A Few Worrisome Signs for the Jobs Market

It is well known by now that the two most recent payroll report misses were due to the bad weather that economists and economic forecasters apparently failed to notice. They presumably never go outside, possess no thermometers and don't watch the weather channel, or any other news programs for that matter – this is the best explanation we have for this particular forecasting failure at the moment.

Forecasting with the help of a ruler pressed to a chart (i.e., by simply extrapolating the most recent trend) apparently isn't working so well when global warming makes the weather extremely cold.

Anyway, employment is usually a lagging economic indicator. This is so because employers can never be sure whether incipient signs of economic weakness are merely a short term blip or the sign of worse things to come. One doesn't simply fire people on every little blip in sales. And yet, there are recently a few signs – subtle as they may still be – that the US labor market is deteriorating. A friend pointed us to this recent post by Lee Adler, who is an avid follower of unadjusted, raw economic data and has long stressed that there is an uncanny inverse correlation between US unemployment claims and the stock market. He writes:

“The headline number for initial jobless claims came in at 339,000, which was close to Wall Street economists’ consensus guess of 335,000. On the surface it looked like a non event, with the stock market seemingly viewing it as such, but the actual data shows a more troubling picture.

[...]

The DOL reported that “The advance number of actual initial claims under state programs, unadjusted, totaled 358,914 in the week ending February 8, an increase of 3,727 from the previous week. There were 361,417 initial claims in the comparable week in 2013.”

The current number was down 2,500 or 0.7% year over year. This was a sharp deceleration from the prior week’s annual rate of decline of 8.6%, and it’s near the top of the range of weekly annual percentage changes of the past 15 months. Furthermore, the week to week increase of 3,727 was atypical for this week of February. Last year that week had a drop of 27,000. The average change for this week over the previous 10 years was a drop of 6,600. Instead of dropping this week, as is the norm, claims rose. By any measure this was not a good performance.”

Adler goes on to point out that in 2007, divergences between the stock market and unemployment claims data preceded the market decline (and we should add, the beginning of the recession). Hence this sudden change in trend is worth keeping an eye on.

We recently showed a long term chart of the inverse of the weekly claims data (the seasonally adjusted version) and the SPX. Here is a close-up:

Weekly UE claims and the SPX – the recent increase in the volatility of claims could be observed both near major stock market peaks and major lows in the past – click to enlarge.

Of course the increased volatility in unemployment claims does not have to indicate a pending trend change…ironically, because it may well be due to the bad weather.

There are a few other signs that things are not going as well as planned. Below are two search query charts from Google, the US jobs index and the US unemployment index. Here is Google's explanation of how these charts are derived:

The Google Jobs Index tracks queries related to "jobs, resume, salary, career, interview, employment" and so forth. The index is set to 1.0 on January 1, 2004 and is calculated and displayed below as a 7-day moving average.

The Google Unemployment Index tracks queries related to "unemployment, food stamps, social security, edd, disability" and so forth.  The index is set to 1.0 on January 1, 2004 and is calculated and displayed below as a 7-day moving average.”

Both charts  depict only US search traffic. Recently there has been an interesting, if still small, change relative to 2011-2013:

The Google US Jobs Index (search term on Google: GOOGLEINDEX_US:JOBS). Note that our annotation should also include '2013' – the current spike early in the year, which is actually a typical seasonal phenomenon, was higher than in the years 2011-2013 and is comparable to the spike seen in early 2008 – click to enlarge.

The Google unemployment index (search term: GOOGLEINDEX_US:UNEMPL). There has been a sudden spike to levels last seen in 2009 and 2010 – click to enlarge.

Along similar lines, Gallup's survey based indexes of the labor population, unemployment and underemployment have recently shown a significant uptick. This survey has in fact contradicted the official data for quite some time.

At present all three series have moved back to levels last seen in early 2012, a full two years ago and shortly after the euro area debt crisis hit its peak. The absolute levels are however less important than the direction – which recently appeared to be confirmed by the official payroll data as well.

Gallup's workforce and employment survey data series - click to enlarge.

Pockets of Weakness

We're not sure yet what to make of all this, except to say that it is consistent with the data points that were in evidence prior to the onset of US recessions on previous occasions. At the same time, a number of other recession indicators are not yet sounding the alarm. It would be accurate to say that the overall picture is still mixed, but pockets of weakness have become more numerous.

In this context, let us not forget the recent sharp deterioration in the manufacturing ISM and especially its new orders component. While both data series still indicated expansion (by remaining above the 50 level), it was the sharp decline from previous levels that was so remarkable.

Note also the following charts from a recent JPM economic report that show global retail sales volume (stalling out in December for the first time since the beginning of the most recent upswing), goods consumption in the 'G3' (US, Japan and euro area), as well as consumer intentions with respect to durable goods purchases in Japan.

  Global retail sales volume ('saar'= seasonally adjusted rate) - click to enlarge.

G3 goods consumption – sharp falls in the euro area and Japan, a milder decline in the US - click to enlarge.

Durable goods purchase plans in Japan – Abenomics fail? – click to enlarge

The Japanese durable goods purchase intentions chart strikes us as especially interesting, as it proved to be a long leading indicator for the global economy in 2007.

Conclusion:

Nothing is definitive yet, as all these recent signs of economic weakness may yet turn out to be temporary blips. Still, these developments are noteworthy, not least because they fly into the face of the consensus, which is that 'economic growth will accelerate' globally. Maybe it won't.