Skip to content

Birks – Mankiw Chapter 28: Unemployment

↓ Jump to responses

A commentary on Mankiw Chapter 28: Unemployment (Mankiw 7th edition)

Mankiw, N. G. (2015) Principles of economics (7th ed.) Ch.28
Principles of macroeconomics
(7th ed.) Ch.15
Mason, OH: South-Western Cengage Learning


When reading the chapter, here are some aspects to consider:

1.Mankiw’s approach

Mankiw classifies unemployment into three main types, frictional, structural and cyclical. Be careful. The same terms are used elsewhere, but with different meanings. Note his definitions:

Frictional unemployment:

  1. unemployment that results from the process of matching workers and jobs, or
  2. unemployment that results because it takes time for workers to search for jobs that best suit their tastes and skills

The first description is in the text, while the second is highlighted in the margin. The wording is loose and they are not the same. Employers are also searching for suitable employees, but this is not mentioned in the second description.

Structural unemployment:

Unemployment that results because the number of jobs available in some labor markets is insufficient to provide a job for everyone who wants one.

This is further explained as wages being above the equilibrium level in that specific labour market. Three reasons are suggested, minimum-wage laws, unions, and efficiency wages. The implication is that, with sufficiently flexible wages, there would be no structural unemployment. There is an underlying (pro-market) ideological focus here. Shortages and surpluses only occur because prices are not flexible enough. Reasons not mentioned by Mankiw, but central to more conventional approaches, are changes in employment patterns due to changing technologies, resource endowments, consumer tastes, etc.. These might call for policies other than freeing up markets.

Cyclical unemployment:

The deviation of unemployment from its natural rate

To estimate the natural rate, Mankiw says that is what you have when there are frictional and structural unemployment. Deviations from this are then due to short term fluctuations in economic activity. This raises a question as to how are we are to measure the natural rate. What are the magnitudes of his type of frictional and structural unemployment? They may vary, in which case should we simply accept the “natural rate” as natural, even if it fluctuates?

2. The traditional approach

Mankiw’s approach differs from that found in earlier textbooks and several current ones also. The traditional definition of unemployment is “being able and willing to work at the going wage, but not in a job”. So a crucial issue is whether the going wage is appropriate. Do unemployment numbers change as you change the going wage? That question is not addressed in this approach. This is a weakness, but there is a practical reason for it. To analyse the real world unemployment position we have to use available data. Unemployment numbers reflect the situation at the prevailing wage rates.

There are advantages in the traditional approach. The types of unemployment in the traditional structure have the same terms as in Mankiw, but their meanings are different. Consider a situation where there are unemployed workers and there are job vacancies. The number of unemployed may not match the number of vacancies, and there are a number of sub-markets reflecting different skills, locations, etc.. For frictional unemployment it is possible to match workers and available jobs. They are for the right skills and are in the same locations. This would not generally be the case for all unemployed and vacancies. The traditional definition of structural unemployment then considers situations where it is not possible to match the unemployed to available vacancies. There are still vacancies, but not for the type of workers available. Hence workers could have the wrong skills (postal workers rather than teachers, say) or be in the wrong location (Mississippi instead of North Dakota [July 2014 figures]). Cyclical unemployment then considers the difference between the number of unemployed and the number of vacancies. Even if all vacancies were filled, there may still be some unemployed.

This is summarised in the following table:

Type of unemployment Vacancy exists? Vacancy matches?
Frictional yes Yes
Structural yes No
Cyclical No
  • With frictional unemployment there is a vacancy available for which the worker is suitable, but the unemployed worker and the prospective employer are not instantly matched up.
  • With structural unemployment there is a vacancy, but it does not match the unemployed worker (in terms of skills, or location, say). No agreement could be reached between the worker and the employer.
  • With cyclical unemployment there is no vacancy.

The classification suggests that policies to reduce unemployment should be as follows:

  • Frictional: improve the search process
  • Structural: change the characteristics (e.g. skills, location) of unemployed workers or unfilled jobs
  • Cyclical: create more jobs

Mankiw includes the traditional concept of structural unemployment under frictional unemployment. You can find this in section 2a, “Why Some Frictional Unemployment Is Inevitable”. This might explain the lack of consideration of a range of possible policy responses.

Commentary by Stuart Birks, last updated 17 November 2014


1 response

Respond to this commentary

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Please note that your email address will not be published.