Just about everyone (myself included) who ventured a payrolls forecast was crushed by the scant December gain of just 74k. How much should you adjust your outlook on the basis of just this one number? Not much, if at all. It is important to watch for trends in the data, and always keep Barry Ritholtz's warning in the back of your mind:
...we know from each month’s revisions that the initial read is off, often by a substantial amount. It’s a noisy series, subject to many errors and subsequent corrections.
To put this into some context, consider what it is we are measuring: The change in monthly hires minus fires. A monthly change in a labor force of more than 150 million people. That turns out to be a tiny net number relative to the entire pool -- about one tenth of one percent.
This is why I continually suggest ignoring any given month, and paying attention to the overall trend. That is the most useful aspect of the monthly NFP data...if you focus on the monthly numbers, you will be given so many false signals and head fakes that you cannot possibly trade on this information in an intelligent manner.
Indeed, the December number was mitigated by an upward revision to November, leaving the growth pattern looking very familiar:
One interpretation of the December outcome was that it was largely weather related. One would think, however, that such an event would have a forecastable negative impact on payrolls. Regardless, the bigger message is that the monthly change in payrolls is a volatile series, and one should be wary of putting too much emphasis on either small or large gains.
Perhaps the real story then is that another existing trend in the data, the downward pressure on the unemployment rate from a falling labor force participation rate, continues unabated:
Moreover, the pace of improvement in alternative measures of labor utilization is not accelerating and arguably appears to be slowing as might be expected if the formally cyclically unemployed increasingly become structurally unemployed:
Altogether, I think the report can be neatly summed up as 1.) indicative of a more modest improvement in activity than suggested by actual and estimated GDP numbers for the final half of 2013 and 2.) suggestive of structural change in labor markets.
The employment report generally complicates monetary policymaking. Not the nonfarm payrolls numbers so much; that number will largely be written off as anomalous in the context of the overall trend. Indeed, this was the first word from Fed officials. St. Louis Federal Reserve President James Bullard, via the Wall Street Journal:
"I would be disinclined to react to one month's number. I think it's important to get future jobs reports and see if new trends are developing," said Mr. Bullard at a press briefing following remarks here to local business leaders.
Richmond Federal Reserve President Jeffrey Lacker offered similar sentiments:
“As a general principle, it’s wise not to overreact to one month’s employment report,” Lacker said. “Employment has been growing along a pretty steady trend this year. It takes a lot more than one labor-market report to be convincing that the trend has shifted, and in my experience one employment report rarely has an effect by itself on monetary policy.”
I think the Fed is generally committed to winding down asset purchases this year, and will not want to be overly sensitive to just one report (that said, they will be overly sensitive to one number if it fits their preferred policy path). Only a more significant change in the overall tenor of the data will alter the pace of tapering.
The drop in the unemployment rate, however, is something more of a challenge. The Evans rule simply isn't looking quite so clever anymore:
Monetary officials generally believed not only that 6.5% unemployment was far in the future, but also that policy would become much more obvious as we approached that target because inflation pressures would be evident. Neither has been true. Not only has unemployment fallen more quickly than anticipated, but inflation remains stubborningly low. With regards to the former, officials increasingly see the decline in labor force participation as largely structural and outside the purview of monetary policy. Bullard, via the article quoted earlier:
Mr. Bullard signaled he wasn't particularly alarmed by a decline in labor force participation, saying it appears at the right level given current demographics.
And, via a nice Wall Street Journal interview with San Francisco President John WIlliams by Jon Hilsenrath:
We’re still working hard on this issue of employment-to-population. Everybody is struggling with the puzzle of why the employment-to-population ratio has stayed low. To what extent are movements in labor force participation structural or cyclical? And to what extent can monetary policy have an influence on those developments? I think the majority of the decline in the participation rate is due to structural factors related to the aging of the population and people going into disability. Very few people come back into the labor force from that. I do think part of it is cyclical. The data in the next year or so are going to inform us better about what is the trend.
With each passing month policymakers are increasingly comfortable taking the unemployment rate at face value. That means they increasingly expect the inflation numbers to pick up. Back to Williams:
As the unemployment rate continues to come down, utilization continues to go up, as the economy continues to improve, I would expect the underlying inflation rate to track back towards 2%.
But he clearly recognizes the potential for inflation to remain low:
The second question is why is inflation so low? To what extent does it reflect just some transitory influences, such as health care costs, and to what extent is it really reflecting a persistent ongoing inflation trend that is too low? And again how can monetary policy affect that? We’re in this world where inflation doesn’t move around a lot around 2%. It has become hard to model and to know exactly what are the factors causing inflation to be too low and which are the ones that are going to help bring it back to 2%. That gets to the downside risk question. If inflation does stay stubbornly low, that obviously is an argument for more monetary accommodation than otherwise.
Likewise, Bullard shares these concerns:
Mr. Bullard said he continues to watch inflation closely, saying it should rise as the economy picks up and the jobless rate continues to fall. But the central banker added he wants to actually see that rise come to fruition as the Fed assesses further tapering of its bond-buying.
"If inflation stepped lower in a clear way, I think that would give me some pause," Mr. Bullard said. "I'm looking for signs inflation is going to come back."
So where does this leave us? First of all, I think the Evans rule is already for all intents and purposes defunct. The unemployment rate is just a hair away from 6.5%, and the Fed has no intention of considering raising rates anytime soon. Second, there probably isn't a replacement for the Evans rule in the works. Bullard:
He expects the Fed for now to hold its threshold for unemployment at 6.5%. The Fed has said it won't increase interest rates until the jobless rate falls below that level so long as inflation stays contained.
"Moving (thresholds) around too much is likely to damage our credibility," Mr. Bullard said
And Williams on not setting a lower bound for inflation:
My view is the current [Fed policy] statement does a good job of capturing the fact that once unemployment gets below 6.5%, then obviously we’ll be taking seriously what is happening in inflation, we will be looking at what is happening with employment and growth and everything, and then we’ll be judging what is the appropriate stance of policy. It just gets very complicated quickly when you start adding more and more clauses about what conditions would you or would you not raise interest rates. Unfortunately, that is the game we’re playing … the FOMC statement has gotten awfully long. It has gotten awfully complicated. The statement is probably better used to try to emphasize the key points as opposed to trying to explain everything in our thinking.
My sense is that they thought the Evans rule was clever and simple, but it turns out that fixed numerical objectives are not quite so simple. Well, multiple numerical objectives are not quite so simple. The ironic outcome to the Evans rule experiment is that policy communication would arguably have been smoother if the Fed simply emphasized an inflation target. Policymakers could have been agnostic on the reasons for the declines in labor force participation; it was irrelevant given the path of inflation. Perhaps the focus on the unemployment rate was something of an unnecessary complication that now needlessly leaves the impression that policy will soon turn more hawkish than is the case.
Thus, the third takeaway is that policy is now largely about inflation (although arguably it always is always about inflation). Ann Saphir and Jonathon Spicer at Reuters:
Stubbornly weak inflation is shaping up as the wild card for U.S. monetary policy makers this year, with top Federal Reserve officials stumped by why it has lingered so low for so long and at odds as to what to do about it.
As the Fed wrestled through last year with deciding when to start trimming its massive bond-buying stimulus, the bulk of attention was focused on the unemployment rate, which until recently has been slow to fall following its spike up to 10 percent during the recession.
By last month, policy makers had grown confident enough in the job market to dial back on the program. Figures released Friday showed the jobless rate fell to a five-year low of 6.7 percent in December, despite the smallest monthly job gains in three years. With much of the hiring slowdown attributed to bad weather, however, many analysts say the Fed will stay on track with plans to end bond buying by late this year.
But there is a hitch: inflation has been drifting down for much of the last two years, measuring a feeble 1.1 percent in November by the Fed's preferred gauge.
As long as inflation reverts to target slowly (with a caveat to be noted below), the Fed will not be quick to hike rates. But the Fed will be increasingly nervous that a sudden burst of inflation means they are behind the curve. WIlliams:
Whether we cut purchases by 10 billion a month or not, we still have a very accommodative stance of policy and that is going to stay with us for quite some time. That is where I worry. If the economy really picks up or inflation or risks to financial stability really do start to emerge in a serious way, we need to be able to move policy back to normal, or adjust policy appropriately, in a timely manner. It’s always a difficult issue. This time it is just a much greater risk because we’re in a much more accommodative stance of policy.
I think it will be the sensitivity to positive inflation surprises that has the potential to add a hawkish tenor to policymaking. WIthout those surprises, policy will continue along current expectations. There is a caveat - note that Williams essentially admits that the possibility and willingness to use monetary policy to address financial stability. Triple mandate. Watch for it.
Bottom Line: I don't see much in the employment report to indicate any fundamental change to existing trends. Nor do I anticipate any change in policy. Tapering is likely to continue at its modest pace. As expected, the Evans rule is defunct, and it doesn't seem like policymakers are inclined to replace it with another set of fixed numerical guidelines. The primary driver of policy is now the pace of incoming data relative to the inflation outlook. Financial stability is probably something like a third-order concern at this point, at least as far as monetary policy is concerned. But that could change.
