Scott Sumner  

A breath of fresh air from John Cochrane

PRINT
The News and Me... Sunday Morning Audio: The Germ...

Over at TheMoneyIllusion I've been running a series of posts that are critical of popular views of inflation. I claim inflation is determined by shifts in the supply and demand for money, and that factors like the Phillips curve and interest rates are not central to the inflation process. Think of money supply and demand models as fundamental, and Philips curve and interest rate models as contingent.

John Cochrane makes some similar arguments, although he uses a very different theoretical framework:

Why is it so hard? The standard story goes, as there is less "slack" in product or labor markets, there is pressure for prices and wages to go up. So it stands to perfect reason that with unemployment low and after years of tepid but steady growth, with quantitative measures of "slack" low, that inflation should rise, as Ms. Yellen's first quote opines.

That paragraph contains a classic economic fallacy, that of composition; the confusion of relative prices and the level of prices and wages overall. If labor markets get "tight," companies finding it hard to find workers, then yes, one expects wages to rise. But one expects wages to rise relative to prices. You only tempt workers to move to your company by offering them wages that allow them to buy more. Similarly, if there is strong demand for a company's products, its prices will rise. But those prices rise relative to other prices and to wages. Offering a company higher prices when its wages, costs, and competitor's prices are all rising does nothing to get it to produce more.


This is one of those cases of two things that look superficially similar but are actually radically different, like eels and snakes. When money became very tight in 1921, 1930, 1938 and 2009, the equilibrium price level fell sharply. Nominal wages also needed to adjust downwards. Unfortunately, nominal wages are sticky, so wage growth slowed much too gradually to prevent high unemployment. Thus even though nominal wage growth slowed in all four cases, real wages actually increased sharply. Cochrane's right that the wage changes we see in this sort of labor market have nothing to do with microeconomic models where a high level of demand means rising prices and a low level of demand means falling prices. Those micro models refer to real or relative prices, not nominal prices.

So why do wages often rise quickly during "tight labor markets"? One reason is that because wages are sticky, when they are rising they (paradoxically) tend to be too low, and when they are falling they tend to be too high. Suppose 10% of worker contracts are adjusted to equilibrium each month. Then if a sudden monetary shock causes the equilibrium wage to immediately rise by 4%, wages would rise 0.4% after one month, 0.8% after two months, and reach the new equilibrium after 10 months. Again, when you see wage growth accelerating then wages are often too low, and when you see wages falling they are often too high.

Now we can see the connection between rising wages and tight labor markets. When wages are rising they are too low---and those excessively low real wages cause companies to want to hire more workers---hence the high level of employment. It has nothing to do with tight labor markets causing higher wages---as Cochrane says that's an example of the fallacy of composition. And that's why it's not a reliable model of inflation--it's not a causal factor.

There are many more gems:

By the way, the oft-repeated mantra that "inflation expectations are anchored" offers no solace. In fact, it makes the puzzle worse. The standard Phillips curve says inflation = expected inflation - (constant) x unemployment. Variation in expected inflation is usually an excuse for a Phillips curve failure. Steady expected inflation means the Phillips cure should work better!
And this:
Is policy tight or loose right now?

You'd think this were an easy question. The newspapers ring with "years of extraordinary stimulus" and "unusually low rates." And indeed, interest rates are low by historical standards, and relative to rules such as John Taylor's that summarize the successful parts of that history.

But ponder this. What does a central bank look like that is holding interest rates down? Well, it would be lending out a lot of money to banks, who would turn around and re-lend that money at higher interest rates. What does our central bank look like? Our central bank is taking in $2.2 trillion from banks, and is paying them a higher interest rate than they can get elsewhere. Right now, the Fed is paying banks 1.25% on their reserves. But Treasury bills are 1%. Even commercial paper is 1.13-1.2%. It looks every bit like a bank that is pushing rates up. And has been doing so for a long time.


It's good to finally see a prominent economist point out that monetary policy has not been easy, a point I've made about 1000 times since 2008.

Cochrane ends with this:

If you just plot inflation and interest rates, they seem to move together positively. Teasing out the notion that higher rates lower inflation from that graph takes a lot of work. My best guess, merging theory and empirical work, is that higher rates -- moved on their own, not in response to economic events -- temporarily lower inflation, but then if you stick with higher rates, inflation eventually rises. And vice versa, which accounts for very low inflation after interest rates have been stuck low for a long time. Maybe yes, maybe no, but even this much is not certain.
Finally I've found something to criticize. This paragraph perfectly encapsulates what's right and what's wrong with NeoFisherism. Cochrane is right that higher rates often are associated with lower inflation in the short run, but higher inflation in the long run. But he stills falls a bit short in my view.

Let me start by quibbling over a minor point---Cochrane's reference to interest rates moving "on their own". I think I know what Cochrane means, but I'm going to throw a temper tantrum anyway, and then explain why.

Interest rates never move around on their own, as the economy is not some sort of Ouija board where things happen "on their own" without there being an "economic event", to use Cochrane's terminology. Now I'm pretty sure that what Cochrane meant by "on their own" was a change in the interest rate caused by monetary policy. But that doesn't help as much as you might assume.

Screen Shot 2017-10-21 at 9.24.44 PM.png
When Cochrane said that higher interest rates may initially reduce inflation, he probably had in mind a contractionary monetary policy. But in that case it's not the higher interest rates that lower inflation, it's the monetary policy. Thus a sudden decrease in the monetary base is contractionary, and may lead to both higher interest rates in the short run and lower inflation. Ditto for a higher interest rate on reserves. (The interest rate paid on reserves is not really a market interest rate, it is an administered price (subsidy).) For any given monetary base, a higher fed funds rate is inflationary, as it boosts velocity (assuming no IOR).

When Cochrane refers to a policy of "sticking with higher rates" he runs into problems. As noted, he seemed to start by considering a contractionary monetary policy that led to higher interest rates in the short run. But if you "stick with" a contractionary monetary policy then rates will end up lower over time. It makes no sense to talk about "sticking with" higher interest rates, because interest rates are not a policy, they are the effect of various other policies.

To get higher rates to persist you'd need to switch from a very contractionary to a very expansionary policy. But that switch will often temporarily depress interest rates, before causing them to rise. A good example occurred in 1967, when rates temporarily fell (easy money) before rising to semi-permanently higher levels during the Great Inflation of 1966-81. That persistent inflation is what NeoFisherians have in mind when they equate a high interest rate policy with high inflation. It's a valid point, and an important critique of Keynesian economics. But unless and until the NeoFisherians fully incorporate both the liquidity effect and the Fisher effect into their models, the analysis will remain frustratingly incomplete. And to do that we need to return to monetarist economics, something neither Cochrane nor his New Keynesian critics seem to have any interest in doing.

HT: Tyler Cowen


Comments and Sharing






COMMENTS (5 to date)
Benoit Essiambre writes:

>"That paragraph contains a classic economic fallacy, that of composition; the confusion of relative prices and the level of prices and wages overall. If labor markets get "tight," companies finding it hard to find workers, then yes, one expects wages to rise. But one expects wages to rise relative to prices. You only tempt workers to move to your company by offering them wages that allow them to buy more."

This seems wrong to me. You don't need to offer workers more real wages, you only have to offer more than other businesses offer, or if you are giving wage increases to existing employees, more than the counterfactual where you wouldn't have been able to sell at higher prices or wouldn't have been able to get easier financing. I don't really see a fallacy of composition here and I'm really trying to think on the aggregate though I may be failing at it.


"Similarly, if there is strong demand for a company's products, its prices will rise. But those prices rise relative to other prices and to wages. Offering a company higher prices when its wages, costs, and competitor's prices are all rising does nothing to get it to produce more."

That is not true either. All prices can rise in concert except the real cost of financing and the company will hire more people and produce more. Prices rising in concert also tends to lower the real cost of financing.

The cost of capital calculation seems to me to be crucial here. Maybe it's because I'm an engineer and the only economics course I ever took was "economics for engineering" and the whole course was about doing cost of capital calculations where you started with financing terms and tried to determine which projects were worth doing given other costs, prices and risk. The whole course was about calculating which bridge is worth building with (i) being the governing variable in all equations. For businesses that do long term investment projects, real interest rates are at the root of the decisions to start more projects or not. This affects all businesses and is regulated by central banks, where is the fallacy of composition here?

The lever that central banks directly have their hands on is the one of financing. That is why they are called "banks" and not something like "money factories".

What I am not getting? It is unsettling to me to see smart people easily step over these holes. I am tempted to just assume I took a wrong turn somewhere but I don't understand where. argh!

Ram writes:

If you add a money equation to the New Keynesian model, you can easily see where NeoFisherism goes wrong:

m - p = (a * y) - (b * i) + e, a > 0, b > 0

where m is log(money supply), p is log(price level), y is log(real output), and i is a short-term nominal interest rate.

"Raising interest rates", in the conventional sense, corresponds to temporarily reducing m. With sticky prices, some of this shows up as lower y, some as higher i, and some as lower p. This leads to a short-term correlation between higher i and lower p.

"Raising interest rates", in the NeoFisherian sense, corresponds to permanently raising i. With fixed m, some of this shows up as higher y, and some as higher p. This leads to a long-term correlation between higher i and higher p.

The problem with conflating the two is that "raising interest rates" in the first sense actually *lowers* i in the long run (due to reduced inflation expectations). To permanently raise i, you need to increase expected inflation, which means you need to "lower interest rates" in the first sense.

The difference in perspectives is the role of the inflation target. In the first sense of "raising interest rates", we *pursue* a preset inflation target by adjusting our interest rate target. In the second sense, we *adjust* the inflation target by adjusting the interest rate target.

I don't really have an opinion on whose language is better, but I think conflating the two is where the surprise of NeoFisherism comes from, and the money equation is useful for clarifying this.

Jose writes:

Great post, Prof. Sumner, you make monetary analysis seem so easy.

Do you think that with many developing countries growing fast and wealth increasing as well the ratio ( financial wealth / base money ) matters as something to be tracked ?

Scott Sumner writes:

Benoit, You said:

"This seems wrong to me. You don't need to offer workers more real wages, you only have to offer more than other businesses offer, or if you are giving wage increases to existing employees, more than the counterfactual where you wouldn't have been able to sell at higher prices or wouldn't have been able to get easier financing. I don't really see a fallacy of composition here and I'm really trying to think on the aggregate though I may be failing at it."

Offering more than other firms is equivalent to offering higher real wages. The point is that this argument applies to real variables, not nominal.

You said:

"That is not true either. All prices can rise in concert except the real cost of financing and the company will hire more people and produce more."

This is also false. Consider a currency reform where all prices fall by 99% overnight. That does not provide a boost to output, nor does it affect the real cost of financing. If money is neutral then it should not affect output.

You said:

"The lever that central banks directly have their hands on is the one of financing. That is why they are called "banks" and not something like "money factories"."

No, central banks do not control the long term real interest rate, which is the cost of financing. They actually should be called "money factories", as their role has nothing to do with banking.

Ram, I agree that NoeFisherism is wrong, but no more wrong than conventional Keynesian monetary theory.

You said:

"To permanently raise i, you need to increase expected inflation, which means you need to "lower interest rates" in the first sense."

I wouldn't say you "need" to cut rates, I'd say you often cut rates when easing monetary policy. But sometimes easy money leads to higher rates, even in the short run.

Jose, Thanks, I'm not sure--how do you think it matters?

Jose writes:

Maybe as people get wealthier around the globe they end up holding larger cash balances. So, in order to grasp if a central bank balance sheet is large or small one should at least look at the ratio financial assets / base money.

POST A COMMENT




Return to top