Melberg, Hans O. (1992), Unemployment: Micro- or Macro-theories?

[Note for bibliographic reference: Melberg, Hans O. (1992), Unemployment: Micro- or Macro-theories?, http://www.oocities.org/hmelberg/papers/921120.htm]
[Comment: This paper is not the best I have ever written and it should not be read as a general overview of unemployment theories and their weaknesses. The reason why I have published it here is that it includes an attempt to distinguish between micro- and macro-theories which I still think is valuable.]

Unemployment: Micro- or Macro-theories?

The Argument
This essay will argue that the classification of unemployment under either a micro- or a macroeconomic heading is wrong. First of all the distinction between micro- and macro-economics is ambiguous and secondly unemployment is a complex phenomena caused by both micro- and macro-factors. Though, if one must to place the main emphasis on one factor, this essay would argue for a macroeconomic view. This argument will be presented in more details in the paragraphs below.

Defining unemployment
Before any discussion one must clarify the concepts involved. The official definition of unemployment is the proportion of the labour force which does not have a job and are actively seeking one. However, using this definition the concept of voluntary unemployment makes little sense (Can you be actively seeking a job you voluntarily do not want?). It has therefore become customary to speak about involuntary unemployment. This is defined as the proportion of the labour force which is actively seeking a job at the existing wage level, but unable to get one. Further problems with the definition of unemployment exists because of the discouraged worker effect (if one wants a job but is not actively seeking it because it seems hopeless) and underemployment (taking a job below your level of skills. Both mechanisms leads us to underestimate the true unemployment figure. The problems in defining unemployment are, as this paragraph has showed, significant and the distinctions should be kept in mind through the essay.

Micro- v. macro-economics
The distinction between macro- and micro-economics are thought to be more clear cut, but this also turns out to be a problematic issue. Macroeconomics is defined as the study of the economics as a whole i.e. the aggregates. Microeconomics, on the other hand, is defined as the study of individual units in the economy, in particular individuals and firms. The importance of a macroeconomic theory arises because of the effects of interaction between the units in the economy as for example in the Paradox of Thrift (when we all try to save more the effect is that we all become poorer. The is so because my income goes down when other people spend less (buys less of my goods). When people save more they spend less and hence I become poorer.) However, the distinction is theoretical and is becoming increasingly blurred as macroeconomics is build on microeconomics. The question is then whether the source of unemployment is to be found within the economics of the firm/the individual or if the source lies in the interaction between different units in the economy.

The Classical view: Too high wages
Traditionally the view has been that unemployment is a macroeconomic problem i.e. it has been looked upon in terms of aggregates. The Classical view was that involuntary unemployment was a short term phenomena resulting from a discrepancy between the price level and the wage level. Involuntary unemployment was the result of too high real wages. In time the wage level, in the Classical view, would be reduced and there would be no involuntary unemployment except for frictional search unemployment caused by time delays between quitting one job and starting another.

The Keynesian view
Against the Classical view, but still using a macro-view, there is the Keynesian's who argue that unemployment is a result of several factors classified as structural, seasonal, cyclical, frictional and demand-deficient. The most important is the demand-deficient which says that involuntary unemployment may exist even in the long run because of a lack of aggregate demand. The lack of aggregate demand leads to falling sales; falling sales leads to a fall in investments (and expectations); this leads to another fall in aggregate demand and more unemployment which leads to a further fall in aggregate demand etc. So the vicious circle in a free market economy continues unless the government intervenes. Because of the liquidity trap investments are not responsive to lower interest rates so not even the Pigou wealth effect can rescue the economy from the depression.

This Keynesian view is clearly a macro-view because it emphasises the paradoxical relationship between the different units in the economy. A single factory may sell more if it reduces its prices relative to other firms, but if all firms reduce their prices no one is better off relatively. In fact by cutting wages to reduce prices they reduce the aggregate demand for goods in the economy. Thus, the overall view of the economy is important to solve the unemployment problem, not the view of one unit (one factory).

Micro-theories
So far this essay has argued that the traditional view of unemployment is a macroeconomic view because it is coached in terms of aggregates. Against these traditional theories there are several other theories which emphasise the micro-economic aspects of unemployment. These aspect include problems with information, moral-hazard, adverse selection and uncertainty. The paragraphs below will present some of these theories.

1. Efficiency wages
One micro-economic explanation of unemployment is the efficiency wage theory. The rationale behind the theory is as follows. Assume that workers differ in quality; Not just abilities, but in the probability to shirk. Simply put some people are more lazy than others and are therefore less likely to work hard. Also the effort is a function of costly monitoring i.e. if you are being closely monitored you work harder than if you are not. Now, being an employer you care about the costs of labour (the wage rate). However, the cost is dependent upon the productivity of the worker. So your objective is not to minimize the wage, but the wage divided by productivity (wage pr. unit produced). To do this there are at least two options. Firstly, you can increase productivity by increasing monitoring. Secondly, you can increase productivity by increasing wages. The reason for this is that as wages increases the cost of shirking becomes higher (because if you are caught you are fired and lose your wage and the higher the wage is the more you loose by being fired. A higher wage thus means that you work even harder since it is more important for you not to be fired). Hence, there is a connection between the quality of workers and the wage rate. The higher the wage, the more costly it is to be fired and the less likely is it that the workers will shirk. Another argument using the same reason is that turnover itself is costly (firing, hiring and training) and consequently the employer would want to pay higher wages to prevent high quality workers from leaving.

Essentially the problem is one of moral hazard and asymmetric information in a principal-agent relationship. The employee would not shirk if he/she had to suffer the consequences always, directly. However, because the individual does not bear all the risk, his/her behaviour changes. This means the behaviour is state contingent which is the definition of moral hazard. The second problem is the informational asymmetry. The employee knows how hard he/she work, but the employer does not. Monitoring could reduce this asymmetry, but it might be very costly.

How does this theory explain involuntary unemployment? So far it has been established that it is profitable for an individual factory to offer higher wages than the market equilibrium. However, the factory is not alone in making this discovery. In fact the whole industry is likely to raise the wages to deter shirking. This creates a problem because if all raise wages then the individual advantage of higher relative wages for the firm is going to disappear. The solution to this problem lies in the creation of a permanent group of unemployed. The high real wage level creates an excess supply of labour. The excess supply does not result in a decrease in the wage level because they firms know they need some unemployment to provide an incentive for the employed workers not to shirk. This incentive is produced by making the cost of being unemployed high which is what a high unemployment rate reflects. Thus, the key to the story is that wages preform two functions. One as payment for the use of a resource and another as an incentive not to shirk. As a result of the second role of wages unemployment becomes an permanent equilibrium phenomena.

The theory of efficiency wages is at first sight appealing. It has a certain theoretical and logical "beauty." The problem in my opinion is that the "beauty" camouflages the problems with the theory. One criticism would be that labour mobility is not very high and therefore the need to increase wages to deter shirking might not be very important. The cost of losing your job is so high that there is no need to increase the wages to make the costs even higher. In the real world these costs include the costs of moving, the costs of being stigmatized as a "shirker" (reputation is important in a repeated game), the loss of personal relationships with fellow workers etc. The cost of a somewhat higher wage, I would argue, is only a small part of the cost of losing your job. Hence, there is enough incentive not to shirk without rising wages (which is the cause of involuntary unemployment).

A second problem for the theory, is the hysteresis effect. The creditability of the threat to fire workers if they are shirking, depends partly on the theory that firms can always hire another worker that could replace the one fired. If it cannot do so then the cost of firing is increased and consequently less likely. Now, being unemployed can easily make the worker less suited for work ie. decrease the quality of the worker. Especially if the worker is unemployed for a long time. This could include simple things such as acquiring "wrong" habits (getting up late), alcoholism, personal psychological problems etc. If this is true then the role of the unemployed group as an incentive not to shirk is reduced.

A third, and most serious problem, is the fact that firms do not represent a collective unit. Although it is profitable for firms as a group to create unemployment to deter the workers from shirking, it may not be profitable for a single firm to say no to a worker who offers to work at a lower wage rate than the existing.

These three problems do not mean that the theory is irrelevant, but that it might not paint a complete picture of the unemployment problem.

2. The Insider-outsider theory
An alternative micro-economic theory of unemployment, is the insider-outsider theory. The focus in this theory is the turnover costs of labour. This means that there are significant costs involved in firing, hiring and training workers. Not only are there exogenously determined costs, but the insiders can increase the costs of turnover by refusing to cooperate with hired outsiders (those who are unemployed). The insiders i.e. those who already have a job, gain market power over wages as a result of these costs. They then use this power to gain higher wages. The employers are willing to give the workers higher wages because this is more profitable than the costly process of turnover. There are, of course limits to this process. For example if the insiders are too demanding, then it becomes profitable to fire insiders and hire outsiders. However, as long as the insiders behave within reason they are able to raise the wage above the market level and thus ensure the continuation of involuntary unemployment.

One question one might ask is why not the outsiders set up a firm that could compete with the high cost insiders (because high wage). The answer to this is economies of scale. The insiders and their firms have easier access to credits, the advantage of experience ("learning by doing") and might engage in strategic action to deter entry. Thus, the threat of outsiders creating a competing firm is not large.

The theory also divides the market into the primary and the secondary sector. Clearly not all workers are in a position to "demand" what they want. The workers in the primary sectors are those who exert little market power. These are the jobs with relative low turnover costs. Usually they require little skill, they are often blue-collar and (as the theory would predict) low-paid. The secondary market on the other hand, is characterized by greatly specialised skills, white-collar and (again as the theory would predict) high pay. Thus, the theory does not pretend to explain the whole labour market, only the unemployment resulting from being outside in the secondary sector. The low turnover costs in the primary sector means that the theory has some difficulties with explaining the unemployment in this sector. One possible solution could be unions which increase the market power of the workers.

There are several problems with this theory too. One solvable problem is to determine who should be the insider and who should be the outsider. The solution is simply to make the insiders those with the highest signal value. For example, those with the best grades from university, or those who did best in an interview. Another more serious problem is, once again, hysteresis. One could say that by accepting the demands of the insiders and creating a group of outsiders, the firm is digging its own grave. This is so because the outsiders might, as explained before, become less employable as time passes. This again means an increase in the market power of the insiders. The logical conclusion is even higher wages and even higher costs. With even higher costs it is less competitive and consequently more likely to produce less profit. But, if there is some degree of competition this process is constrained. In fact if the firm faces strong competition there is little room for managers to "demand" even higher wages, Thus this theory does not explain how there could be high unemployment in highly competitive industries. A final weakness is the mentioned failure to explain unemployment in the primary sector where turnover costs are low. Clearly such unemployment exists, empirically speaking, but the theory here has difficulties in explaining why.

So far two main micro-economic theories of unemployment have been discussed. There are still a few others which will be explained briefly in the next paragraph.

3. Other theories: search theory, contract theory, bargaining theory
The "Search-theory" of unemployment argues that unemployment is a result of employers quitting to search for a new and better paid job. This involves a certain optimum time spend searching in order to find the best paid job. While searching the worker is unemployed. This seems to be a rather theoretical explanation of unemployment since only less than 10% of the unemployed actually voluntary quitted their own job. Besides there is little reason to quit your job before applying for a new job.

Another theory is the "Contract-theory" of unemployment. This theory argues that employment is the result of a contractual relationship which involves an insurance for the worker not to be fired easily and a fixed pay. As market conditions might change, these conditions still holds (in general, unless the external change is great). This means that wages are sticky and there is no reason why it should be in equilibrium. If for example external conditions indicated that the wage should be lower, this would not happen because of the contractual relationship and hence there would be a situation of excess supply ie. involuntary unemployment.

Finally there is the Bargaining theory which argues that unemployment is the result of a bargaining process between unions and employees. The unions have preferences for higher than market wages which results in a certain amount of unemployment. It should also be noted that these theories are not mutually exclusive, but might each explain different aspects of unemployment and therefore be complementary.

The speed of technological change
Personally I would favour a different theory of unemployment. One would of course accept that the micro-economic theories presented so far play a role, but in addition there are other factors. One such factor could be the observed "creativity" of a capitalist, free market economy. New inventions, new technologies, changing preferences and external shocks are constantly changing the structure of the markets. As a result of rapidly changing technology, preferences and exogenously determined state of nature conditions I would argue that there is a permanent mismatch of skills in the labour force. To take one simple example. When new technology made it possible to automatize the telephone network, thousands of workers became unemployed. Similarly when preferences change because of a changing age-structure (the population is becoming older on average, less children are born), there is no longer a need for so many new houses as before and hence not a need for so many construction workers. Hence, the dynamism of rapid change (which is characteristic of capitalism) will always produce a mismatch of skills which results in unemployment. The problem could be reduced if re-training was easy, but unfortunately this is not so.

Policy implications
What are the main policy implications of the theories? Regarding the use of unemployment benefits both the insider-outsider theory and the efficiency wage theory predicts that this increases unemployment, though for different reasons. In the efficiency wage theory the reason is that the cost of being fired is less so that shirking and firing becomes more probable. In the insider outsider theory the reason for the increase in unemployment is more workers would maximize their utility by being unemployed because this means an increase in leisure. When it comes to policy implications regarding reducing unemployment there is a clear difference between the theories.

Both theories favour micro-intervention (though different types of it) instead of large fiscal aggregate demand stimuli. However, the insider-outsider theory would be more favourable than the efficiency wage theory to demand stimuli because it predicts a reduction of unemployment as the result of a demand shock (reduces the number of outsiders), while the efficiency wage theory would predict small effects from a demand-shock. This is so because the efficiency wage theory always need an "army" of unemployed in order to provide an incentive for the employed not to shirk. These predictions again contrasts with Phelps search theory in which demand shocks increases the real wage level and makes the optimal search time shorter. A possible consequence in this theory is that the end result is a less efficient economy because the worker's skills and jobs are not as well matched as they would have been if the worker had searched for a longer time.

Conclusion
In conclusion one can must once again emphasise the complementary nature of the theories of unemployment. This implies that unemployment has more than one cause and hence more than one policy instrument must be used to reduce unemployment.

[Note for bibliographic reference: Melberg, Hans O. (1992), Unemployment: Micro- or Macro-theories?, http://www.oocities.org/hmelberg/papers/921120.htm]