently argue that inflation is a highly misleading variable, and should be dropped from macroeconomic analysis. To replace it, we’d be better off looking at variables such as NGDP growth…
Hey, you’ve convinced me, I mostly buy it — but still, *if* the Fed had actually targeted inflation like they said they were, the whole Great Recession wouldn’t have happened or would have been a lot less bad (as David Laidler, who assisted Friedman and Schwartz with the Monetary History, noted in a recent Econtalk podcast.) They had serious deflation in 2008, and after that didn’t let the price level get back up to early 2008 levels until almost 2011. That’s not so good at hitting a 2% inflation target.
If they announce a new NGDP growth-targeting policy right now and are just as good at hitting their NGDP target, we could be in another Great Recession in 2015.
28. October 2013 at 21:08
Beyond that, the larger point is those silly “average people” are being hurt. And they’re being hurt because they face a higher cost of living. Because for most the price declines in housing are completely notional. In practice, most people didn’t get a decline in housing costs, and if they did it was a small one. On the other hand price increases in food and travel are quite real.
Telling some guy he’s not really facing inflation because the price increases he’s faced are offset by the fact his home is now underwater (and thus, someone else could buy it for less even though he can’t afford to sell it) is a dick move, plain and simple.
28. October 2013 at 21:29
Inflation is what I pay, not what I get paid. After all, people usually provide only one good or service but pay for a huge number of them. So, it is not surprising people see inflation as a rising cost of living, not as rising standard of living.
In lay terms, you stand for higher income growth. It is just that economists stick “real” in front of income. (And saying “money income” jars a bit to the lay ear.) So, speaking to your peer audience and speaking to the general populace has some disjunct.
You think expectations about income count and are a matter for central banks to manage: using NGDP makes it clear that you mean money income, not “real” income. Macroeconomists tend to obsess about keeping money “real”, so focus on expectations about the “real” value of money. In their heart of hearts, most macroeconomists would like money to just go away; to be JB Say’s “momentary function” in exchanges of goods and services. That we “really” buy goods and services with other goods and services in a can-be-analysed-as-if-it-were-completely-transparent way; as if each money transaction was just half of the “real” transaction, as Say claimed.
Hence the peer obsession with inflation-credibility and the missing-the-point on income credibility. The BoJ, the Fed, the ECB have always had heaps of credibility on the former–indeed, far too much–it is the latter that has been lacking.
But money transactions are complete transactions. That is the point of having money: it is the asset you hold to complete transactions. Money forms a “real” function; there would be a lot less transactions without it.
That’s why societies kept inventing it and why, according to AH Quiggin’s “Survey of Primitive Money” such a huge variety of things have been used as money (shells, beads, salt–including stamped salt cakes–cloth, iron and other tool metals in a wide variety of shapes, die cakes, gold dust, weighted gold, teeth, feather coils, strings of coconut discs, carved stone, animal skins, cattle, grain, pigs, coconuts, buffaloes, seeds, slaves, silver in shapes or lumps, tea, tea bricks, …). Not to mention rum (in the Sydney Cove colony), cigarettes in PoW camps and prisons, tinned fish in no-smoking prisons …
Recessions and depressions are transaction collapses. Not surprising that money is at the heart of such.
When you say money counts, you mean it rather more profoundly than most macro-economists do. As you have noted before, it is a lack of a common language problem. (I regularly come back to your “The Economics Babel” post.)
But it is not only a lack of common language among macroeconomists. It is a lack between both between MMs and other macro-economists and between economists and the general public.
When you say “inflation is a poor indicator” you mean it in a way that points directly at really fundamental differences.
29. October 2013 at 04:20
Mike, houses are a hedge against inflation, and will also correlate over time, as with wages.
29. October 2013 at 05:47
Mike,
You and I know that inflation helps the hoarder of assets and those who can borrow against such and that it harms all others.
The stagflation of the 1970s was real. Millions of living Americans can give witness of it and the economic record can be reviewed by all. But the Monetarists will insist that this inflation will be different! How so? Who knows. Just trust them!
29. October 2013 at 06:51
Scott,
If the textbooks want to talk about rising prices, they should use a different word and stop corrupting the English language. Should we say that you rising price a balloon or the universe rising priced after the big bang? Defining inflation as rising prices is utter crap and confuses people.
Lowering inflation in the early 1980s is one of many data points. You can cherry pick data points to agree with you but I believe I can find more data points where inflation is easy on the rich and hard on the poor. Besides basic reasoning would say that owners of property and companies that produce real goods would be better able to adapt to inflation than people who have to rely on pensions, socials security, or live off the savings. What if you had invested your whole life savings into a life insurance policy?
By the way, I’m sure it’s not a coincidence that the economists who become central bankers have such a strong incentive to mislead the public. The corruption of the word is a big part of the reason they can always blame someone else for their failures. Something you’ve talked about at length.
29. October 2013 at 06:52
kebko, Yes, but it’s also true that those people who are buying houses are getting them dirt cheap.
MikeDC, You said;
“The current price of apples is a good approximation of what people have actually paid for their apples. The current market rental price is a much less good approximation of what people have actually paid for their housing.”
But that’s exactly my point. The CPI is not supposed to measure what people HAVE paid for their houses–sunk costs don’t matter. It’s should measure the current market price of housing.
Jim Glass, Good point.
MikeDC, Inflation doesn’t hurt people in net terms. Every dollar spent by one person is earned by another. What hurts people are decreases in aggregate supply, or perhaps a more unequal distribution of real income. Better to keep inflation out of the picture entirely, and focus on what matters.
BTW, you ignore the fact that lots of people refinanced at lower rates. They ignore that fact when they think of higher gas prices. Cognitive bias.
Lorenzo, Very good comment.
Dan, That’s not what monetarists claim at all. We are not even targeting inflation. I expect you to come back here and admit you were wrong when the high inflation does not arrive. Only a few of my commenters from 2009 have come back and admitted they were wrong.
BTW, I predict even worse real outcomes than the 1970s, regardless of whether the Fed does monetary stimulus not. And they won’t.
29. October 2013 at 06:55
[…] Scott Sumner writes, […]
29. October 2013 at 08:33
“Oil shocks don’t tend to raise NGDP, and hence aren’t expansionary.”
I’m not so sure about this. It is obviously true away from the ZLB, because any NGDP-raising effect will be offset (and probably more than offset) by the policy response. But at the ZLB policy makers behave perversely. If oil shocks (irrationally) raise inflation expectations, they will tend to reduce the demand for base money, and at the ZLB, the central bank is not likely to fully offset this change.
29. October 2013 at 09:23
I agree with the basic premise that inflation is not per se bad. Why should an annual 4 percent inflation rate be any worse than an annual 2 percent inflation rate *if those rates are constant and predictable*?
The problem, as I see it, is inflation (or other price changes) that are erratic and unpredictable. Are these types of changes ever good (or even not bad)?
As far as historical wages keeping up with historical inflation, I guess that’s the case over time; however, I would might have some sympathy with the thinking of the “average” person whose wages are re-set annually (or less frequently). Let’s say Mr. Average has an annual salary of $100 and during the initial year price inflation is 2 percent. In year 2, Mr. Average’s wages are increased to 2 percent. Mr. Average might think that in the initial year he’s lost out on his anticipated purchasing power by 2 percent of his annual wages (or, if he’s somewhat above average in his thinking, about 1 percent). But, wait, one might say, of course Mr. Average and Mr. Employer must have anticipated some rate of inflation over the upcoming year and built that into the wage negotiation. Frankly, I don’t think for most rank and file employees this is the case, and it certainly isn’t for Social Security recipients. And, 1 percent over 1 year is no big deal. And, even if future anticipated inflation is part of the negotiation process, let’s then assume that in year 2 price inflation unexpectedly hits 10 percent. Mr. Average might be somewhat justified in thinking that he’s lost, say, 5 percent of his anticipated purchasing power for that year.
He (and social security recipients) may have their wages adjusted for *historical* inflation after one year, but are they incorrect in thinking that they’ve lost some purchasing power in the interim? Doesn’t this explain Mr. Average’s attitude towards inflation? And, isn’t it sudden and unexpected price inflation more than anything else that exacerbates this problem and adds to general disruption in an economy?
29. October 2013 at 10:25
My employer shed around 50% of the jobs in 2009-2010 timeframe. About 1/3 through a fairly generous voluntary layoff, 1/3 though attrition, and 1/3 through actual layoffs. It was not all macroeconomic related….but a good part was.
I’ve stuck it out…and it seems to me that although I have not had much in terms of wage growth since 2008…my family budget has improved considerably due to falling interest rates (mortgate refi) and now with the recovery of my 401K…I’m not waking up as many nights at 3:00 a.m.
Inflation measures have always seemed a little dubious and simplistic especially when they are at low levels < 5% to me..mainly because the contents of a family budget keeps changing rapidly, products are changing, mortgage interest rates are big portion of the typical family budget as are consumer interest rates…
Take that $700 lease from '78. Part of this is that autos only lasted 1/2 as long, part is the high consumer interest rates…and part is changing nature of world trade….
But your main basket of items you need to survive ..food (other than Ethanol related increases), shelter (monthly payments on a 30 yr.mortgate), and transportation (price on a 3 year lease) have not really increased nearly as much as the inflation rate).
I think inflation is very important..I'm just dubious of how it is measured and it's accuracy. So if you are modifying policy based on wheither it is 2% or 3%..I think that is a problem.
29. October 2013 at 12:48
Andy, I don’t think higher inflation expectations reduce the demand for money, I think higher NGDP growth expectations reduce the demand for money. Money demand stayed very high in 2008 despite high inflation, because NGDP growth (and hence interest rates) remained low.
Vivian, Unexpected inflation can absolutely be a good thing if it helps keep NGDP growth steady. Inflation doesn’t matter, only NGDP growth matters.
You said;
“He (and social security recipients) may have their wages adjusted for *historical* inflation after one year, but are they incorrect in thinking that they’ve lost some purchasing power in the interim? Doesn’t this explain Mr. Average’s attitude towards inflation?”
I don’t think so. I’m convinced that people equate “inflation” with “supply-side inflation.” Hence they think inflation is bad, when what they really should complain about is falling AS. Suppose there’s a massive supply shock because of an oil embargo, and money is kept so tight that the CPI falls by 1%. Do you think people will be happy because of the deflation? No, they’ll scream bloody murder than their living standards are falling because of high prices, and they’ll be 100% wrong. Right now inflation is 1.2% and people are complaining–much more than in the mid-1990s when it was 2% to 3%.
BTW, Anyone on Social Security right now has actually been HELPED by the adjustment lag, as inflation has fallen well under 2%. Ask them if they understand this fact.
Mike, Good point.
29. October 2013 at 14:05
” Unexpected inflation can absolutely be a good thing if it helps keep NGDP growth steady. Inflation doesn’t matter, only NGDP growth matters.”
Scott, I’m not at all convinced by conclusory statements such as the one quoted above.
As for the rest, two things:
1. In reading a number of things you’ve written here previously, I was led to think that you believe that wages tend to be “sticky” and that that somehow “matters”.
2. One of the things that smart economists might be prone to forget is that economics is largely a behavioral phenomenon. How “Mr. Average” thinks and responds to various economic stimuli has a big effect on the economy and economic cycles. While Mr. Average might be theoretically wrong about some of his perceptions about economic theory, what he does in practice as a result of those misperceptions is just as important, and cumulatively likely more important, than what Mr. Economist thinks. Being in touch with what Mr. Average thinks and how he will likely respond is not only important to politicians, but to economists as well. In other words, there may well be a very wide gulf between what you think Mr. Average should think and do and what Mr. Average actually thinks and actually does. The latter has a much larger effect on the economy than the former.
29. October 2013 at 15:06
“BTW, Anyone on Social Security right now has actually been HELPED by the adjustment lag, as inflation has fallen well under 2%. Ask them if they understand this fact.”
Scott,
I just asked a Mr. Average retiree and she does not understand the “fact” that due to inflation falling below 2 percent that she is actually HELPED by an adjustment lag. Perhaps you can explain it further?
If your employer were to give you an inflation adjusted-pension and you were given the option of having that adjustment for inflation made monthly or annually over your lifetime, which would you choose?
29. October 2013 at 16:05
Am I just stupid or should Coibion & Gorodnichenko be writing Et-1(INFt) instead of Et(INFt+1) in equations (1), (2) and so on?
More substantially, I can’t see why wage setting shouldn’t be equally or more affected than price setting if their story about the oil price and expectations is true.
29. October 2013 at 16:58
Macroeconomists need to start ignoring nominal aggregates. And they need to start focusing on the nominal aggregates that really matter, market generated NGDP and market generated nominal hourly wages. It’s the market process, stupid. (And always has been.)
29. October 2013 at 16:59
Never reason from a spending change.
30. October 2013 at 06:39
Vivian, Sticky wages and sticky pensions hurt you when inflation is more than expected, and help you when inflation is less than expected. That’s standard theory. The initial starting point is built on the assumption of average inflation (say 2%), so if it’s less you come out ahead.
I agree that what average people think is important, which is why I focus on NGDP, not inflation. When they go in radically different directions, average people act like NGDP matters. Thus wage increases and interest rates were low in 2008, because NGDP growth was low and falling, even though inflation was high.
Sorry if I was to abrupt, I’ve seen you comment here before so I thought you already knew my views on inflation.
You asked;
If your employer were to give you an inflation adjusted-pension and you were given the option of having that adjustment for inflation made monthly or annually over your lifetime, which would you choose?”
If it was monthly, you’d be forced to start at a lower level. I can collect my social security at 62, but I must take a lower amount.
Declan, I also wondered about that. Maybe someone else can help.
30. October 2013 at 07:08
“If it was monthly, you’d be forced to start at a lower level. I can collect my social security at 62, but I must take a lower amount.”
Sorry Scott but this is completely off point and non-responsive to the question. If you start at 62 your on-going annual COLA adjustment to benefits is made annually (precisely 3rd quarter over prior 3rd quarter), just like any person collecting at the normal retirement age.
So, while it should not be needed, let me restate; Assume you retire at age 66 (presumably your normal retirement age). Would you be better off thereafter to get annual inflation adjustments to your starting benefit than you would if you were allowed to get monthly inflation adjustments (known officially as COLA) adjustments? In other words, over your lifetime, which system would result in higher total benefits?
I believe this is a question of simple arithmetic.
And, as to your point about “expected inflation of 2 percent” and “sticky wages”, I don’t buy the initial premise merely because I don’t think that most wages for Mr. Average are set with *future* inflation expectations in mind but are set against the historical backdrop. But, for the sake of argument, let’s assume they do. Wouldn’t your statement support my initial point that it is *unexpected* inflation (of say, 10 percent) that is harmful to Mr. Average’s real earnings in that high inflation year?
31. October 2013 at 12:51
Vivian, We will have to agree to disagree, I believe wages do reflect expectations of NGDP growth (or inflation). And the analogy to Social Security is apt. Nothing in economics is “simple arithmatic.” Change one thing, and other things in the system also change.
As far as “Mr Average,” if a boost in aggregate demand leads to higher RGDP, then by definition average income rises. Real average hourly wages may fall, but that’s a different question.
31. October 2013 at 13:04
Thanks, Scott, but I wouldn’t classify your last comment as a “disagreement” to my rather specific question.
8. November 2013 at 09:08
Total nominal wages don’t contain any information (or at least any changing information over timescales of 50 years) — they are almost exactly proportional to NGDP (NGDP/NW ~ constant)
http://informationtransfereconomics.blogspot.com/2013/10/sticky-wages.html
The far more relevant ratio is NGDP/LS (LS = total number of employed people) which is directly connected to the CPI. Another ratio that works is NGDP/U which gets us the Phillips curve:
http://informationtransfereconomics.blogspot.com/2013/10/the-phillips-curve.html
8. April 2014 at 07:37
“”If all prices rise by 10%, and all incomes also rise by 10%, has the cost of living actually increased?” Almost everyone gets it wrong. And these are good students.”
You, I assume, are thinking nominal cost of living. Designated in dollars as the unit of account. Obviously that measure has increased.
They, I assume, are thinking real cost of living, designated in, say, work hours required to buy a basket. Hours as the unit of account. Obviously that measure has not increased.
By dismissing their answers, you are saying, effectively, “any labor-based theory of value is, prima facie, false.”
This succinctly reveals the kind of unexamined, unconscious assumptions that underly so many definitive declarative statements by so many economists.
8. April 2014 at 07:45
Just to add: since the question doesn’t mention productivity — $/hour worked — they (and I) would naturally assume that it is the same before and after. There is an implicit ceteris paribus embedded.
8. April 2014 at 07:51
Which takes me here:
http://en.wikipedia.org/wiki/Complex_question
A complex question, trick question, multiple question or plurium interrogationum (Latin, “of many questions”) is a question that has a presupposition that is complex. The presupposition is a proposition that is presumed to be acceptable to the respondent when the question is asked. The respondent becomes committed to this proposition when he gives any direct answer. The presupposition is called “complex” because it is a conjunctive proposition, a disjunctive proposition, or a conditional proposition. It could also be another type of proposition that contains some logical connective in a way that makes it have several parts that are component propositions.[1]
Complex questions can but do not have to be fallacious, as in being an informal fallacy.[1]
14. March 2020 at 04:16
[…] there is no well-defined rate of price inflation (finally Scott Sumner wins on that one), can there be a well-defined real interest rate? Is it very high or very low? Or just “a […]
14. March 2020 at 09:38
[…] there is no well-defined rate of price inflation (finally Scott Sumner wins on that one), can there be a well-defined real interest rate? Is it very […]