Simplify, simplify, simplify

How my approach diverges from the mainstream

[Nothing new here for long time readers, but a more compact way of characterizing my views.]

While I was getting ready for a recent podcast interview with David Beckworth, I started thinking about how I would describe my approach to monetary economics. What distinguishes my framework from that of the mainstream? In the end, I decided that my approach is best characterized as simplify, simplify, simplify.

Here are three examples:

1. Many of the most important macro issues can be explained by a simple model of nominal GDP (NGDP) growth.

2. Monetary policy is the only important determinant of NGDP in a fiat money regime.

3. We should return to the simple pre-2008 regime, where 98% of the monetary base was currency.

Simplify macro: Macroeconomics focuses on three primary issues: long run economic growth, the business cycle, and inflation. In the US, the business cycle and inflation are mostly determined by changes in NGDP. Unstable NGDP causes business cycles because hourly nominal wages are sticky in the short run. When NGDP changes, companies initially respond by adjusting their demand for labor.

In my view, most economists overrate the importance of other determinants of the business cycle. With the notable exception of Covid, supply shocks rarely play an important direct role in the business cycle. At best, supply shocks occasionally cause monetary policymakers to misjudge the situation, leading to monetary policy errors (as in 2008.) Fiscal policy also plays very little role in the business cycle. And financial crises are mostly a symptom of falling NGDP, not a cause.

The IS-LM and AS/AD models are not particularly useful, unless one follows the monetarist approach of defining aggregate demand as nominal GDP. Unfortunately, most economists approach aggregate demand from a more Keynesian perspective, occasionally conflating aggregate demand with real expenditure. This is an example of reasoning from a quantity change. Real expenditure can just as well rise from a positive supply shock as from a positive demand shock. Aggregate demand means nominal spending, or it has no coherent meaning at all. Similarly, the IS-LM approach leads many economists to engage in the fallacy of reasoning from an interest rate change.

Simplify NGDP determination: When I wrote a book on the Great Depression, I looked at a wide variety of factors that influenced nominal output. That’s because under the international gold standard no single central bank could control the global price level and/or NGDP. But once the world shifted to fiat money, it because very easy to model changes in NGDP. Central banks now had unlimited ability to offset changes in money demand (and velocity), and hence it made sense to view monetary policy in terms of outcomes. Easy money was excessive NGDP growth and tight money was inadequate NGDP growth. What could be simpler?

Again, mainstream economists make things much more complicated, examining how things like financial shocks, consumer sentiment, trade shocks and fiscal policy might affect NGDP growth. But fiat money central banks try to offset those changes, so in the end an NGDP shock is nothing more than a monetary policy error.

Simplify monetary policy: After 2008, monetary policy was made much more complex. Instead of controlling the path of NGDP by adjusting the supply of base money, the Fed also began trying to manipulate the demand for base money by paying interest on bank reserves. This led to much larger central bank balance sheets, and many central banks began to engage in complex credit policies, buying assets other than risk-free Treasury securities.

Complexity leads to confusion:

In my view, most of the monetary policy errors that we see in the real world occur due to needless complexity being added to monetary theory and policy. Consider the big inflation overshoot of 2021-22. Some prominent economists argued that supply shocks were largely to blame for the inflation spike. That’s because they were fooled by things like bottlenecks at the ports, which pushed up prices. But that’s another example of the fallacy of reasoning from a quantity change. In fact, the flow of goods through our ports increased sharply during this supposed “supply shock”.

The actual problem was that a highly expansionary monetary policy caused rapid NGDP growth. This increased the demand for imported goods, straining our ability to handle the increased flow of imports.

If economists had single-mindedly focused on NGDP growth rates, they would have seen that the overshoot of NGDP growth was roughly the same size as the overshoot of the price level, meaning that virtually all the excess inflation of 2021-23 was caused by monetary stimulus. K.I.S.S.

The opposite error occurred in late 2008, when economists were so obsessed with the financial turmoil that they failed to notice that NGDP was falling sharply. Throughout history, falling NGDP generally causes or at least worsens financial crises. (In the 1930s, falling NGDP was the sole cause of the multiple financial crises, whereas in 2008 it dramatically worsened what had until then been a very mild financial crisis.)

Monetary policy and NGDP have almost no effect on long run economic growth. But for the macroeconomic events that make the headlines (business cycles and inflation), it’s almost all about monetary policy and NGDP. Monetary policy drives NGDP, and everything else is downstream of NGDP shocks.

In a perfect world, we would not have intro courses in both micro and macro. We’d have an economics textbook with about 30 chapters, including one chapter on long run economic growth, then a chapter on inflation/NGDP, and then a chapter on the business cycle. That’s right, macro should be treated relegated to at most 10% of economics, not 50%. I’d rather have students actually understand 3 chapters on macro, rather than be hopelessly confused by 15 chapters on macro:

AD = C + I + G = GDP = M*V = P*Y

LOL, do you think one student in a thousand understands the equation I just wrote?