Friday , June 5 2020
Home / S. Sumner: Money Illusion / Why does money matter?

Why does money matter?

Summary:
If someone asks you why monetary policy is so important, you might mention the business cycle, and/or financial crises. But suppose they ask you how you know that monetary shocks cause business cycles? How would you answer? One reason that market monetarists remain in the distinct minority is that this question is really hard to answer, indeed harder than almost any other question. You can’t just say “because blah, blah blah”, you have to carefully lay out a very complex argument. In this post I’ll try to present the general shape of this argument, and also explain why it’s so complicated. To begin with, you must explain two basically unrelated points, each of which is non-obvious: 1. Why do we believe that money determines nominal fluctuations? 2. Why

Topics:
Scott Sumner considers the following as important:

This could be interesting, too:

Tyler Cowen writes Solve for the equilibrium

Tyler Cowen writes Thursday assorted links

Tyler Cowen writes Chesapeake travel notes

Scott Sumner writes The Fed cannot allow another lost decade

If someone asks you why monetary policy is so important, you might mention the business cycle, and/or financial crises. But suppose they ask you how you know that monetary shocks cause business cycles? How would you answer?

One reason that market monetarists remain in the distinct minority is that this question is really hard to answer, indeed harder than almost any other question. You can’t just say “because blah, blah blah”, you have to carefully lay out a very complex argument. In this post I’ll try to present the general shape of this argument, and also explain why it’s so complicated.

To begin with, you must explain two basically unrelated points, each of which is non-obvious:

1. Why do we believe that money determines nominal fluctuations?

2. Why do we believe that nominal fluctuations create business cycles?

The first is much easier to explain, because it fits in with basic (classical) economics, indeed basic supply and demand theory. Because nominal values are measured in money terms, changes in the value of money affect all nominal values. And as with any other good, changes in the supply and demand for money impact its value.

Thus it’s pretty easy to explain how monetary policy could create fluctuations in NGDP. Just assume that the central bank changes the supply (QE) or demand (IOR) for base money in a way that destabilizes nominal aggregates such as the CPI and NGDP.

Still, even though this is really basic stuff, the person inquiring about why money is important has to have internalized this relationship before proceeding to the much more difficult question—why do monetary shocks have real effects? And because of “money illusion”, even this simple classical argument is really hard for many people to grasp. Most people don’t think of inflation (or NGDP growth) as a fall in the value of money.

I’m going to break the second question (nominal/real interaction) down into two components, to make thing simpler:

1. The core set of theory and evidence in favor of nominal shocks creating business cycles.

2. Auxiliary evidence in favor of nominal shocks creating business cycles.

The core theory and evidence has 4 distinct parts, which in combination strongly point the finger at nominal shocks creating business cycles:

a. The existence of “odd” looking business cycles, with unemployment occurring for reasons that are not immediately obvious. (Note that today we are at the beginning of the first non-odd business cycle I’ve ever experienced. Unlike with most cycles, the reason for the high unemployment is immediately obvious right now.)

b. An obvious correlation between nominal shocks and business cycles, something noticed even by David Hume.

c. Off the shelf theories of price floors and price ceilings creating surpluses and shortages, right out of EC101.

d. Prices and especially wages that are obviously somewhat sticky in the real world.

None of these 4 points are particularly persuasive, considered one at a time. But in combination, they are incredibly powerful. When know from point #1 above that contractionary monetary policy can reduce NGDP, and we generally also see RGDP fall at the same time. This is associated with millions of workers no longer working, even though they seem to wish they were working. That looks like the “disequilibrium” that occurs under a price floor. Labor supply exceeds labor demand. And we know that nominal wages are sticky, in which case an unexpected fall in NGDP should cause mass unemployment.

The four pieces of this argument fit together like a fine Swiss watch. Everything clicks into place.

That’s actually enough for me to be sold on the basic monetary model of the business cycle. But in fact there are a number of other auxiliary arguments, which also point in the same direction:

3. Natural experiments: Monetary policy is hard to “identify”. But there are cases where central banks intentionally create monetary shocks, and then we experience the business cycle impact that is predicted. In 1981, Volcker’s Fed set out to reduce inflation with a tight money policy. Unemployment rose to 10.8%, as expected by the basic model.

One particularly interesting sort of natural experiment is switching from a fixed to a floating exchange rate regime. If money were neutral in the short run, then this policy switch should have no impact on the volatility of real exchange rates. In fact, real exchange rates become dramatically more volatile under a floating rate regime than a fixed rate regime, as we saw after the end of Bretton Woods, or when the US left gold in 1933. The only plausible explanation is that money is non-neutral in the short run, and sticky wages and prices are the only plausible explanation that has been offered for that non-neutrality.

Another very interesting natural experiment is that countries tended to recover from the Great Depression after they left the gold standard. There’s also some evidence (not as strong) that countries in Europe did better if not in the euro, as they could devalue to boost output.

4. The Natural Rate Hypothesis was developed by Friedman and Phelps in 1967-68, and later the model was confirmed by events. This model assumes that money is non-neutral in the short run. If the model were not true, then why was it confirmed by later events?

5. I’ll end with a piece of evidence that is more tenuous but still interesting. It seems like the short run non-neutrality of money is believed by a wide variety of people that approach the issue from very different perspectives. Consider central bankers, economists and investors.

Central bankers actually make adjustments in monetary policy. That’s their job. They then see the economy react in ways that convinces them that money is non-neutral. You don’t typically see extreme RBC-types running central banks, as the impact of their policy decisions would often contradict their theories. They’d be confused.

Most economists believe in the non-neutrality of money for the reasons listed at the top of this post.

And investors (as a whole, not individually) also seem to believe in the non-neutrality of money. Asset prices react to unexpected money announcements in a fashion consistent with market monetarism being true. That’s not surprising, give that MMs believe that market responses are optimal forecasts of policy effects. But this suggests that the “wisdom of crowds” also in some sense believes money is non-neutral in the short run.

To summarize, there is no brief “elevator speech” for the monetary theory of business cycles. The core argument has 5 components, a nominal model and 4 pieces of theory and evidence supporting real effects. This alone is a pretty convincing argument, perhaps even for Ray Lopez. But there are also a number of auxiliary arguments, only some of which are listed here.

A few arguments against short run non-neutrality of money have been offered, especially in the early days of real business cycle theory. But all have been addressed, and none are now viewed as persuasive.


Tags:

 
 
 
Scott Sumner
Scott B. Sumner is Research Fellow at the Independent Institute, the Director of the Program on Monetary Policy at the Mercatus Center at George Mason University and an economist who teaches at Bentley University in Waltham, Massachusetts. His economics blog, The Money Illusion, popularized the idea of nominal GDP targeting, which says that the Fed should target nominal GDP—i.e., real GDP growth plus the rate of inflation—to better "induce the correct level of business investment".

Leave a Reply

Your email address will not be published. Required fields are marked *