The Truth Behind the Numbers.

There has been much gnashing of teeth and the tearing of hair over the recent jobs figures. First, we saw a stunning drop in the headline unemployment rate on Friday, from 8.2% to 7.9%, a new four-year low. This morning, we got a blockbuster drop in the weekly jobless claims of 30,000 to 339,000, the lowest since 2008.

Right-wing ideologues led by retired General Electric (GE) chairman Jack Welsh, are claiming foul. They are vociferously asserting that the Obama administration faked the numbers, and that the employment picture is really much worse than presented.

Well Jack, I think it’s time to find a new hobby in your retirement. Maybe your golf swing needs some work? You are just a step away from turning on the gas burner in the kitchen and forgetting about it, a frequent, and sometimes fatal senior oversight. Economic analysis is clearly not your forte.

If the data were fabricated, it was done so by Republicans. These monthly figures are prepared by career civil servants at the Department of Labor’s Bureau of Labor Statistics. The same team has been producing these numbers for the last 20 years, through the Bush I, Clinton, Bush II, and Obama administrations. To claim they are manufactured now would require a conspiracy that defies the imagination. Get a life, Jack!

As for the weekly jobless claims, The DOL has already told us that the anomaly was caused by one state failing to turn in their numbers one time. It was not made clear whether this was California or Rhode Island. Suffice it to say that we have to wait another week to get a more accurate read on short-term employment trends.

There are far more important reasons why these numbers may be off that I have been warning you about all year. The problem arises with the seasonal adjustment process that the government has been using for the last several decades.

Seasonal adjustment employs a complicated statistical model that averages in data for the last five years. Half a decade ago, the world was a pretty horrific place.  In the waning years of the Bush administration housing was in free-fall and the subprime crisis was showing its ugly head.

Companies were dumping their work forces as fast as they could write the pink slips while the economy was entering the worst recession in 80 years, taking the monthly nonfarm payroll up to mind-boggling losses of 700,000 a month. The data in 2009 was so horrific it caused once a century spikes in the data flow, skewing monthly reports for years to come. Like it or not, those numbers still live with us today, thanks to the seasonal adjustment process.

The net effect of all of this massaging of the numbers has been to overstate the economic data from November to May every year, and to understate the unemployment rate. That’s because the bad data peaked during this period four years ago. The flip side of a flawed seasonal adjustment system has been to achieve the reverse during the June-to-October period, that of artificially low economic growth and high unemployment.

It just so happens that this year, the cusp of this biannual data bias coincides with the final month of the presidential election — hence, all the kerfuffle over the figures. This is why “Sell in May and Go Away” has worked like a charm for four years in a row. It is also why I urged readers that they had to buy the August bottom in all risk assets wherever that was, be they stocks, precious metal, or energy. This has been the wellspring of my 20% performance gain since then.

There is another factor that long-time labor analysts are starting to notice. In the old days, large companies deliberately over hired. As the economy was expanding, they used to maintain an internal pool of excess labor to accommodate the expected expansion business.

Then we entered a negative demographic trend in 2006. This is when a structurally declining rate of consumer spending is unable to generate an economic growth rate high enough to employ a generational over supply of young workers entering the labor force.

As a result, employers have ditched their internal excess labor pools, confident they can hire whenever they want from a chronically glutted army of unemployed. Cutting costs to boost profits was also a major driver. How big was this safety cushion? A few million workers. The end result of this policy was all time record profits, with S&P 500 earnings at $100/share, in the face of a chronically high unemployment rate.

Adding gasoline to the fire in the employment picture has been the 6 million jobs we shipped to China since 2000. They are never coming back, no matter who runs the country or whatever economic policy is pursued. And 6 million lost jobs means you have 15 million missing consumers, adding a further drag to economic growth already impaired by the demographic picture. This is why I started out 2012 with a lowly 2% GDP forecast and cut it mid-year to a subterranean 1.5%.

Let me throw one final thought out there as a half-century-long observer of economic data. One should always take any information originated by a government with so many grains of salt, be it the United States, China, or Lower Slobovia. These are rough estimates at best, especially when they are based on surveys, are highly volatile, and prone to error. If you bet your life on them, make sure you carry a hefty insurance policy.

My college math professor used to tell me that “Statistics are like a bikini bathing suit. What they reveal is fascinating, but what they conceal is essential.” (This is back when bikinis were brand new and considered shocking). Maybe Jack should consider purchasing a new Speedo, or joining a nudist camp.


Those Statistics are Not as Impressive as They Look

Comments are closed.