The Unemployment Debate

Here is a debate posted on the Economist talking about the issue of cyclical or structural unemployment.

Here is another good blog post talking about the increase in structural unemployment.

It is important to find a reasonable estimate of our potential level of output. This includes understanding the unemployment dynamics and what economists call the natural rate of unemployment. Right now the current unemployment rate is 9.7% if the potential level of output corresponds with a natural rate of unemployment of 5% than we would need to use some form of monetary/fiscal policy to eliminate the output gap. Now the natural rate of unemployment implies no workers are left unemployed due to the economic downturn, there is no cyclical unemployment only structural and frictional. Now suppose over the last decade there has been a structural shift in the make up of our economy. Suppose the United States has a large segment of the population that lacks the skills to find employment in the new economy. This would show up as an increase in structural unemployment. The natural rate of unemployment might be 7% instead of the original 5%. We would still need to use monetary/fiscal policy to close the gap, but with a higher natural rate of unemployment the policy needed would be smaller. So you can see, understanding the make up of unemployment has important consequences for understanding the magnitude of policy.

Further, the level of potential output in our economy, which is based off of the natural rate of unemployment, is used to project future levels of output, government spending, and tax revenues. So as we debate tax reforms it is a necessity to have an understanding over the levels of cyclical and structural unemployment. If we believe the natural rate of unemployment is 5% this would cause estimates of the potential level of output to be subsequently higher leading to higher tax revenues and poor policies today. To give you an idea over the range of estimates this note by the San Francisco Federal Reserve shows nearly every possible estimate of the natural rate to be around 6.3 to 8.4% (ignoring the CBO unemployment based). Notice the low estimate of 6.3% comes for the CBO, who also acts as the government forecaster for future deficits.

As most of us know the unemployment rate is still hovering around 10% and is not showing any sign of dropping. I don't think many economist would disagree over the economy turning out to be worse than we expected, so now we need to ask ourselves why is it at 10% and what can we do to lower it? When we look at policy we need to have an understanding over the type of workers that are unemployed. Are these workers in the manufacture, service, or construction industries? Additionally, we need to ask if they are unemployed for cyclical or structural causes. If all the workers were unemployed because of the recession, i.e. the cyclically unemployed, than the priority of the government should be on the economic recovery and finding programs to increase GDP. Here is where you could advocate for direct government spending programs. Unfortunately, I suspect the high unemployment has more to do with a structural shift in the labor market. Essentially, those looking for jobs lack the available skills to find employed. I will discuss this more in a bit.

Before getting to these answers it will be helpful to review some key figures. The first figure shows our the national unemployment rate (in percent, blue line) and median duration of being unemployed (in weeks, red line) dating back to 1970. There are a couple things to notice. First the unemployment rate is decreasing, but it remains near that 10% level. Unlike previous recessions the unemployment rate has not dropped suddenly. The median duration of unemployment shows no signs of decreasing. The median worker has been unemployed for nearly 26 weeks with a large number of workers unemployed for over a year. 

Workers are having difficulty finding jobs.  The graph on the right shows that workers are slowly moving into the group unemployed for more that 27 weeks (blue line). Nearly 7 million workers are in this situation. Further these numbers are not accounting for the million or so workers that have left the workforce and fall into the discouraged worker category. So where did these workers come from and what can we do to find them jobs?

To fully understand the unemployment problem we need to look at our economy over the last decade. Starting in 1999 manufacturing sectors began losing jobs. From 1999 through 2003 nearly 4 million manufacturing jobs were lost. Now many people will point the finger at China and claim they stole our jobs. Unfortunately, that is just not true. If the jobs didn't go to China they would have moved to Mexico, Brazil, or a number of other Asian or Latin American countries. China didn't steal our jobs as much as we handed our jobs to them.

A quick side bar: following the Asian, Russian, and Latin American crises throughout the 1990s many foreign investors began seeking safe investments within the United States. As they increased their demand for foreign assets the US dollar reached record highs. Yes, a large purchaser of US assets was China, but they only represent a modest amount when considering the many European, oil producing, Asian, and Latin American economies that were also seeking safe US assets. The large appreciation in the dollar made US goods more expensive on the global market but made foreign goods cheaper for Americans. As producers found it more difficult to sell their products to foreign and domestic buyers it made sense to move operations abroad. Now this situation only tells part of the story. Firms will only shift operations if they believe the exchange rate appreciation was a long-term event. The large appreciate was more likely the final straw. For years US manufacturing firms have been struggling to remain competitive due to union demands. Many unions hold monopolies within their industries (the United Auto Workers Union is the union for all US auto workers). This gave them tremendous leverage when negotiating wage contracts. Wage distortions created by the unions made it extremely difficult to remain in business. This left many firms shifting operations overseas. Back in 2005, NPR did a great comparison between GM and Toyota facilities operating in the United States. You can view it here. This comparison highlights the difficulties union firms have when facing competition from domestic non-union firms.

The graph on the right shows the large decline in manufacturing. Starting in 1999 the economy lost 3 million manufacturing jobs. The jobs losses slowed down during the housing boom, where we see the construction industry providing a temporary home for displaced manufacturing workers. As you can see today the recession has caused the manufacturing industries to lose another 2-3 million jobs but this time the construction industry has also lost 2-3 million jobs. 

At the peak in 2000 manufacturing and construction industries provided over 24 million jobs. This is nearly 20% of our labor force. Today the two industries have 17 million workers. We have lost over 7 million construction and manufacturing jobs. So what can be done to help the economy recover. Given these job losses it will be nearly impossible for the economy to return to the days of 5% unemployment. The structural factors we very likely keep unemployment around the 7% levels we discussed above.

In the short run we need to create projects for these workers. We needed to spend the $800 billion in stimulus funds on badly need infrastructural upgrades. This could have been building or updating schools, hospitals, highways, bridges, and/or the power grid. These projects could have been conducted using private/public partnerships and paid for through user fees.

Nevertheless these types of programs only address the immediate needs of the economy, they do not address the large job losses that have occurred in manufacturing. We need to spend considerable amounts of time, energy, and money retraining our labor force. We need to encourage more vocation training at the high school level and find incentives for small business to continue hiring. Pressuring China into letting the yuan appreciate will have a modest impact on the manufacturing jobs and ultimately place greater burdens on households in the form of higher prices. We need to break up the monopoly unions have in the labor markets (if we want to continue with a unionized labor force let's look at Germany and Japan for ideas, more on this later) which will allow American manufacturing to become relevant again.