US inflation may trend lower than the Fed would like

All this year the Federal Open Market Committee (FOMC) has been confidently stating that the US economy was strengthening and that any inflation weakness was “transitory,” but it is now starting to look more than simply transitory.

Aug 31, 2017 | John Greenwood

Persistently slowing inflation
The headline consumer price index (CPI), a major measure of inflation, peaked at 2.8% year-on-year in February this year, but by June it had slowed to 1.6%. The core CPI (eliminating the volatile food and energy components) was 2.3% year-on-year in January, and had slowed to 1.7% by June. A similar trend was observed for personal consumption expenditure (PCE) deflator too. This matters because expectations about future interest rate hikes by the Fed and plans for balance sheet shrinkage could be deferred if the nominal growth of GDP slows persistently. While growth itself is not yet in doubt, inflation could well be trending lower than the Fed would like. If we concentrate on different measures of core inflation, the problem remains. The Trimmed Mean CPI produced by the Cleveland Fed, the Atlanta Fed’s Core Sticky CPI measure and the Trimmed Mean PCE calculated by the Dallas Fed all slowed down in May or June from a peak at the beginning of the year.

A break with past patterns of inflation
The fact that all five measures of core inflation have now been declining for five or six months creates a problem for FOMC members. Continued growth of the US economy over the past eight years since the Global Financial Crisis has gradually raised employment, lowering unemployment to 4.3% in July, below traditional measures of “full employment.” But this has not been accompanied by the rise in wages and prices that the Phillips curve - adhered to by the Fed's chairperson Janet Yellen, most central banks and many academic economists - had predicted, which is also the rationale behind the FOMC's “transitory” comment.

In the current business cycle expansion since mid-2009 and continuing to the present, the wage and price Phillips curves have become almost flat despite a big decline in unemployment (from 9.9% in 2009 Q4 to 4.3% in July 2017). Wage growth has broadly remained stable between 1.8% and 2.2%, though with brief spikes in 2015 Q1 and 2017 Q1. However, these “spikes” hardly constitute sufficient upward movements to threaten a general outbreak of inflation. Consequently, core PCE inflation has remained subdued, well below the Fed’s target rate of 2%, and mostly in the range 1.3% to 1.7% since 2012 Q3.

Phillips curve: not a reliable predictor
The problem with the Phillips curve as a model for predicting inflation is that both the tightness of the labor market and the growth of wages are symptoms of deeper, underlying forces driving the economy. The Phillips curve relation is not really a theory of inflation, but rather an empirical observation – that wage acceleration due to tightness in the labor market has in the past been associated with an acceleration of inflation. But this needs not always be the case.

The key point is that wages are a relative price, whereas inflation of the overall price level is ultimately a monetary phenomenon. In my view wages are being driven by more fundamental factors such as globalization of the labor market since the 1990s and slow money and credit growth. Currently not only is M2 (money supply) growth still in the mid-single digit range, but wider measures of credit such as shadow bank credit growth remain in negative territory. In this environment it is no surprise that the Phillips curve is failing to follow past norms. Given the continuation of globalization and slow money and credit growth, I expect US real GDP growth to fluctuate around 1.5%-2% while inflation remains at or below the Fed’s target for much of the next two years.