Why sticky wages matter
Based on reaction to my previous sticky-wage post, it might be useful to consider how sticky wages operate at the industry level. For simplicity, assume a unit elastic demand for cars (nothing crucial will hinge on that assumption—it makes the math easier.)
Now suppose that Americans decide they want to spend 10% more on cars in 2012 than 2011. Because demand is unit elastic, the demand curve will shift right (and upward) by 10%. How will automakers react to this increase in demand? You can construct plausible scenarios where there is no impact on unit sales. Suppose we were at full employment, and both wages and prices had been trending upward at a 10% rate (like 1979.) In that case car companies might react to the extra nominal demand by merely selling the same number of cars at a 10% higher price.
But suppose the situation is more like America circa 2012. Lots of slack in the economy. Wage contracts reflecting very low expected NGDP growth. Nominal wages are highly sticky. In my view under current conditions the automakers would respond to 10% more demand for cars by selling lots more (say 7% or 8% more) and raising prices only slightly (say 2% or 3%.) And I think sticky wages are one reason why this sort of “nominal spending shock” in the auto industry would boost output, although not the only reason.
The evidence that nominal wages are sticky in the short run is so overwhelming that I’d like to take it as a given. Instead, let’s consider various possibilities for how car prices would react to more demand:
1. Suppose the auto industry is perfectly competitive and sticky wages are a big part of costs (including wages for the making of components.) With sticky wages the MC curve will shift upward by less than the demand curve, and equilibrium output will rise. Prices will rise somewhat, and the pseudo “real wage,” i.e. wages/car prices, will fall. This is the classic case of wage stickiness leading to counter-cyclical real wages, as companies move along a stable labor demand curve.
2. Suppose the auto industry is monopolistically competitive. In that case prices might be very sticky. It’s possible that both wages and prices don’t change, or both might rise by something like 2%. Again a 10% nominal spending shock will cause much higher real output in the car industry, but in this case the “real wage” in terms of auto prices will be roughly unchanged, and this will seem to refute the sticky-wage model of the business cycle. In fact, all it will really show is that the perfect competition model doesn’t fit the auto industry. Back in 1921 when the US economy was somewhat closer to perfect competition, real wages rose sharply when NGDP fell.
The bottom line is that the sticky wage model that relies on W/NGDP is almost always a good explanation of fluctuations in employment. Because nominal wages per hour are sticky, when nominal spending falls sharply we can expect hours worked to fall. This is one of the most reliable regularities in macroeconomics. But we can’t go much beyond that. Theories that try to explain fluctuations in terms of real wages (W/P) will work well in some settings, and not others. It might work in perfect competition, but not monopolistic competition. It will depend on whether prices are also sticky, and on how you measure things like housing prices. The official CPI numbers show housing prices up 7.5% in 5 years, whereas Case-Shiller has them down 32% over 5 years. Housing is a third of the CPI, so it makes a big difference which housing numbers you use. On the other hand NGDP is more cut and dried—just the dollar value of expenditure on final goods–no assumptions about prices are necessary.
Nominal GDP shocks aren’t the only factor that explains business cycles, but I believe they explain the majority of the cyclical fluctuations. Because of sticky wages, a stable path for NGDP would make RGDP growth more stable, and would also make financial crises less severe.
Tags:
22. February 2012 at 11:38
Suppose wages aren’t sticky at all, and there is a negative NGDP shock. Now what happens?
22. February 2012 at 13:17
TA, I believe wages and prices fall to a new equilibrium and employment remains unchanged.
22. February 2012 at 13:22
If the share of NGDPPC going to wages has been relatively constant, shouldn’t the Price Markup over Unit Labor Costs be falling over time? If labor captures a consistent proportion of real growth, real wages should be growing.
If W/NGDP was what was expected to stay constant, sticky wages can explain unemployment: NGDP was 9% lower than expected but wages are sticky so W/NGDP suddenly shot up and hours worked needed to fall. In this case you could also see W/P fall if prices rose more quickly than wages.
But if it’s W/P that is expected to remain constant, the story doesn’t work. Prices continued to rise so W/P actually fell, so if wages are sticky why wouldn’t employment rise?
Maybe I’m not understanding something, but it looks like the question is whether W/P or W/NGDP(PC) makes more sense to look at as the factor that’s expected to remain roughly constant. If so, while W/NGDP sounds like a better story, the P/W graph seems to contradict it.
22. February 2012 at 14:05
The evidence that nominal wages are sticky in the short run is so overwhelming that I’d like to take it as a given.
Without mentioning that they are sticky due to a combination of pro-union legislation, minimum wage laws, taxpayer financed unemployment insurance, and forming price inflation expectations that are not conducive to a monetary deflation period?
Back in 1921 when the US economy was somewhat closer to perfect competition, real wages rose sharply when NGDP fell.
Nominal wages fell something like 20% in the depression of 1920-1921. If the Fed didn’t reinject the economy full of funny money, nominal wages would have fallen more to eliminate unemployment. Monetary policy to raise NGDP is not only not necessary, but is positively destructive (since it sets the economy up on another false boom and inevitable bust later on).
Again a 10% nominal spending shock will cause much higher real output in the car industry, but in this case the “real wage” in terms of auto prices will be roughly unchanged, and this will seem to refute the sticky-wage model of the business cycle.
And why would nominal spending suddenly collapse by 10%, even though the Fed didn’t stop printing money?
Crickets…
Because nominal wages per hour are sticky, when nominal spending falls sharply we can expect hours worked to fall. This is one of the most reliable regularities in macroeconomics.
You’re mistaking correlation for causation.
It’s not the case that NGDP falls and then wages fall as if the former causes the latter. NGDP and wages fall together, for the same prior cause that you’re not addressing. And no, it’s not because the Fed isn’t printing enough money. It falls because the allocation of labor and capital resources are NOT where the voluntary consumer wants them, both cross sectionally and in terms of their temporal projections of completion.
Nominal GDP shocks aren’t the only factor that explains business cycles, but I believe they explain the majority of the cyclical fluctuations. Because of sticky wages, a stable path for NGDP would make RGDP growth more stable, and would also make financial crises less severe.
It would make them as severe as would result from the amount of inflation necessary to keep NGDP growing at a 5% clip, and in period of correction, when prices and wages and costs would otherwise fall, and when purchasing power of cash balances would otherwise rise, targeting NGDP will only exacerbate the very problems that targeting NGDP caused.
22. February 2012 at 14:07
Hi!
I would really like to hear about that evidence on sticky short-term prices, since most econbloggers who assume sticky prices rarely bother to write about it.
Another blogpost or a link-or-two to the relevant books or studies will be appreciated.
Thank you 🙂
22. February 2012 at 14:09
From the firm’s perspective, Ngdp is simply their total current or expect revenue from output… Right? But how do you know that wages a set based on expected ngdp and not prices?
22. February 2012 at 16:07
Talking NGDP around an Austrian is like going to Cape Cod during greenhead season. Swat! Swat!
22. February 2012 at 16:15
Ok Major Freedom, we get it. You don’t like Sumner’s economics. We’re all very clear on that by now. But do you really need to come here and piss on every post Scott makes? At what point does your commenting here border on obsessive?
You and Greg Ransom are creepers.
22. February 2012 at 16:52
Ok CA, we get it. You like Sumner’s economics. We’re all very clear on that by now. But do you really need to come here and fawn over every post Scott makes? At what point does your commenting here border on obsessive?
You are a creeper.
22. February 2012 at 17:15
“Nominal GDP shocks aren’t the only factor that explains business cycles, but I believe they explain the majority of the cyclical fluctuations. Because of sticky wages, a stable path for NGDP would make RGDP growth more stable, and would also make financial crises less severe”.
More “evidence” that it does:
http://thefaintofheart.wordpress.com/2012/02/22/the-60s-x-the-90s-golden-age-x-great-moderation-who-wins-the-economic-gold-i-could-also-call-the-dispute-obsession-w/
22. February 2012 at 17:29
ja, “sticky wages” is precisely the statement that wages do NOT hold a constant ratio with other elements of the economy. Wages “fall out of sync” whenever there’s a shock.
Wages adjust slower than almost everything else. NGDP changes much faster than hourly wages. That’s why we talk about “sticky wages”, because the wages are sticky even when the rest of the economy isn’t.
“Wages are sticky” is a short way to say, “a big change in prices or output produces a surprisingly small initial change in hourly wages, especially if the direction of the change is downward.”
Over the long term, wages often do catch up to prices or output. But not always. In particular, over a span of a single recession, wages will usually adjust much more slowly than prices or output. The gap is made up by a change in unemployment or inflation.
Our current rising price/labor markup is just what you’d expect with unemployment high and inflation low.
22. February 2012 at 18:57
Not that it matters too much for the examples you are using, but you might want to say “downward nominal wage stickiness” rather than just nominal wage stickiness. Both are sticky in the very short run, but from a behavioral and theoretical perspective at least, wages should respond to above average NGDP growth more quickly than to below average NGDP growth.
I think the shifting Phillips curve in the 1980’s is pretty consistent with that concept. It is also consistent with a lot of survey data asking people about how they would feel about nominal wage cuts when inflation is much lower than expected vs nominal wage increases that fail to keep up with inflation. People consistently respond more negatively to the form than the latter, even if their real wage is rising in the face of the nominal cut, or rising in the face of the nominal raise.
Just some food for thought though.
22. February 2012 at 19:38
Two errors in the last paragraph there.
1. It is also consistent with a lot of survey data asking people about how they would feel about nominal wage cuts when inflation/NGDP growth is much lower than expected vs nominal wage increases that fail to keep up with EXPECTED inflation/NGDP growth.
2. People consistently respond more negatively to the former than the latter, even if their real wage is rising in the face of the nominal cut, or FALLING in the face of the nominal raise.
Sorry for cluttering the comments with a second post. I always lament the lack of an edit function for comments.
23. February 2012 at 10:50
Any sticky variable will do:
One thing that bothers me is that although I competely agree with this analysis, I hardly think that fully downward-flexible prices and wages would remedy this problem. What about asset prices and debt contracts? Because those are also sticky, or denominated in fixed nominal terms, ideally if you had downward wage flexibility you would also want the debt burdens of households to fall in step with their income.
Although on the real product/output side of things flexible wages improves the situation, from a household balance sheet perspective, it almost exacerbates them. I have a hard time swallowing the argument that increasing real debt burdens will lead to a more robust recovery.
To me it seems like you need to make all variables equally “mobile” which is why inflation (which would erode all nominally denominated contracts) is seriously preferable to increasing wage flexibility. In many ways, elevated inflation is the most fair/equitable way to achieve this same outcome. Shame on the Fed.
23. February 2012 at 11:16
A recession is really just about one sticky price–the price of money. The price of money is very sticky, and it’s not because of any psychological mumbo-jumbo or money illusion. The price of money is sticky because it doesn’t have a single unambiguous price at all–every other price is a price of money. Therefore, a change in the price of money is rather more roundabout and momentous than a change in any other price.
No individual experiences the shortage of money directly, but rather just a fall in demand for whatever they happen to sell. While it would benefit everyone to reduce their prices in unison by the same proportion, the decentralised character of market prices means that they must grope toward the new equilibrium.
There is a who-goes-first-problem with an excess demand for money. Whoever moves first may actually be at a disadvantage in the short-run, since they are selling at the new lower price but must still buy at the old higher prices until everyone else catches up. In other words, it may not matter how many more widgets are being sold if they are forced to sell each at a loss because other prices have yet to fall.
23. February 2012 at 13:46
[…] Sumner on sticky wages. […]
23. February 2012 at 15:05
TA, It depends whether prices are sticky. If not, lower NGDP simply reduces prices. A good example is a 100 to 1 currency reform. Both wages and prices are completely flexible, and both NGDP and the price level immediately fall by 99%.
ja, I’m not sure what you are arguing. Kling did not provide any evidence of P/W.
Major Freeman, Yes, wages fell 20% in 1921, but prices fell even more.
I said nominal spending rose by 10%, not fell.
catbert, Maybe someone else can help you.
Joe, The data I look at says wages are more correlated with NGDP, at least when there are supply shocks. Wages didn’t rise sharply when inflation rose due to the 2008 supply shock. Wages in China follow NGDP much more closely than inflation (I mention China because it’s a country where NGDP and inflation are vastly different.)
Steve, Yes, But Hayek favored NGDP targeting.
CA, Don’t worry, I’m helped by the (weak) arguments made by my opponents.
Marcus, Thanks, good post.
Blake. That’s right, I agree but I sometimes forget to mention that distinction.
Shocking, You said; I agree. I favor stable NGDP growth, rather than a futile attempt to achieve stability by adjusting wages.
Lee, Good point.
23. February 2012 at 16:12
The Reserve Bank of Australia’s Governor just got ask about NGDP targeting in Parliamentary testimony! Will post the link when the transcript is posted…
23. February 2012 at 17:27
Lee, it’s much more complex that than.
There is no one “money” — there are many kinds of money and near monies and substitutes for money and shadow monies and things the provide money services — and various things that shift value across time like money debts.
See Keynes’s Treatise for some examples.
Therefore, a change in the size and value of all these different kinds of money and near monies and financial assets and debt obligations is rather more roundabout and momentous than a change in the size and value of any other kinds of things.
23. February 2012 at 19:42
OT, but the right-wing os often very concerned about tax rates the richest people pay. Fair enough, we all have our concerns, and a nation does need capital for business expansion.
We Market Monetarists have made slim headway on the right, due to their extreme fixation on inflation, and even an increasing genuflection to gold (don’t worry, the left is full of economic dopes too).
Okay, so how about this argument: QE monetizes debt, and thus cuts income taxes that rich people have to pay. If we would monetize about 25 percent of the national debt (through QE) that would remove a huge IOU that rests on the shoulders of the wealthy, and thus they would increase their propensity to invest and also enjoy life (since they are represented as unusually miserable).
Of course, the develeraged economy and higher growth and inflation rates would benefit everyone else immensely as well. But in today’s times, the right-wing only buys arguments that are shown to favor the rich.
23. February 2012 at 20:39
Scott,
I recently discovered a beer called “Steel Reserve”. I’m obviously out of the loop because it has existed since about 1990. It is typically sold in 24 ounce cans (in the East) and has an alcohol content of 8.1%. It also does not have the metallic taste that is common with other cheap weaker brews such as “AB Natural Ice” or “Pabst Piels”.
Well the bottom line is it costs less per ounce of alcohol on a per unit basis than Vodka up to a 375ml unit sale. Moreover it tastes much better.
Well in my home town (Hockessin, DE) it goes for $1.65 a can. In Newark, DE (8 miles from my house and home of the U of DE) it goes for $1 a can.
My cars get about 8-11 MPG on premium which runs about $4 a gallon these days.
The bottom line is I have to schedule my beer purchases acordingly.
You can categorize this under the “sticky alcohol and gasoline consumption” category.
Thanks,
Mark
23. February 2012 at 20:56
Greg,
Even the nearest of near monies is still bought and sold in money. This is not true of money itself (except in currency exchange markets). Near monies have prices of their own that can change relative to everything else, like an interest rate. While changes in the supply of and demand for money near monies can effect the demand for money proper, it is only when such an effect leads to monetary disequilibrium that the shit hits the fan.
23. February 2012 at 20:56
Mark,
Let me introduce you to:
http://homedistiller.org/forum/viewtopic.php?f=14&t=9981
Read carefully, follow along – this recipe will un-stick your prices, and make you a Vodka snob.
Productivity gains everywhere.
23. February 2012 at 20:59
Scott, well said.
23. February 2012 at 20:59
Mark S.–
The guys near my shop buy Steel Reserve 24-packs. I wonder who makes the stuff? Good investment?
My other investment idea is to open up a bar where Chinese tourists like to go.
Perhaps stocked with Steel Reserve.
23. February 2012 at 21:06
Morgan,
Truthfully, I am a Vodka snob. A colleague of mine knew as much and bought a 1.75L bottle of Belvedere recently which I managed to finish nearly all by myself in one night.(I warned him before hand.) It was like water to me.
Clearer and purer than water. And more purifying still.
23. February 2012 at 21:10
Benjamin Cole,
According to my local liquor shop they can hardly keep the stuff in stock.
My advice is invest.
23. February 2012 at 21:36
Give it a shot Mark, 185 proof neutral spirit cut 40% with distilled water is nearly tasteless. Just burns like glass.
How I got into this hobby, I saw a taste test where Popov was filtered four times through a Brita filter and people chose it blind over G. Goose. Googling led me to that site. It’s a bunch of chem geeks pretending to be hillbillies.
23. February 2012 at 21:56
Morgan,
I’m a hillbillie pretending to be a chem geek so I guess its only fair.
Mark
23. February 2012 at 22:13
Morgan,
Let me reinforce my hillbillie credentials.
I spent a night with a Slavic grandmother. She knew the recipe. We drank it. The next day the only people out of our whole crowd (numerous) that woke up in our own beds were my father and myself.
I don’t remember how I got to my own bed but I’m sure my father was responsible.
P.S. No one knows where my mother spent the night.
23. February 2012 at 22:38
Let me be clear. My father and I had no intimate relations (Muy Tata?). Nor do I believe my mother engaged in adulturous relations. The point was we simply don’t rememember what happened that night. Watch out for that Slavic Rocket Fuel.
23. February 2012 at 22:39
Kelly,
The money in your pocket is different than the money reserves institutions have in the Federal Reserve which is different than the money I have at Bank of America, etc.
I’m talking about different kinds of money money as well as all sorts of inter-relations between different sorts of money and various sorts of near monies and much also.
And note well:
There are _always_ money disequilibriums of one sort or another in the market.
I simply don’t buy this the premises built into this picture:
Kelly writes,
“Even the nearest of near monies is still bought and sold in money. This is not true of money itself (except in currency exchange markets). Near monies have prices of their own that can change relative to everything else, like an interest rate. While changes in the supply of and demand for money near monies can effect the demand for money proper, it is only when such an effect leads to monetary disequilibrium that the shit hits the fan.”
24. February 2012 at 05:25
The WSJ is blaming monetary policy for the rise in oil prices. In an unintentionally ironic twist, this opinion piece is titled “‘Stupid’ and Oil Prices.”
http://online.wsj.com/article/SB10001424052970203918304577241623995642182.html
Fed officials and Mr. Obama want to take credit for easy money if stock-market and housing prices rise, but then deny any responsibility if commodity prices rise too, causing food and energy prices to soar for consumers. They can’t have it both ways, as not-so-stupid Americans intuitively understand when they buy groceries or gas. This is the double-edged sword of an economic recovery “built to last” on easy money rather than on sound fiscal and regulatory policies.
24. February 2012 at 06:46
Sorry Steve!
Peggy Noonan knocks it out of the park!
“Mr. Obama yesterday blamed rising demand from the likes of Brazil and China, and there is something to that as well. But this energy demand is also not new, and if anything Chinese and Brazilian economic growth has been slowing in recent months.
Another suspect””one Mr. Obama doesn’t like to mention””is U.S. monetary policy. Oil is traded in dollars, and its price therefore rises when the value of the dollar falls, all else being equal. The Federal Reserve throughout Mr. Obama’s term has pursued the easiest monetary policy in modern times, expressly to revive the housing market. It has done so with the private support and urging of the White House and through Mr. Obama’s appointees who are now a majority on the Fed’s Board of Governors.
Oil staged its last price surge along with other commodity prices when the Fed revved up its second burst of “quantitative easing” in 2010-2011. Prices stabilized when QE2 ended. But in recent months the Fed has again signaled its commitment to near-zero interest rates first through 2013, and recently through 2014. Commodity prices, including oil, have since begun another surge, and hedge funds have begun to bet on commodity plays again. John Paulson says he’s betting on gold, the ultimate hedge against a falling dollar.
Fed officials and Mr. Obama want to take credit for easy money if stock-market and housing prices rise, but then deny any responsibility if commodity prices rise too, causing food and energy prices to soar for consumers. They can’t have it both ways, as not-so-stupid Americans intuitively understand when they buy groceries or gas. This is the double-edged sword of an economic recovery “built to last” on easy money rather than on sound fiscal and regulatory policies.”
24. February 2012 at 06:48
drudge lied to me, that’s not Peggy. God, I love her.
24. February 2012 at 09:04
Thanks Antipodes.
Ben, Yes, I sometimes forget to mention that monetary stimulus now would reduce future tax liabilities.
Good luck Mark.
Steve, That’s typical of them.
25. February 2012 at 18:23
Ransom:
Commerical paper is a near money. It has a market price and a yield. Now, is there any guarantee that the particular project a firm funds with its commercial paper will be profitable? Of course not. For example, they may be planning to produce an good that will become obsolete becaues a new product will be introduced which is better for many buyers.
Or, perhaps their product will require the use of a complementary product whose supply will be disrupted because of a natural disaster.
And on, and on.
It is possible that they plan to sell a capital good to some firm that has a project that will only come to fruition in the more distant future, and that buyer ends up not buying because some other project with a payoff in the less distant future looks more profitable.
Nothing can avoid these errors. Even so, the commerical paper market clears. And these mistakes are not due to the use of commerical paper for financing.
If firms fund their investments with equity–shares of stock–and some firm makes a bad investment, well, there is a bad investment.
Since equity is perpetual, does that mean that projects funded with equity never have to come to fuition? No.
Your examples of money were all media of exchange: Currency, reserve balances, and checkable deposits at Bank of America.
Shortages of currency matching surpluses of reserve balances are quickly corrected. Surpluses or shortages of checkable deposits issued by any one bank are easily corrected. If a central bank creates an excess supply of base money, it soon generates an excess supply of all checkable deposits as well.
If a single bank raises the interest rate it pays on deposits, monetary or not, and this raises the demand to hold these deposits, and the bank then lends the funds out to some firm or other that borrows to fund a project that will not pay off, it has not created monetary disequilibirum. The other banks, who lost market share, lend less. There has been a shift of resources and, by assumption, a malinvestment. But focusing on the money and the bank is pointless.
If there is a decrease in the demand to hold money, and an increase in the demand for a near money, like commercial paper, then there is no direct process by which the quantity of money proper will directly decrease. People will accept the money even though they don’t want to hold less money. That is what makes it true money.
The problem isn’t with the market for the near money, the problem is with the market for the media of exchange.
If the demand for base money is positively related to nominal expenditure on output, (which is likely for a variety of reasons,) then the result in a shortage of base money and so the banks are forced to contract the amount of deposits issued to match the demand for them, bringing nominal expenditure back to equilibrium. Then, fewer projects are funded by bank loans funded by bank money and more projects are funded by “near money,” like commercial paper.
In the end, the problem is with excess supplies or demands for base money.
27. February 2012 at 15:08
ssumner:
Major Freeman, Yes, wages fell 20% in 1921, but prices fell even more.
Which prices? Asset prices? Of course. They were previously bid up faster than consumer prices. Of course they have further to fall during a correction.
Without inflation, there would have been a further correction and at some point, a cessation of falling prices.
28. February 2012 at 05:35
Major, The WPI.
28. February 2012 at 11:45
Major, The WPI.
Yes, and if the WPI fell, and if wages fell, then business costs fall as well, yes? By the same amount that wages and the WPI fell?
Wage earners don’t buy from the WPI anyway, so it doesn’t matter that wages take up more business costs and WPI related goods take up less business costs. Since both were falling, it means business costs were falling, and so there is no reason to advocate for inflation to raise the demand for goods as such.
1. March 2012 at 12:01
The problem is that wages fell more slowly than prices and NGDP, so unemployment rose.
1. March 2012 at 22:21
ssumner:
The problem is that wages fell more slowly than prices and NGDP, so unemployment rose.
Prices AND NGDP? Love it how you sneaked in NGDP like that. Hilarious.
It can’t be both prices and NGDP. It’s got to be one or the other.
Wages fell more slowly than prices? Over what time period? If we had a free market past 1922, and the Fed did not reinflate, then wages would have kept falling relative to prices, and unemployment would have been eliminated the old fashioned way. Instead, the Fed reinflated the money supply, which pushed down wages artificially, and the result was the roaring 20s inflationary (of the money supply, check the aggregate money stocks) boom resulted, culminating in the end of decade crash.