During the latest June FOMC press conference, Fed Chair Jerome Powell was confronted with a question regarding the non-accelerating inflation rate of unemployment (aka NAIRU). That is, where exactly he estimates the natural rate of unemployment to be. The fact that NAIRU receives so much attention lately points to an important dynamic.
With the U.S. economic expansion in the late stages of a very long cycle, the Fed appears to be determined to continue to normalize its policy stance back to pre-crisis 2008 levels, when an historically unique housing shock caused series of unprecedented monetary policy moves.
This raises the question of how to define ‘normal’ or ‘neutral’ policy:
- Is it a Fed Funds rate close to the real neutral interest rate, now at 2.9%?
- Or is it an unemployment rate close to the NAIRU?
- Or is it a balance sheet that is back to its level in pre-quantitative easing?
Put differently: Fed policy in the last few years seemed to be centered on the question of the timing of next interest rate hike.
Now, however, a new question arises: how long do we continue to raise rates? This appears of particular importance in the context of long-term trends.
If you consult the graph below, an interesting bifurcation pattern occurs, in the sense that U.S. rate fluctuations at the short end do not seem to affect the long end. That is, the long end of the curve appears to be a on a downward trend ever since the record inflation period around the Volcker era in the 1980s:
It is essential to get a sense for the ‘neutral’ short end of the curve. To that end, we know that this ‘neutral’ short-term interest rate level should be low enough to allow the huge U.S. economy to remain in that stable sweet spot of a sustainable growth path -- below the level of overheating -- AND strong enough to avoid recession dynamics.
We view the following 2 yardsticks to be important gauges of what the new normal could be:
1) NAIRU: This stands for non-accelerating inflation rate of unemployment. This is a theoretical concept with empirical underpinnings. Economists refer to NAIRU as the lowest level of unemployment that can be supported before inflation will emerge.
This idea is related to the relationship summarized by the Phillips curve. That is, when unemployment falls below a certain level, employers will find they compete for workers and gradually increase wages. This will subsequently lead to inflation.
To what degree is that relationship still existent and relevant? The empirical evidence for the Phillips curve relationship offers mixed results that have especially in recent years not been very stable (Consult last month’s Zacks Economic report).
At the same time, NAIRU has varied over the years -- as you can see below -- which depicts the NAIRU estimate of the Congressional Budget Office (CBO):
Notice that since the 1980’s, the CBO estimate of the NAIRU (the blue line) has gradually decreased to 4.6%, which is far above the current unemployment rate (the red line). If you take the estimate at face value, the Fed should continue to raise rates.
But a number of reasons make us believe a more accurate estimate of NAIRU should be lower. With increasing globalization, workers are competing globally for jobs, thereby reducing pressure for employers to raise wages during times of lower worker supply, making the relationship of the NAIRU less stable.
Furthermore, as technology advances (online job networks, etc.), frictional unemployment (one element of the NAIRU) is becoming smaller, as individuals need to spend less time between jobs. We therefore think that we might not actually be that far off from the natural unemployment rate as the graph above might lead you to believe.
2) Fed Balance Sheet: After the last recession, the Fed’s balance sheet grew to $4.5 trillion portfolio, consisting mostly of MBS and treasury securities. At the end of last year, the Fed announced plans to shrink the portfolio gradually over the coming years, by not reinvesting the proceeds or maturing securities -- setting the goal at $3 trillion in 2020.
Some Fed officials -- such as Boston Fed President Eric Rosengren -- question whether it might be necessary to let the Fed portfolio wind down that much. The size of the portfolio before the financial crisis would typically remain below $1 trillion and would be a function of deposits kept by commercial banks with the Fed.
Where exactly the right portfolio size is will be important. The current process of gradual shrinking will drain resources from the system, and might put mild pressure on interest rates. This will have a tightening effect one way or the other.
It will be important, as the U.S. might reach a point in its lengthy business cycle that does not call for further tightening. That is, in Zacks’ view, it might not be as simple as setting the benchmark of the Fed’s balance sheet portfolio size to what it was before the financial crisis. The Fed might need to find a way to identify a new target portfolio size.