Book club: Scott Sumner’s ‘Alternative Approaches to Monetary Policy’

Chapters 4 and 5: ‘The end of macroeconomics as we know it’

It is in Chapter 5 of AAMP that Sumner’s ambition becomes apparent: nothing less than the ‘end of macroeconomics as we know it.’ He proposes a new vision for macroeconomics built around a nominal income futures targeting regime in which ‘the instrument is the indicator is the goal.’ In other words, nominal GDP futures serve as monetary policy instrument, indicator of the stance of monetary policy and the goal of monetary policy. In this conception of monetary policy, market NGDP expectations are ‘monetary policy itself.’

To get to there, Sumner elaborates on the price of money approach in Chapter 4: Which Approach to Monetary Policy Works Best?  Sumner shows why interest rates and monetary aggregates are ambiguous indicators of the stance of monetary policy through the lens of exchange rate dynamics. As Sumner notes in Chapter 1, the exchange rate is the price of money in terms of other countries’ currencies and is therefore a more reliable indicator of the stance of monetary policy. The exchange rate has been used as a monetary policy instrument historically and is still used this way contemporaneously by countries like Singapore. The point is not to advocate that the US or other countries adopt an exchange rate targeting regime. Instead, Sumner uses the exchange rate as a foil to illustrate the incoherence of alternative approaches, most notably, interest rate targeting. My sense is that this material would actually work better as part of Chapter 1 or as Chapter 2.

In Chapter 5: From the Gold Standard to NGDP Futures Targeting, Sumner cleverly walks us through a nine-step process from a pure gold standard to nominal GDP targeting. Each step involves only a small change in regime, while remaining consistent with the price of money approach.

Sumner makes an important distinction between NGDP targeting and NGDP futures targeting. Only the latter is consistent with the price of money approach. The former is just a variation on an instrument rule, like an interest rate rule, conditioned on macroeconomic variables, where the variable of interest is nominal GDP. Again, Sumner steps us through the process of transitioning from a discretionary FOMC employing an interest rate instrument rule to a nominal GDP futures targeting regime, which Sumner likens to an expanded FOMC with one dollar per vote and up to eight billion members (ie, no frictions that would preclude participation in the market). The NGDP futures market essentially becomes the central bank.

But Sumner shows that NGDP futures targeting does not otherwise entail a radical change in monetary institutions. Indeed, it is fully compatible with existing institutions and a dual price stability and full employment mandate. It just requires a re-conceptualisation of the monetary policy process. This is how Sumner describes the conceptual transition:

1. The Fed has a dual mandate to address inflation and employment. It puts equal weight on each side of the mandate. As a practical matter, that means the Fed puts roughly equal weight on inflation and real GDP growth.

2. NGDP targeting puts equal weight on inflation and real growth. Thus, stable NGDP growth at a modest rate (say 4 percent a year) is roughly consistent with the Fed’s current policy goals.

3. Fed officials say they rely on both internal models and market forecasts when setting policy.

Once we get to stage three, we can adjust the weight placed on internal models and market forecasts in forecasting nominal GDP. The internal models are a crutch to win over central bankers who do not fully trust market forecasts.

The forecast is then stabilised in the same way that a managed exchange rate regime stabilises the exchange rate: the central bank does whatever it takes to keep the exchange rate at the targeted price in real-time. Nominal GDP forecast targeting is essentially the same process. Rather than periodically adjusting an administered price for inter-bank overnight credit (the official interest rate), sitting back and hoping for the best, monetary policy would operate in real-time through open market operations to stabilise an otherwise flexible market price for NGDP futures. Sumner’s ultimate goal is to remove interest rates entirely from the policymaking process. This approach is fully robust to the zero bound problem in the official interest rate.

While not a radical departure from existing monetary institutions, Sumner shows that reconceptualising monetary policy in this way has profound implications for macroeconomics. Sumner’s vision for macroeconomics is as follows:

‘there are no more mysterious “demand shocks”—all shocks are real shocks. No more misdiagnosis of the stance of monetary policy. No more worry about the so-called inherent instability of capitalism. No fear that bank failures would spill over into inadequate aggregate demand. No reason to claim that bailouts of failing firms would boost GDP. No reason for people to expect changes in taxes and transfers to affect aggregate demand. No worries about the paradox of thrift or the paradox of toil. No reason to teach Keynesian economics. Our macroeconomic textbooks could be written by real business cycle theorists.’

Sumner characterises his vision as ‘absurdly ambitious,’ but clearly outlines what it takes to get there. It is only ambitious in the sense that it requires a reconceptualisation of monetary policy, not because it involves profound institutional change or is operationally or technically difficult. Ideas can be hard to shift, but Friedman did much of the heavy lifting in an earlier era. In many ways, Scott’s work is the completion of Friedman’s project and its logical end-state, a market-determined monetary system liberated from Keynesian doctrine.

It is the second monetarist counter-revolution.

---

---