A national labor market is something enormously complex. Labor is hardly a commodity, the markets are rife with imperfections and information asymmetries (does an employer know beforehand what exactly he is buying?). Demand and supply for specific categories (of which there are numerous) are in constant flux, so it's no wonder there is always a fraction of people unemployed (so called 'friction unemployment'), which is simply the result of market imperfections that can be reduced, but not eliminated.
Much worse is a structural mismatch between supply and demand, which can be caused by rapid sectoral change, a mismatch in education and labor demand, or localized crisis. Another problem is when there is simply insufficient demand for goods and services, resulting in lower labor demand. We're experiencing something like this right now as a result of the financial crisis and deleveraging households.
There is a wide variety in how different institutions in different countries deal with these problems and it's insightful to discuss a few noteworthy examples of these.
National differences in flexibility
Companies need to rely on some flexibility in order to adapt to changing circumstances. There are interesting differences between countries with respect to that:
- The US, is a 'hire-and-fire' economy, where cost of firing employees is low. This provides companies with so called 'numerical flexibility,' in times of falling demand, companies can easily lay-off employees to adapt
- Denmark also makes firing cheap, but puts large efforts in so called 'active labor market policies,' basically retraining of laid-off employees, providing 'flexicurity'
- Japan still has a tradition of more or less life-time employment, but it has very flexible wages through a bonus system, providing 'pay flexibility'
- Germany also makes it quite expensive to fire employees, but it has policies which makes it easy for firms to reduce working hours, providing 'work-time flexibility'
The Japanese and German models do not provide much in the way of 'numerical' flexibility, that is, it is difficult and expensive for companies to quickly change their workforce. The disadvantage, from a national point of view, is that labor allocation over the economy is slower. This can slow down economies to adapt to changes in the composition of demand for its products and services.
However, they have alternative ways of flexibility to deal with that, which is to change working hours (Germany) or pay (Japan), so this disadvantage can be overstated. The advantage of both the Japanese and German model is that labor relations are more of a long-term relationship, rather than a 'one-night stand.'
This gives companies more of an interest in developing the skills of their employees. It isn't a surprise that both countries, more than most developed economies, have maintained a thriving export manufacturing sector that is skill based. Germany has good complementary institutions in its vocational training system.
In the US, where it is much easier to fire people, labor can be rapidly redeployed in sectors that are growing, although there is one quite important barrier to that. If the skill base of the different sectors is too much apart, workers need retraining for this efficient re-allocation to take place, and that's where the advantage of the Danish model lies.
The US model has three important implications:
- Prolonged slumps lead people to withdraw from the labor market altogether, effectively reducing labor supply ("hysteresis")
- This reduces potential output as well
- As a result, stimulus policies can run into capacity problems
The US model's disadvantages become serious during a prolonged period of weak growth like we're experiencing now after the financial crisis. Employees are neither on shorter hours (Germany) nor with reduced bonuses (Japan), nor are they retrained (Denmark). They simply withdraw from the labor market altogether.
From an economy wide perspective (not to mention the perspective of the people involved), this is a rather massive waste. It means that every year an X percentage of people withdraw from the labor force of which a declining fraction will be able to return the longer it will take the economy to fill the output gap, the difference between potential and real output. Potential output is the output produced if all factors of production are employed.
This loss of human capital (a similar mechanism is operative of physical plant that has been idle for too long) is producing a situation in which prolonged below capacity output reduces the potential output the economy can produce, simply by effectively withdrawing human and physical capital from the economic process.
So here we see how important it is to keep the economy humming close to capacity. Prolonged periods of below capacity production will start to reduce the capacity itself, by making a certain percentage of human and physical capital obsolete. This 'hysteresis' effect is an important behind the models developed by Summers and DeLong, who argue that stimulus spending could, under certain conditions, pay for itself.
Stimulus policy effectiveness
Apart from the personal tragedies and the waste of human capital, there is another problem with people withdrawing from the labor market. After a prolonged period, it effectively reduces the labor supply. This can very well unfavorably shift the so called NAIRU, the non-accelerating inflation rate of unemployment, basically the rate of unemployment at which inflation is constant.
The task of stimulus policies is to bring demand back towards potential supply. But if demand running below potential supply for a prolonged period of time, this reduces the potential supply itself, then the two ultimately meet at a lower output (and more stimulus would trigger inflation). There is a bit of a debate going on about to what extent this is happening right now.
Two ways out
The logic of the above argumentation also provides two ways out, although unfortunately it's a little late in the game already for either of these.
First, demand should never be allowed to run for so long and so much below potential output, as that causes human and physical capital to become 'obsolete,' which reduces potential output. We should have gone for an even bigger stimulus early on.
Second, this doesn't happen to the same extent in any of the other labor market models described above. Rather than firing people, they just work fewer hours in Germany or cut bonuses in Japan, while in Denmark people are retrained to keep them in the effective labor supply.
Japanese and/or German labor market institutions are probably a couple of steps too far for the US, as these are quite alien to the hire and fire culture. However, the Danish model is quite 'doable.' Pity it isn't done. Quite the contrary:
Across the country, work force centers that assist the unemployed are being asked to do more with less as federal funds dwindle for job training and related services. (NYT)