Scott Sumner  

There are two kinds of people . . .

Adam Smith on Stationary Bandi... What Is Tyler Smoking?...

Ideology is perhaps the most common way of dividing economists into groups. You have Allan Meltzer on the right, and Paul Krugman and Joe Stiglitz on the left. Today I'd like to argue that in one very important way it is Krugman and Meltzer who should be grouped together, and Stiglitz placed in a separate category.

Back in 1988 Meltzer wrote a book on Keynes. In a footnote (the most important part of the book, BTW) Meltzer wondered why Keynes didn't recommend an inflation target high enough to prevent a liquidity trap. After all, if the trend rate of inflation were high enough, then nominal interest rates would never fall to zero. In that case, the central bank could stabilize the economy, and fiscal policy wouldn't be needed. The entire General Theory would be superfluous.

At about this time, the Keynesian community began to assume that inflation was high enough that we didn't need to worry about monetary policy ineffectiveness, and "new Keynesianism" was born. A few years later Japan hit the zero bound, and Paul Krugman began worrying that it could happen here as well. When it did happen here, Krugman made essentially the same point as Meltzer, why not raise the inflation target high enough so you don't face the zero bound problem? So did Brad DeLong. In that case the central bank could stabilize the economy, and there'd be no more "Great Recessions."

I believe Meltzer and Krugman and DeLong are right that a high enough inflation target would prevent the zero bound problem. But I also believe that all four of us are very much in the minority. Most economists don't think in nominal terms, they think in real terms. Stiglitz is much more representative of the profession as a whole. They think in terms of bank loans, demographics, inequality, deregulation, investment, and a million other factors that might impact the credit market, but have no impact on demand if the central bank is inflation targeting. In the majority view, monetary policy is not the all important policy tool that steers the nominal economy, but rather just one of many factors to put into the equation, to explain changes in "aggregate demand." They might see declining Japanese population as something that reduces aggregate demand, whereas to me, it obviously reduces aggregate supply.

Why do so many people overlook these strange affinities and differences? I think one reason is that when monetary policy does fail, and we end up at the zero bound, people like Paul Krugman and Brad DeLong start saying things that sound a lot like what Joe Stiglitz says. Paradox of thrift, fiscal stimulus, etc., etc. In contrast, Meltzer continues to assume that monetary policy steers the nominal economy. So we put them in separate boxes.

When I read other economists I immediately put them into one of two groups---people who think in terms of nominal shocks when discussing AD, and people who think in terms of real factors. Like most economists, I arrogantly assume that I am right, and hence I believe that the nominal people like Meltzer, Krugman, DeLong and me are a sort of "inner sanctum" who have access to the "secrets of the temple." The rest are just like average people, who think in real terms. Too many houses were built in 2006, or not enough Japanese babies were born in recent decades---that sort of thing.

Real AD? That term makes my hair hurt. When the AS curve shifts to the right (positive supply shock) what happens to the real quantity of goods and services that are demanded? Look at the graph. What should we make of that change in quantity demanded? What does it tell us about "demand"?

Think about it.

Comments and Sharing

COMMENTS (13 to date)
dannyb2b writes:

"Meltzer wondered why Keynes didn't recommend an inflation target high enough to prevent a liquidity trap. After all, if the trend rate of inflation were high enough, then nominal interest rates would never fall to zero."

It may have been harder for zlb to occur. How would a higher inflation target have avoided the financial crisis and breakdown in intermediation we recently experienced?

trent steele writes:

Sometimes I think there are two types of economist: Those who believe in private property and an economy based on free exchange, and policy engineers who believe the economy is separate from the people in it (and thank god, because then they would be advocating theft and the arbitrary distribution of wealth!).


michael pettengill writes:

Years of very very very easy money has created demand for debt and shares of stock.

On the one hand, lots and lots of debt has been supplied, trillions in government debt to pay for tax cuts in most cases, but also lots of private debt to pay for going on vacation and going shopping.

On the other hand, lots of demand for shares, but that demand is selective and demands higher and higher profits, or extremely high rates of growth of something that can't be met by making physical things. But the high profits end up being used to reduce the supply of shares of stock because firms can't increase the demand for their shares by buying real stuff with the profits, like a nuclear power plant or a new car factory or a new fighter like the F-35 because the risks are so high, the stock price will crash, so those investments are done by getting money from the government, and so you end up using the profits to buy shares of stock to reduce the supply and drive share prices higher.

If the Fed were to demand tax backed infrastructure bonds with 50 year terms on new capital assets with 75 year useful lives, then monetary policy would create demand for labor.

You can't build rail lines without steel and concrete and heavy equipment. You can't build water tunnels (like those supplying NYC) without tunnel boring machines and the concrete and steel to line them.

You can't build bridges from the new bridge in Minneapolis 2008 to the Golden Gate without lots of steel and concrete and heavy equipment.

But the Fed buying of debt is not going to build capital, but only to pump up stock and bond prices which either buy old assets or fund consumption of food and housing because no demand for labor making steel and concrete generate income to pay for food and housing.

Michael Byrnes writes:

Great post. I think the real/nominal divide is very hard for many people to grasp, even economists who should know better.Same with the difference between "inflation" and "real cost of living".

foosion writes:

@dannyb2b, higher inflation makes it easier for the Fed to lower real rates, stimulating the economy. The zlb is further from 4% than from 2%.

@trent, there aren't many who don't believe in private property. The issue is often the definition of private property. For example, how much energy should the govt put into protecting patents and copyrights and how long should those rights last, or do doctors and lawyers have the right to have their competitors arrested, or is an implicit to big to fail guarantee a property right? Scott recently posted on govt policies which favor the rich and otherwise effecting "the arbitrary distribution of wealth".

AD writes:

Can someone elaborate on the last paragraph? What are we supposed to make of that change in quantity demanded?

Lorenzo from Oz writes:

michael pettengill: prosperity and easy money are not the same thing. Low inflation and low interest rates are generally indicators that money has been tight, not easy.

Besides, I find the notion that the asset booms of our time are the results of "easy money" remarkably silly. It is as if no one has studied C19th economic history, when a series of dramatic asset booms and busts occurred under the gold standard, the ultimate in "hard" money. Rising capital accumulation and technological uncertainty are easily enough on their own to explain asset booms and busts.

Mike Lorenz writes:

The "Krugman/Sumner inner sanctum" always appear to me as guys holding a match under a thermometer because the last time it said 70 we were all warm. It doesn't work that way.

Ben J writes:


This post is incredibly lucid. Reading you for the past few years in conjunction with my grad studies has put me in the same position - dviding people who discuss the macroeconomy into people who know what a nominal shock is and those who don't (or don't seem to understand what it is). Thanks for helping me to internalise those ideas - that the most important part of understanding the business cycle is understanding the distinction between the real and the nominal, and having a central bank that makes the latter have as little effect as possible.

I remember each time I made this leap for a different concept. I'll never forget that comment that Vaidas Urba made on one of your posts,

The availability of capital is a real constraint that determines the real size of the banking industry (the ratio of nominal bank assets to nominal GDP). The availability of reserves is a nominal constraint that determines the nominal GDP.
Brian Albrecht writes:

As with most aspects of modern economics, I think the majority viewpoint is completely linked with how graduate programs teach topics.

From what I can tell, most first-year graduate courses don't deal with nominal issues until students work extensively with models based on real variables. To change the majority, changing the teaching seems the first priority.

How would a nominal-focused economics graduate look like? I can partially imagine the macro course from reading blogs over the last years. I'm sure this is being done at some schools, although I'm not aware of any.

However, that's only macro. My main question is what would a nominal-focused micro look like? How does one think of standard micro models in terms from a nominal-focused point of view. If it can't be done, does that require a breaking of the link that has been constructed between micro and macro over the past decades?

tesc writes:

"When the AS curve shifts to the right (positive supply shock) what happens to the real quantity of goods and services that are demanded? Look at the graph. What should we make of that change in quantity demanded? What does it tell us about "demand"?"

Nothing, AD has not change. AS moved to the right along AD droping price inflation changig quantity demanded but not demand.

RPLong writes:

If we just fixed gas prices high enough, we'd never have to worry about shortages...

Scott Sumner writes:

Brian, My initial thought is that it makes no sense to use nominal quantities in micro, except when addressing wage and price stickiness.

Comments for this entry have been closed
Return to top