Going Beyond Jobs Numbers Noise
I keep telling investors to ignore the monthly frenzy surrounding the monthly jobs report (see this, this and this). It isn't significant for their holdings, at least not in any actionable way. By the time we know for sure that the economy has accelerated or slowed, stocks will have long since reflected this in earnings and then prices. It is only partially a joke to note that economists are often the last to know.
Why should investors ignore the nonfarm payrolls numbers? Consider what must be done in order to either profit or avoid a loss based on changes in the employment situation. Investors would have to:
• Predict what the nonfarm payrolls will be;
• Guess if this will be above or below consensus (which changes often);
• Conjecture how much of this is already reflected in stock prices;
• Arrange your portfolio in light of all of the above.
To win this game, you must get each of those four steps right. And, each one is dependent upon your getting the prior step correct, then pyramiding that prior lucky guess with another. Someone usually gets this correct, but I would suggest this is a function of random luck, not skill. That isn't an attractive basis for putting risk capital to work.
Rather than play a no-win game, let me suggest something else: Try to better understand what is going on right now, rather than create a series of improbable guesses about the future.
We have previously discussed the dangers of overemphasizing the monthly employment situation. This month, let’s take the flip side of that discussion, and review what signals are contained in the noisy monthly series.
First, some data: Today’s nonfarm payrolls report shows that the U.S. economy created 257,000 jobs in January. Unemployment ticked up to 5.7 percent, from 5.6 percent a month earlier, as more than a million people returned to the labor force. Wages rose 2.2 percent.
There is some good signal among the noise, if you know where to look. If you must obsess over this data series, here is where I suggest you focus:
Trend vs Data Point: I keep exhorting readers to ignore the monthly numbers, instead focusing on the longer-term trend. For 2014, the average job growth was 255,000 a month. That is a significant improvement over 2013’s average monthly gain of 194,250. For those with short memories, in 2009 the U.S. was losing 424,000 jobs a month.
It's noteworthy that the U.S. stock market was up a lot more in 2013 than 2014. And consider that in 2009 the Standard & Poor's 500 Index rose 23 percent, despite those monster job losses and an economy that spent the first half in recession.
Wages, Temporary Help, Hours Worked: My readings of the data over the years suggest these three metrics are the closest thing within the NFP to a leading indicator for future employment gains.
Hours Worked: The first thing a company can do when it sees an increase in new orders is increase the hours of existing employees. (This is especially true in manufacturing).
Temporary Help: The next thing a company can do -- especially if it doubts that the economy really is improving -- is add temporary workers. There is no long-term commitment, and management can cut them at will. A temp worker who is a good fit can be added permanently if the demand is there.
Employment of temporary workers tends to fall earlier in a slowdown, and decline faster and further than total nonfarm employment. Companies also make more use of temp workers in the early stages of a recovery.
Rising Wages: This suggests companies have rising sales and earnings, and would rather pay up for their existing workers than lose them to competitors. Retaining talent is typically cheaper than finding new bodies. Average weekly earnings were strong last month, up 2.2 percent, the biggest gain since November 2013.
The above three elements are the closest thing within the NFP to a leading jobs indicator. But some others are worth noting.
Private vs Public-Sector Gains: This can help you determine the sources of job growth. That there were more private sector jobs created than the total (267,000 versus 257,000) tells us that government job losses remain a drag on the economy.
Benchmark Revisions: They can provide an important counterpoint to the Bureau of Labor Statistics models relative to the actual health of the job market. When we see consistently lower revisions among strong monthly numbers, it could be a sign of underlying trouble. Similarly, when the revisions are consistently higher even in the face of weak numbers, it can suggest a slow healing is taking place. Although this can provide insight about the health of the economy, it isn't exactly a hard quantitative science.
Labor Force Size: The pool of available workers has been declining for the last few years. Demographic factors and the aging of the baby boomers account for some of this reduction. Secular trends such as globalization and technology also have been driving people from the work force.
This month we saw a bit of a countertrend: More than 1 million people returned to the work force. That can be construed as positive. The labor force Participation Rate rose 0.2 percent to 62.9 percent. Note this is just a bit higher than the lowest level since the late 1970s. Whether this is the start of a new trend bears watching.
Understanding where we are in the employment cycle is vastly superior to any given monthly data point. More significant is what's within the data. The alternative is a useless guessing game that creates far more heat than light.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Barry L Ritholtz at email@example.com
To contact the editor on this story:
James Greiff at firstname.lastname@example.org