Unemployment by the Numbers

Since spring I’ve called for a cyclical recovery in the US and other developed economies, a prospect that would power a substantial rally in global stock markets.

Although I agree with many of the bears’ long-term concerns--for example, excessive household and government debt--the Index of Leading Economic Indicators (LEI) suggests that at the very least a temporary, cyclical improvement is underway. And while any recovery in the US economy will likely be weak, mounting evidence indicates that certain foreign and emerging markets are enjoying a particularly robust recovery. Because S&P 500 companies generate an increasing share of their earnings overseas, this foreign recovery is bolstering corporate profits.

Critics usually cite unemployment statistics when questioning my call for a cyclical economic recovery. I’m often asked how I can call for economic growth when the US unemployment rate has leapfrogged 10 percent, the highest level in decades.

Rather than listen to the endless hyperbole that perpetuates in the media’s discussion of employment statistics, let’s examine the actual data and unemployment trends during prior recessions.

The most widely watched report of US employment conditions is the Bureau of Labor Statistics’ (BLS) monthly Employment Situation report that’s released on the first Friday of each month.

The first step is to define unemployment. The US unemployment rate is defined as the total number of unemployed divided by the labor force. This measure depends both on how one measures the total number of unemployed and how one measures the size of the labor force. The BLS’ survey omits many people that one might consider unemployed.

To qualify as unemployed one has to be looking for work. For example, a “discouraged worker” isn’t working, nor has he or she recently looked for a job. But he or she does indicate that they want a job and have looked for work in the recent past.

Discouraged workers include potential workers that just don’t feel they can find full-time employment. Discouraged workers aren’t considered part of the labor force and, consequently, don’t factor in to unemployment statistic. BLS also excludes workers who can’t find a full-time job and instead take on part-time work. Although these workers are employed, there’s a strong possibility they’re not making as much as they’d earn in full-time employment and, in fact, would rather be in a full-time job.

This graph depicts the total unemployment rate, including discouraged and part-time workers.

Source: Bloomberg

As you can see, the unemployment rate stands at 17.5 percent when you include these omitted categories. Although this percentage is shocking, it’s the most useful statistic because the BLS only publishes data going back to 1994--it’s tough to make a meaningful comparison.

Nevertheless, by this measure the unemployment rate is almost twice what it was at the height of the 2001 recession. That being said, the 2001 recession was barely a recession and is widely considered to be one of the mildest in US economic history. This isn’t a particularly useful comparison; it would be more meaningful to compare the current unemployment rate to the recessions that occurred from 1973 to 1974 and 1981 to 1982.

One valid conclusion we can extrapolate from data is that this measure of unemployment topped out in late 2003, roughly two full years after the 2001 recession ended. In other words, the full unemployment rate appears to be a lagging indicator.

Along these lines, the conventional measure of the unemployment rate tends to lag economic recoveries. The last spike in unemployment came after the 2001 recession, which ended in November 2001; in that instance the unemployment rate didn’t top out until June 2003. This pattern held true for the early ‘90s recession, which ended in March 1991; in that case the unemployment rate didn’t top out until June 1992.

This lag is hardly a new phenomenon. In the vicious economic downturn that occurred from July 1973 to March 1975, the US unemployment rate peaked at 9 percent in May 1975.

To make a long story short, the recent jump in the unemployment rate--no matter how you look at it--isn’t inconsistent with economic recovery. I wouldn’t be surprised if unemployment trended higher well into 2010.

The rate of change in the unemployment rate is another useful metric.

Source: Bloomberg

To create this chart, I compared four different recessions: 1973 to 1974, 1990 to 1991, 1982 to 1983, and the current contraction. I examined the official US unemployment rate six months before the beginning of each recession as well as several months after the unemployment rate topped out. Because the recession of the early 1980s was a double-dip, I counted that cycle as a single downturn.

There’s no way to sugarcoat this chart. One would expect the current downturn to be worse than the comparatively mild recession that occurred between 1990 and 1991, but the so-called Great Recession is worse than the downturns that occurred in the 1970s and 1980s. In the first half of the graph, unemployment appears to behave normally, but the situation deteriorated rapidly at the end of 2008--the height of the credit crunch.

Here’s yet another way to analyze US unemployment.

Source: Bloomberg

This chart depicts the percent change in monthly initial jobless claims from six months before each recession. Weekly jobless claims data is released every Thursday. This data comes from individual states that record the number of workers filing for first-time jobless benefits. Unlike the unemployment rate, initial jobless claims tend to be a leading indicator of US economic activity. The last spike in initial claims topped out in September 2001, before the end of the 2001 recession. Back in 1991 claims data topped out about a month prior to the recession’s end.

The graph depicts only the three worst recessions in the post-war era: 1973 to 1974, 1981 to 1982 and 2007 to 2008. As you can see, initial jobless claims appear to have spiked at a faster face in this cycle than in the early 1980s but at a slower pace than in the mid-1970s.

The current pattern appears to be playing out much like what occurred in 1973 to 1974.

Debates over unemployment tend to be filled with hyperbole. The reality is that the unemployment rate tends to rise months after recessions end; it’s not valid to claim that falling unemployment is a prerequisite for recovery.

Finally, my comparison of the recent contraction to prior recessions shows it to be just as bad but arguably no worse 1973 to 1974 and 1982 to 1983. The bottom line: These statistics don’t preclude a cyclical upturn but do offer further reason to expect tepid consumer spending and a cyclical economic environment in coming years.

Comments

[Old Trader]

"Discouraged workers include potential workers that just don’t feel they can find full-time employment. Discouraged workers aren’t considered part of the labor force and, consequently, don’t factor in to unemployment statistic. BLS also excludes workers who can’t find a full-time job and instead take on part-time work."

Unfortunately, both of those groups (discouraged, and involuntary part-time) are becoming an ever larger part of the population/work force.

2009 Nov 24 09:26 AM

[thiazole]

Could you point me to some examples where a recovery started, but high unemployment stopped the recovery? Historical examples are much more convincing that just "saying it".


On Nov 24 06:39 AM chris coonan wrote:

> What is distressing is that unemployment as a correlation to recovery,
> is not necessarily a lagging indicator. The excessive unemployment,
> as you point out, and I contend is even worse than the BLS represents,
> can stall out a recovery all together. With an importing nation,
> depending upon consumption, and credit spending....unemployment is
> a HUGE FACTOR. Credit is also the difference this time around, people
> don't have any, it has all been devoured by the housing collapse.
> That is what is different this time around, and why this time should
> be correlated more closely to 1929/1930 than to the 1973 event.<br/>
>
> I think the author is overly optimistic.

2009 Nov 24 10:28 AM

[Elliott Gue]

Thanks for the comments. A lot of investors confuse business cycles with long-term secular trends.

I am hardly optimistic about the US economy longer term. I have written several posts on SA explaining the dangers posed by excessive household and government debt, unemployment and other factors. These remain huge risks longer term.

However, as I noted above and in prior articles, I don't think these longer term risks preclude a cyclical recovery and rally. In fact, we have proof that's the case -- the US is seeing an economic recovery, albeit a rather weak one, and the stock market has rallied as a result of improved economic conditions (real and perceived).

I would submit that in a market that's likely to offer, at best, sub-par returns for some time, investors must catch rallies such as we've seen off the March lows if they are to generate decent returns. Investors must be flexible and willing to suspend long-term arguments to participate in cyclical moves. Some of the most ardent bears, at least the successful ones, are willing to do this.

Personally, I don't think the US stimulus package represents good economic policy -- or, at least, it did not in the form passed early this year. And, as I noted before, I see America's rising public debt as an anathema. However, the stimulus package is a bit like giving drugs to an addict -- the long-term effects aren't healthy but it can make the patient feel better for a time.

2009 Nov 24 12:35 PM

[The Alchemist]

There's only 1 minor problem in that assessment. China was not part of the picture in past recessions. In the past, the economy would create too much goods, so people lose their jobs and the goods gets consumed because less are produced, then inflation sets in as less outputs are available, and the economy grows because of demand, which the economy then hires people to produce more, and the cycle repeats. This occurs because the economy is so vast that it is impossible to be managed, so the economy instinctively fluctuates around an equilibrium. This is why that while wages and earning grew the purchasing power has changed little in terms of real purchasing power. This was also why the rich poor gap was not as significant as it is now. Globalization changed that, that is to say the inflationary pressure was removed and people were allowed to do as they please, because honestly, most people would rather not do manufacturing. The economy became much more service oriented in the last 2 decades, all thanks to the result. It works both ways of course, lack of inflationary pressure means this time it will be especially difficult to revive the economy, because the service industry services real economic activities, which is also why manufacturing had always been the sector to pull the economy out of a trough. It can still go either way at this point, depending on what happens, though I would say the odds of a V or even U shape recovery is a lot less likely than a lof of dubyas. In the perfect world, China would be happy to manufacture for the world and receive services from the world, but China does not like the dirty work any more than other people, and would prefer white collar jobs as well, hence, don't count on the Chinese miracle. For this reason, even if this recovery were to last a decade (which it just might once it finally, finally gets off the ground), it will eventually come crashing down again, and next time, the stimulus won't work.

2009 Nov 24 04:03 PM

[teamlysol]

Figures don't lie, but liars figure, and I figure that the reason that unemployment is higher is because people don't have any money, either real, or promissory, and, as a result, sales volumes are lower, which means lower operating revenues for businesses, which means less job positions, lower salaries, and generally a lot less money in circulation. Comma. 10.2% UNemployment also translates as: 89.8% EMployment, and, if you were looking at the whole thing differently, say the conventional grading scale, that would be a B+. Not a bad result for having just watched 14 trillion dollars wash out of the Con Me. But, if the truth of the matter is closer to 18%, as some have theorized, then you're talking about a B-, there, which is a little less good, but still better than some countries, where unemployment is something like 60%, which is a prolonged exercise in hunger and poverty, usually accompanied by famines, rampant disease, civil war, locusts, the whole program, there.

Could the United States get to 20% unemployment, if we really unapplied ourselves, if everyone quit their jobs tomorrow, and bought a VW van with their last 1200 dollars, and wore hempen clothes, smoked dope, and dropped out of society? I don't know, that's a lot of dope, and the economy would probably trundle on, just from the sale of all-natural foods, herbal remedies, sunglasses, medical marijuana, and VW parts and memorabilia. I think we'll survive the 21st century economically, maybe not in the style to which some people were accustomed, but soup is good food, and economic projections make for some great birdcage liner.

2009 Nov 29 11:51 PM