The Wall Street Journal reports that most economists still expect the Fed to raise rates in September:
Financial market participants tend to be less confident, with odds of a September hike running around 35%. Still, the consideration of any rate hike may seem odd given the lackluster nature of the US economy. Notably, inflation wallows below trend and anemic wage growth suggests significant remaining labor market slack. The Fed, however, looks at the progress towards its goals, which on the unemployment side has been substantial, as well as the perceived need to act ahead of actual inflation.
In short, the Fed believes the risks to the economy are shifting toward overheating, even if the economy is not yet overheating. And, as Greg Ip at the Wall Street Journal identifies, this has important consequences for monetary policy:
Risk management suggests they ought to start in September, because then they retain the option of tightening once or twice before the end of the year. But if they wait until December, they forgo that option. (This assumes they do not move at their meeting in October, which is not followed by a press conference.)
This is probably the best argument for a September rate hike. Federal Reserve Chair Janet Yellen made it fairly clear in her Congressional appearances this week that she would prefer to move earlier but more gradually than later and more rapidly. And even if you think she only anticipates a single rate hike this year, that outcome is not precluded by a hike in September. Yellen has also said we should not expect a clearly identified path similar to the last tightening cycle. They can hike in September and pass on December.
Paul Krugman thinks the Fed's logic is completely backwards. From his Bloomberg interview this week:
If the Fed waits too long to raise rates, then we get a little bit of inflation. If the Fed raises rates too soon, we risk getting caught in another lost decade. So the risks are hugely asymmetric. I really find it quite mysterious that the Fed is eager to raise rates given that, they're going to be wrong one way or the other, we just don't know which way. But the costs of being wrong in one direction are so much higher than the costs of being the other.
The Fed, I think, believes the risks are asymmetric in the other direction - that inflation expectations are very fragile to the upside, and hence waiting too long risks a costly rise in actual inflation.
If I had to bet who would be proven right, I would put my money with Krugman. Inflation and inflation expectations have proven substantially less fragile in the past twenty years than the Fed likes to admit. Consider first that inflation has tended to hover mostly below two percent since 1995:
Core inflation averaged 1.70% since 1995, headline inflation 1.86%, both comfortably below target. Note the particular rarity of periods where core inflation rises more than 25bp above target. What used to be one of the Fed's favorite indicators, the 5-year, 5-year forward breakeven rate, isn't pointing toward high inflation in the medium term:
Although the Fed frets about unemployment, even at low levels of unemployment, inflation is more often than not below target:
And neither faster wage growth nor low unemployment triggers higher inflation expectations. Inflation expectations are remarkably stable, with relatively few deviations from 3% which tend to be traced back to gasoline prices:
The Fed would argue that their credibility explains stable inflation expectations. By acting ahead of inflation, the Fed ensures there is no above-target inflation, and that connection between policy and outcomes gives rise to that credibility. I would argue that two decades of generally below target inflation suggests an overly excessive pursuit of credibility at the cost of economic underperformance. We don't reach the target inflation consistently, but we do get recessions and slow job market recoveries.
Also, it seems that Yellen abandoned her enchantment with optimal control models. A recent version from the IMF indicates it would be still preferable to delay rate hikes in favor of a more aggressive normalization path later:
Under the optimal control approach, the Fed would accept the cost of temporarily higher inflation (still within a 25b range of target) in return for a faster return to potential output. Yellen now appears will to tolerate a return to the inflation target from below, rather than above, in order to avoid the possibility of a sharper rise in rates later.
Why the change of heart? Why the gradualist approach? It is reasonable to believe its about financial market instability. Back to Greg Ip:
... the effects of six years of zero rates on leverage and risk-taking are increasingly evident. As Ms. Yellen’s Monetary Policy Report noted, “Credit markets have been reflecting some signs of reach-for-yield behavior, as issuance of speculative grade bonds continues to be strong, yields are low, and credit spreads are somewhat narrow by historical standards.”
Fair enough, maybe it isn't about inflation, but financial markets. But that is kind of disconcerting as well - the gradualist approach didn't work so well last time around. Seems like if you were really worried about financial markets, you would want to follow the optimal control approach and move quickly when inflation warranted a policy shift. The error of the last cycle may not have been in waiting too long to hike, but hiking too slowly when the time came.
Bottom Line: Ultimately, as the crisis fades further into the rearview mirror, the Fed see the policy risks shifting. Many, including Yellen, will shift back toward the central banker's natural inclination to fight inflation, despite the lack of inflation for the past two decades. And that natural inclination keeps the September option alive. Given the Fed's penchant for tight policy on average, the risk is that while they don't trip the economy into recession in the near term, they instead lock the economy into a sub-par equilibrium.