NBER WORKING PAPER SERIES

A THEORY OF DUAL LABOR MARKETS

WITH APPLICATION TO INDUSTRIAL
POLICY, DISCRIMINATION
AND KEYNESIAN UNEMPLOYMENT

Jeremy I. Bulow

Lawrence H. Summers

Working Paper No. 1666

NATIONAL BUREAU OF ECONOMIC RESEARCH

1050 Massachusetts Avenue
Cambridge, MA 02138
July 1985

The authors are affiliated with the Graduate School of Business,

Stanford University and MBER, and Harvard University and NBER
respectively. We have benefited from useful discussions with James
Medoff, James Poterba and Andrei Shleifer. The research
reported
here is part of the NBER's research programs in Labor Studies,
Economic Fluctuations, and Taxation. Any opinions expressed are
those of the authors and not those of the National Bureau of
Economic Research.
 NBER Working Paper #1666
July 1985

A Theory of Dual Labor Markets

with Application to Industrial
Policy, Discrimination
and Keynesian Unemployment

ABSTRACT

This paper develops a model of dual labor markets based on employers' need

to motivate workers. In order to elicit effort from their workers, employers

may find it optimal to pay more than the going wage. This changes fundamentally

the character of labor markets. The model is applied to a wide range of labor

market phenomena. it provides a coherent

framework for understanding the claims
of industrial policy advocates. it also can provide the basis for a theory of

occupational segregation and discrimination which will not be eroded by market

forces. Finally, the model provides the basis for a theory of involuntary

unemployment.

Jerenr I. Bulow
Lawrence H. Sumniers
Center for the Study of
the Econonnj and the State Department of Economics
Harvard University
University of Chicago Cambridge, MA 02138
1101 East 58th Street (617) 495—2l4iL7
Chicago, IL 60637
(312) 962—8779
 —1—

Economists have a clear understanding of how

perfectly competitive labor
markets without any information problems would function. All workers would be

paid their marginal products. Wages of equally productive workers would be

equal and wage differentials would depend only on productivity differentials.

Involuntary unemployment would not be observed because of flexible wages.

Marginal workers would be indifferent to losing their jobs since wages would

just equal the amount they could earn pursuing alternative opportunities. With

perfectly competitive labor markets, subsidies to workers in specific industries

or interferences with free trade would be very likely to reduce economic

welfare.

Unfortunately, none of these characteristics of competitive labor markets

appears to hold in practice. Wages differ across workers in ways that are

inexplicable in terms of productivity differences. After controlling for both

measurable and unmeasurable differences in productivity, econometric studies

consistently suggest that factors such as occupation, firm size, race, sex, age

and tenure all influence wages. Involuntary unemployment is observed and is

found to. rise at times when productivity and real wages fall. Workers value

their jobs and regret losing them. Subsidies to workers in particular

industries, and interferences with free trade are pervasive around the world.

This paper examines one theory of dual labor markets which can account for

these phenomena.1 The deviation from the standard competitive model which

drives our analysis is the inability of firms to costlessly determine how much

effort their workers are putting forth. Firms can elicit more effort from their

1Shapiro and Stiglitz (1984) analyze some related issues with a model which
is formally very similar to the model presented in the first section of this
paper. Yellen (1984) provides an informal discussion of dual labor markets
paralleling parts of the formal discussion presented here. Our initial analysis
was developed before we became aware of these contributions. The idea that
 —2—

workers either by devoting resources to monitoring them more closely or by

raising the cost to them of being caught shirking. The latter objective may

require that firms pay workers more than their opportunity cost. A worker who

is paid only his opportunity cost has little incentive to avoid shirking. Firms

have a strong incentive to make workers value their jobs. They can only do this

by paying more than the "going wage." It is the deviation of workers' wages

from their opportunity costs that gives rise to our model's imperfectly

competitive features.

The potential importance of the linkages between the level of wages and

worker productivity which form the basis of our analysis is well illustrated by

one of the most dramatic chapters in the history of American industrial

relations. This episode also serves to highlight the importance of the effort

elicitation problem as an explanation for non—market clearing wages. In 1914,

the Ford Motor company introduced the five dollar day for industrial workers.

At the time prevailThg wages at Ford and other companies were between $2.00 and

$3.00 a day. Immediately following the announcement, more than ten thousand

persons gathered outside the Ford plant gates looking for jobs. Ford declared

that the motive for the wage increase was "profit sharing and efficiency

engineering".2 A contemporary engineering study of production at Ford, Arnold

and Faurote (1915) described the results of Ford's innovation:

firms need to elicit effort from workers can give rise to a labor market with
imperfectly competitive features has recently been explored in Calvo (1979),
Ishikawa (1984), and Bowles (1984). A particularly clear statement of the role
of non-competitive wages in eliciting effort may be found in Becker and Stigler
(1974). It dates back at least to Marx's "reserve army of the unemployed."

2Quoted in Nevins (1954), p. 538.

 —3-

The Ford high wage does away with all of the inertia and living
force resistance... The workingmen are absolutely docile, and it
is safe to say that since the last day of 1913, every single day
has seen major reductions in Ford shops labor costs.

Alan Nevins (1954) in his history of the early Ford Motor Company concluded

that:

Ford and his associates freely declared on many occasions that

the high wage policy had turned out to be good business. By this
they meant that it had improved the discipline of the workers,
given them a more loyal interest in the institution, and raised
their personal efficiency... Once the Ford factory, like others
had been called "The House of Correction"; now it was temporarily
called "The House of Good Feeling".4

The only available quantitative estimate of the effects of Ford's high wage

policy appears to be the calculation of Ford's chief of labor relations, John

Lee (1916) that productivity increased by 51 percent in 1914 following the

introduction of the high wage policy. One estimate, Levin (1927) suggests that

the high wage policy halved absenteeism. And Nevins reports that discharges

for cause declined very sharply because of the "instant and unquestioning

obedience of men eager not to lose their five dollar day."

The increases in productivity appear to be attributable almost entirely to

increases in worker effort. Nevins (1954) after reviewing the Ford archives

concluded that:

Suggestions that management was actuated by a desire to get the pick of

Detroit mechanics and by anxiety to end a high turnover rate are
demonstrably false and misleading. . .The mass production methods of the Ford
plant made it unnecessary to search for picked mechanics.. .the turnover
problem had been practically solved before the five dollar wage was
adopted. The theory (that management was trying to avert the threat
of unionization) is not supported by real evidence. The union had
shot its feeble bolt long before the wage decision.5

3Arnold and Faurote (1915), p. 331.

4pievins (1954), p. 549.

5Nevins (1954), pp. 551—554.

 —4-

The contemporary importance of variations in effort and the costs of

monitoring them may be conveyed in a number of ways. A large fraction of the US

labor force is involved in supervision rather than the direct production of

goods and services, and the maintaenance of worker morale is a major priority

for most firms. Productivity varies widely across time and space in ways which

are difficult to account for except in terms of differences in effort broadly

defined.6 An HEW task force on Work in America (1973) after reviewing hundreds

of studies of individual plants concluded that "lowering business costs by

reducing absenteeism,tardiness,turnover, labor disputes, sabotage, and poor

quality can increase productivity by up to 40 percent."

Many economists take the position that markets will tend to make efficient

micro-economic decisions about the compostition of national output, and the

means by which it is produced but that stabilization policies of some sort are

required to address the macro-economic problem of cyclical unemployment. An

alternative view is that recessions are just the most obvious symptom of market

imperfections which pervasively distort the allocation of resources. The model

presented here provides a justification for this alternative view. It shows how

involuntary unemployment can result from sources which also badly distort

microeconomic aspects of performance.

The paper is organized as follows. Section I lays out our basic model of a

dual labor market. It shows how equally productive workers can in equilibrium

be allocated arbitrarily between a primary and secondary sector of the economy.

6Nelson (1981) surveys evidence suggesting that standard variables leave a

great deal of productivity variation unexplained.
 —5—

These sectors parallel institutional descriptions of the dual labor market such

as that of Doeringer and Piore (1971). The primary sector is characterized by

high wages and responsible career jobs, while the secondary sector has menial

jobs, low wages and no job ladders. Although workers in the secondary sector

envy those in the primary sector, and are equally productive there is no

equilibrating market force which can erode wage differentials. The model also

explains why firms are extremely Conscious of their

relative standing in setting
wages, and how firms paying different wages can coexist in market equilibrium.

Section II shows that our dual labor market model provides a formal

framework which can justify the arguments of American industrial policy

advocates that the high wage/high value added sectors of an economy should be

subsidized and protected from foreign competition. Unlike most of the arguments

in the industrial policy debate, ours explains why workers and not just

imperfectly competitive firms may benefit from industrial policies.

Section III discusses theories of wage discrimination -- the observation

that workers' wages are affected by their demographic status as well as by their

productivity. Unlike standard theories of discrimination, the model presented

here explain why discrimination endures in spite of market forces tending to

eliminate it. Our theory of discrimination also provides an explanation for the

empirical observation that disadvantaged

groups receive "equal pay for equal
work" but unequal work. We also show that "affirmative actiont' policies which

subsidize desirable jobs f or members of disadvantaged groups are likely to raise

total welfare.

Section IV extends the model to provide a theory of involuntary

unemployment. We argue that it is a natural consequence of the rationing of

 —6-

primary sector jobs for incentive reasons. By postulatinf that the primary

sector does not hire workers who are employed but only those who are out of

work,we are able to show that involuntary unemployment with Keynesian features

will result. The model also provides the basis for theories of the cyclical

upgrading of secondary workers, the absence of work-sharing arrangements, and

real business cycles.

Section V concludes the paper by suggesting directions for further

theoretical and empirical research.

I. A Theory of Dual Labor Markets

Doeringer and Piore (1971) have described the American economy as having a

dual labor market. Jobs fall into either the primary or secondary sector. Jobs

in the primary sector are "good jobs" characterized by high wages, job security,

substantial responsibility and ladders where internal promotion is possible.

Jobs in the secondary sector are characterized by low wages, casual attachments

between workers and firms, and are menial. Workers in the secondary sector envy

those in the primary sector who have both better jobs and higher wages. A

typical example of a primary sector employer is a large manufacturing

establishment, while small service firms such as fast food outlets typify the

secondary sector.

As Doeringer and Piore and others have documented, these descriptions

accurately capture many aspects of the labor market. Yet they raise an obvious

question. If workers in the secondary sector envy those in the primary sector,

why are not wages in the primary sector bid down? The logic of competitive

economics denies the possibility of equally skilled workers receiving different

 —7—

wages in different jobs. The model in this section provides an explanation for

the existence of dual labor markets. The central idea is that primary sector

firms may find it advantageous to pay more than the "going wage" it helps them

in eliciting effort from their workers.

By interpreting the secondary sector in
the model presented below as home production, it can also be viewed as a theory

of involuntary unemployment.7

The Model

We assume that the economy is comprised of N identical infinitely lived

agents. Each can supply one unit of labor and produce w units of output in

either sector of the economy. Consumers maximize

lifetime welfare which is
given by:

(1.1) U = —r(v—t) dv
JtU(xiix2_as)e
where x. is the number of Units of sector i
output consumed in period t, r is
the discount rate, and s is an indicator variable equalling zero when the worker

works, and one when he shirks. Thus it is assumed that only two levels of

worker effort are possible. Shirking workers are assumed to produce no output.

Note that we have assumed that shirking and consuming extra units of x2 are

perfect Substitutes. This assumption simplifies the analytic treatment

considerably without altering any of the Substantive conclusions. Note that

in equilibrium no workers shirk.

We further assume that preferences are homeothetic and normalize so that

U(O,O)=o. Risk neutrality is also assumed. it follows that

7Such a model is provided

by Shapiro and Stiglitz (1984). We suggest an
alternative and we believe more satisfactory
theory of involuntary unemployment
in Section IV.
 -8-

U(Ax1,Ax2) = AU(x1,x2). Note also that our assumptions imply that all consumers

will consume the economy's two goods in the same proportion. We ssume that

consumers can neither borrow nor lend. As we discuss below, this assumption

restricts workers from posting bonds against the possibility of their shirking.

We normalize the price of secondary sector output, x2 to be unity. Using

the hometheticity of the utility function and production assumptions noted above

we can write:

x E E
(1.2) p1 = f() 2
=
2
=
1
i
g(E1); g < 0

where E. workers are employed producing wE units of output in sector i.

Competition will insure that secondary sector workers receive a wage equal

to their marginal product, w = w. In keeping with the discussion above, we

assume that secondary sector workers are monitored perfectly and thus have no

possibility of shirking. This reflects the idea that secondary sector jobs are

menial.

The key aspect of our model is that detection of shirkers in the primary

sector is difficult. This is a natural consequence of the responsible character

of primary sector jobs. Both false positives and false negatives may result as

firms try to detect shirkers. We assume that a worker who is not shirking has

an instantaneous likelihood d1 of being falsely labelled a shirker. A worker

who is shirking is assumed to have an instantaneous likelihood d2 of being

identified as a shirker. It follows that the probability of not being labelled

a shirker over an increment of time is (d2-d1)dt greater for those who are not

shirking than it is for those who are shirking. It is also assumed that all

employment relationships have an exogenous separation rate q. Separations occur

 -9-

either due to worker quits in order to relocate of withdraw from the labor force

or due to employer induced seperations caused by changing patterns-of product

demand.

We assume in the discussion below that firms are restricted to paying all

workers a common wage. Subsequently, more elaborate compensation schemes are

discussed. Firms in the primary sector need to induce their workers not to

shirk. The only penalty that they have at their disposal is the threat to fire

workers caught shirking, since by assumption workers cannot post a bond. It

will become apparent that it is never desirable to fire workers not caught

shirking. Denoting the present value of lifetime utility for workers in the

primary and secondary sectors respectively by PV1 and PV2, workers will shirk at

any given instant unless the condition

(1.3) a
(d2—d1)(PV1—Pv2)

is satisfied. The left hand side is the instantaneous gain in utility from

shirking while the right hand side is the product of incremental probability of

being dismissed if a worker shirks and the loss in lifetime utility from being

dismissed. The values of PV1 and PV2 may be calculated from the recursive

equations:

— -
—(r+q+d 1 )t -(r+q+e 1 )t
(1.4) PV1 - = (w1-) fe dt +
(q+d1)(Pv2—)70e dt

E
— E —(r +
2
(1.5) PV2 — = 1(q+d1)(PV1-) fe dt

Equation (1.4) holds that the present value of the surplus from a primary sector

job is the sum of the discounted surplus from the current job and the present

value of the future surplus if in the secondary sector, discounted to reflect

 -10-

the time until the secondary sector will next be entered. Since workers in the

secondary sector get a wage w, equation (1.5) is simply the present value of the

future surplus if in the primary sector discounted to reflect the time until the

primary sector is next entered. In forming equation (1.5) we have made use of

the steady state assumption that the flow of workers into and out of the primary

sector is equal, so that the flow rate out of the secondary sector is
2
Solving equations (1.3) through (1.5) yields the no shirk condition on

primary sector wages:8

w -w=
- ar a(d1+q)N
(1.6) — + - -E

Equation (1.6) has several plausible implications. As the utility from

shirking, a, increases, firms must pay more to induce their workers not to

shirk. As the probability of successfully detecting a shirker, d2 declines,

firms must also pay higher wages. Likewise, if the rate of turnover among

non-shirkers, d1 + q increases firms must pay higher wages. This is because the

value of maintaining a job is reduced if future turnover is more likely. The

greater the number of primary sector jobs, E1, the higher wages must be to

maintain the opportunity cost of losing a job, because the time a worker must

spend waiting to return to the primary sector if fired is reduced. Finally,

increases in r raise required wages because they reduce the present value of the

loss from being fired.

Notice that equation (1.6) implies that in equilibrium, primary sector

wages will exceed wages in the secondary sector even though all workers are

8Another way to derive (1.6) is to note that the difference in present

value of a primary and secondary sector job, PV1 -
PV2, equals
[w1 —
]/(r+d1+q+ (E1(d1+q)/N-E1]
 —11—

identical. Workers in the secondary sector will envy those •in the primary

sector, but it will not be possible for them to bid f or primary sector jobs by

being willing to accept lower wages. For if they accepted lower wages, they

would have an ircentive to shirk. Hence firms will not offer lower wages.

The model here implies that firms will be extremely conscious of relative

wages. Primary sector firms need to observe wages offered in the

secondary
sector in order to insure that their wages are set high enough so that their

workers have no incentive to shirk. In standard competitive models, firms need

only observe their own labor supply curves and have no need to learn about

either wage or employment levels at other firms. Indeed, one of the major

arguments for competitive markets as an allocation mechanism is that they reduce

the costs of information acquisition because each agent can act knowing only the

prices it faces.

In fact, firms are extremely concerned with assessing where they stand in

the wage distribution. The Handbook of Wage and Salary Administration (1984)

notes that "Salary surveys are indispensable to an effective compensation

program.. .The establishment of sound compensation policies depends on an

accurate assessment of an organization's place in the salary marketplace." It

goes on to give an idea of volume of survey information available

by noting that
there are almost 100 available surveys on the compensation of accounting clerks,

almost 50 surveys with salary data on the Chicago metropolitan area and 9

This is the incremental wage from being in a primary sector job, w1 -

discounted by the interest rate, the chance that a primary sector worker will
drop into the secondary sector, d1+q, and the chance that a secondary sector
worker will find employment in the primary sector,
E1(d1+q)/(N-E1).
 —12—

surveys with specific data on salary for accounting clerks in Chicago.9

Equation (1.6) also provides an explanation additional to standard human

capital arguments for the existence of enduring attachments between workers and

firms, which does not rely on the productivity enhancing effects of experience.

Reductions in the exogenous seperation rate q, reduce the wage which primary

sector firms pay in order to insure that workers will not shirk. Firms

therefore will have an incentive to minimize this separation rate. At a minimum

this means that they will not replace workers not determined to be shirking.

Firms also have an incentive to give laid off workers first priority for new

jobs, and to provide alternative jobs within the firm for poorly matched

workers.

Equation (1.6) provides one relation between primary sector employntent and

wages. A second relation is necessary to characterize the market equilibrium.

This relation (5) results from the requirement that workers in the primary

sector are paid their marginal product. It holds that

E
= =
(1.7) w = p1 wf(j1) wg(E1)
2

The determination of equilibrium by the intersection of these product market

equilibrium condition and .the no shirk condition is depicted in Figure 1. It is

apparent that in competitive equilibrium, identical workers will be paid

different wages. We defer until the next section an analysis of the efficiency

properties of this equilibrium. It is clear however that the composition of

output will be affected by factors such as the probability of detecting shirking

9Handbook of Wage and Salary Adminsitration (1984), p. 323.

 I
I
PMC
NS C I
I
I

Wages

Wi
I
I
I
I
W -J
I

E1

Employment

Figure 1

The Determination of Equilibrium

 —13—

workers, the interest rate and the utility workers get from shirking. In

equilibrium, of course no shirking will occur. -

The analysis so far has assumed that the only sanction available to firms

when workers are suspected of shirking is dismissal. One can imagine firms

developing various other means for inducing workers not to shirk. For example,

workers in primary sector jobs might be asked to post bonds which would be

forfeited in the event they were detected shirking. Since the probability of

detection is less than one, these bonds would have to exceed the value of the

output foregone because of worker shirking. Bonds of this type are not observed

in practice for two reasons. First, workers simply do not have the requisite

liquidity to post bonds. Second, and probably more important, third party

verification of detected shirking is impossible. Firms would have an incentive

to falsely label workers as shirkers in order to collect their bond. Even if

firms did not do this there would remain the problem of determining in which

employment separations the bond should be forfeited.

While we do not observe the posting of bonds in the actual economy, a

number of the features of actual primary sector firms may perform some of the

same economic functions that bonding might perform. In particular, these firms

are characterized by job ladders and rising age-wage profiles, while similar

phenomena are not observed in the secondary sector. As emphasized by Lazear

(1981) a rising age/wage profile can maintain the present value of a job as a

worker's years tonormal retirement diminish. As Medoff and Abraham (1981) have

argued, it is difficult to ascribe rising age-wage and tenure profiles to rising

productivity.10 Rising age-wage profiles are of course subject to the same sort

10lheir claim has been challenged by a number of authors, notably Braun

(1984). It seems unlikely, however, that all of the observed returns to
experience can be traced to this source.
 —14•-

of incentive problems as bonds with one important addition. Rising age-wage

profiles unlike posted bonds will encourage workers to stay with firms even when

they have higher productivity opportunities elsewhere. As long as firms cannot

fully solve the effort elicitation problem with

rising age-wage profiles, wage
differentials will persist.

It is noteworthy that job ladders appear in parts of the economy where

individual performance is difficult to disentangle but not in parts where it is

easily monitored. Thus John Dunlop (1985) writes "Capital intensive production

methods as in electric power generation, refineries and basic steel tend to be

associated with.. .lengthy promotion ladders and elongated pyramids, while more

labor intensive operations as in light assembly, grocery stores and sewing have

few promotion ladders and flat pyramids." As Doeringer and Piore (1971)

emphasize, promotion opportunities predominate in high rather than low wage

sectors.

We have already shown how taking account of the effort elicitation problem

can potentially explain two features of observed labor markets: the fact that

there are good and bad jobs held by equivalent workers, and the existence of

rising age-wage profiles in primary sector jobs. How important are these

phenomena empirically? One way of thinking about the question is the following:

How indifferent, even in good times, would the typical manufacturing worker be

to the loss of his job? In a competitive market, firms would pay workers no

more than their opportunity cost of working. Hence workers would be indifferent

to losing their job. This is not true in the environment described here where

good jobs are scarce and rationed.

 —15—

Intra Industry Wane Differentials

The analysis presented so far provides an explanation for inter—industry

and inter-occupation wage differences even for equally skilled workers. This

analysis is clearly relevant to empirical observations such as the often

emphasized finding that wages in US manufacturing exceed those in service jobs,

even after controlling for many measures of worker quality. It is also relevant

to the dualism that 'is said to characterize the labor market. Yet, a second

type of anomaly in the wage structure also stands out. There appear to be

substantial, persistent differences in wages within industries. Large

literatures document the correlation of wage rates with unionism (e.g., Freeman

and Medoff (1984)), and firm size (e.g., Brown and Medoff (1985))and there is

substantial idiosyncratic variance as well. The analysis of these wage

differentials poses a difficult issue. Leaving aside the question of why firms

choose to pay different wages, how do high wage firms remain viable in the

market place? The question is a critical one. For typical firms the wage bill

is 4 times profits. If a firm paid even 10 percent above the going wage its

profits would be reduced by 40 percent.

In part, we believe the answer lies in the monopolistically

competitive character of product markets as described in Weitzman (1984). But a

simple extension of our model can provide a further part of the explanation.

Imagine that the probability of detecting a shirking worker in the primary

sector is not given exogenously but depends on the expenditure of resources.

These resources are consumed according to a function $(d2), where 4 > 0. Then

primary sector firms will solve the problem:

(1.8) Mm: w + •(d2) s.t. No Shirk Condition

to produce output most efficiently. It is immediately apparent from the first

order conditions for the solution of (1.8) that first order changes in wages
 —16—

will have no effect on labor costs.11 This is dramatically different from the

standard model where wage changes translate directly into cost chaflges. It

follows that discrete changes in wages will give rise to much less than

proportional changes in labor costs. This helps to explain how high wage firms

endure. From the perspective of this model, the observation that firms in the

same industry pay different wages is no more mysterious than the obser ation

that their factor proportions differ. Wage premia in this model contribute

directly to productivity, just as do standard factors of production.

Indeed the minimization problem, (1.8), provides an explanation for the

union productivity effect discussed by Freeman and Medoff (1984). When wages

are increased, firms are able to conserve on resources devoted to monitoring

workers and so total productivity is enhanced. As predicted by this line of

argument, Brown and Medoff (1978) report some evidence suggesting that the

ratio of non-production to production workers is lower in the unionized than the

nonunionized sector.

Equation (1.8) also provides a basis for a theory of the firm size effect.

Large firms enjoy economies of scale and are able to get reduced prices from

suppliers but because of their size have less favorable 4(d2) functions. They

therefore substitute high wages for supervision. These ideas comport well with

a number of Brown and Medoff's (1985) empirical findings: (i) quit rates are

lower in large and unionized establishments; (ii) large employers have more

pronounced job ladders than do smaller firms and steeper age-wage profiles;

(iii) the wage differential between large and small firms decreases with skill

11The argument here is similar to but different from that of Akerlof and
Yellen (1984). They emphasize that inertia may not be very costly for firms.
Our point is that a variety of compensation-monitoring strategies will be about
equally attractive to firms. The argument here is more general than our model.
It will be valid in any setting where firms choose wages. Models based on
 —17—

level. It is greatest at the bottom of the hierarchy where differences in

monitoring technologies are likely to be greatest.

Wage Dynamics

In addition to contributing to an explanation of static wage differentials,

the model presented here can contribute to the explanation of aspects of the

dynamics of wage differentials that are otherwise puzzling. In a number of high

wage industries, in the US and Europe, notably steel, wage differentials

increased substantially at the same time that the market conditions facing the

industry deteriorated dramatically. In the US for example, steel wages rose by

35 percent relative to those in all manufacturing between 1970 and 1980. No

standard competitive or bargaining theory, would predict a dramatic increase in

relative wages in the face of declining product demand.12 One would expect

reductions in labor demand to lead to both lower employment and lower wages.

The actual outcome is suggested by the model presented here. An increase

in the probability that the employer will initiate a separation will necessitate

the granting of a wage increase in order to insure that the no-shirk condition

(1.6) is satisfied. Mathematically, a layoff probability would serve the same

function in (1.6) as a higher quit rate. Thus failing firms will find

themselves forced to raise wages creating the prospect of dynamic instability.

This may also partially explain why firms tend to give relatively little notice

turnover or labor quality considerations would yield parallel results.

12Lawrence and Lawrence (1985) discuss an "endgame" explanation for this

phenomenon. Their explanation depends critically on union bargaining strategies
while their empirical work suggests that it occurs about equally frequently in
the union and non-union sector. This suggests the possible relevance of the
explanation presented below.
 —18--

before plant closings.13

Summary

The analysis in this section suggests that in any economy where firms

cannot monitor workers perfectly, they will pursue policies which will cause

workers to value their jobs. A necessary consequence of workers being led to

value their jobs is wage differentials unrelated to skill differentials. The

contribution of this factor to wage inequality is difficult to gauge. But some

evidence suggests that arbitrary wage differentials not linked to differences in

skill may be of substantial importance. Studies consistently find that

differences in genes, parental upbringing, years of schooling and IQ all taken

together can explain only a very small part of the inequality 'in wages.14 As

Jencks (1972) and Thurow (1976) argue this suggests the possible importance of

luck in wage determination. The formulation presented here shows how luck can

affect wages in a competitive market.

A labor market where there are pervasive wage inequalities unrelated to

productivity differentials, is profoundly different from a competitive labor

market. In the next section we explore the normative properties of our model.

II. Trade and Industrial Policies

The previous Section provided a model that was consistent with the

many occasions struggling firms do extract wage reductions from their

workers. But the packages frequently include other elements operating directly
on the no-shirk condition. These may include job guarantees or profit sharing
plans which give workers an incentive to monitor each other.

14Results presented in Taubman (1977) estimates suggest that the expected

absolute difference in earnings between identical twins is about two-thirds as
great as between randomly chosen members of the population. Jencks (1972) reports
similar results for a host of other variables, including schooling and IQ.
 —19-

existence of both large inter-industry and intra-industry wage differentials

that are not readily explicable in terms of differences in labor qi.ality. The

standard competitive analysis of industrial subsidies and trade policies

presumes the absence of these differentials. It is natural to ask how this

analysis must be modified in the presence of endogenously generated wage

differentials.

American industrial policy advocates such as Robert Reich and Lester Thurow

take the position that nations should try to focus economic policies on

encouraging high value added sectors. These sectors, it is argued provide "good

high wage jobs," in contrast to the low wage jobs found in other sectors of the

economy. Industrial policy advocates credit much of the economic success of

Japan over the last 20 years to their successful encouragement of high value

added production. They are alarmed by what they see as the de-industrialization

of America and the consequent loss of high value added employment. While

sophisticated industrial policy advocates recognize the economic truth that a

nation must have a comparative advantage in something, they find little solace

in this. In George Meany's picturesque phrase, "you cannot have a successful

economy built on everyone doing everyone else's laundry."

From the perspective of standard neoclassical economic theory the claims of

industrial policy advocates are difficult to understand. Competition equalizes

the marginal productivities of all equivalent workers. There is no such thing

as a good or bad industry. Domestic industrial subsidies distort the

composition of economic activity without redeeming benefit. Foreign subsidies

should be welcomed by domestic consumers. These conclusions rest on the

assumption that competition equalizes wages across sectors. We shall see that
 -20-

the claims of incrtrial policy advocates become explicable in the context of a

model like ours with endogenously generated wage differentials.

It is apparent crom Figure 2 that a laissez-faire economy will not reach a

first-best outcome in a model like ours. The deadweight loss relative to the

first best is equal to the shaded area in the figure. A subsidy to primary

sector production, wh-:h had the effect of shifting the PMC locus to the right

would increase total economic welfare. Indeed, a subsidy large enough to shift

the PMC locus out sufficiently to cause primary sector employment to rise to X.,

would achieve the same pareto-optimal outcome that would be achieved under

laissez faire. Note that an equivalent outcome could be attained by subsidizing

primary sector workers and thereby shifting the NSC outwards.

A more subtle question is whether a subsidy to primary sector employment

which was financed by a lump sum tax would represent a pareto improvement.

Workers in the secondary sector would be taxed to subsidize workers in the

primary sector who were already better off than they were. Nonetheless, it

turns out that such a subsidy would represent a pareto improvement. This may be

seen as follows. The NSC implies that:

dPV1 - dPV2
(2.1) ds ds =0

where s is the rate of subsidy. A change in the subsidy rate must have an equal

impact on the present value of lifetime welfare for workers in both the

• secondary and the primary sector. Equating the present value of the gain in

economic welfare from a change in the subsidy, to the sum of the present value

gains for workers currently in the primary and secondary sectors workers we have:

(2.2)
w- dE1 = dPV1
+ dPV2
+ dE1
r E1 ds E2 ds (PV1-PV2)—-—
 '4-

Wages

Wi

E1 E2

Employment

Figure2
Subsidies and Economic Efficieçy
 —21—

dE
dE1
where we have used the -fact that — =-—
ds ds
Since we know that w1-w > -
-
r PV1 PV2 starting at a zero subsidy rate, it
follows that dPV1/ds =
dPV2/ds
> 0. A small subsidy will represent a pareto-

improvement. It is not difficult to show, however, that a subsidy large enough

to increase employment to its socially optimal level would reduce the present

value of lifetime welfare for workers in the secondary sector. At the

utilitarian optimum an increased subsidy to primary sector workers has a second

order effect on social welfare but the transfer effect is first order; thus

secondary sector workers must lose.15

In some respects these results parallel standard results, surveyed in

Bhagwati and Srinavasan (1983) suggesting that if wages are artificially high in

some part of an economy it is desirable to provide a wage subsidy to counteract

the distortion. There are three important differences. First the wage

differentials considered here are not arbitrarily fixed as in standard models

but instead are determined endogenously and depend on the relative sizes of the

two sectors. Increases in the subsidy raise wage differentials but are

nonetheless desirable. Second, the distortion considered here is not a result

of union power or other institutional forces. Unlike these cases, there is no

cause for concern that subsidizing the high wage sector will lead to greater

distortion. Third, our model permits us to assess separately the impact of a

subsidy on the lifetime welfare of workers in the primary and secondary sector,

because the flow between sectors is explicitly modelled.

The analysis so far has concentrated on the case of a closed economy. In

15These results have implications for the effects of other labor market
policies. A proportional or progressive income tax reduces the utility gain
from holding a primary rather than a secondary sector job, thus shiftirj the NSC
to the left and reducing economic welfare. Note that this effect is first order
 —22-

the case of an open economy, the differences between our formulation and the

'standard competitive one become much more dramatic. This is illustrated

heuristically in Figure 3•16 Imagine that the domestic economy is small and

takes prices as given on world markets. Figure 3 illustrates a situation where

the world price of the primary sector good, p. is less than the domestic price.

This may occur for any of the conventional "natural" reasons why foreigners

might have a comparative advantage in the primary sector, if foreign workers are

less prone to shirk, if foreign governments subsidize the primary sector, or if

foreign firms have found better contracting devices for inducing workers not to

shirk. Each of these possibilities has some relevance to the current U.S.

policy debate. The region ABDE represents the welfare loss to the economy due

to reductions in rents caused by the reduction in primary sector employment.

The angular area ABC represents the gain from free trade, as the price of

i, ,.orted goods is reduced. It is apparent that as long as the domestic primary

sector is "nearly competitive," allowing free trade will reduce national

economic welfare. The lost rents as primary sector employment contracts from OF

to 00 represent a greater loss than the second order to gain to consumers from

allowing free trade.

There is a simple intuition for this result. The gains to consumers from

unlike the second order welfare costs normally associated with labor income
taxes. The difference arises because of the pre-existing distortions present in
our model. Similarly, policies directed at compressing wage differentials will
in a model like this one have unfortunate consequences. If primary sector
workers cannot be bonded to work hard by high wages, the market equilibrium will
of necessity involve a large threat of being confined to the secondary sector.
This in turn requires an expansion in the secondary sector at the expense of the
primary sector.

16The graphical analysis is only heuristic because of its use of

consumers' surplus as a welfare measure and its neglect of the income effects of
policy changes on the demand for primary sector output.
 PMC NSC

Wages
I

I
I

I
p1 B C
I

I
w _____________

0 G F N

Employment

Figure 3

International Competition and Welfare

loss fromtrade gain from trade

1/1/,,,
 —23—

allowing free trade are of second order in the change in the relative price of

primary sector output. On the other hand, the gains in primary sector

employment from restricting trade represent a first order welfare gain because

of the wage differentials between the two sectors.

We can show this result formally by recogniz4ng that in a free trade

environment the no shirk condition (1.4) must still be met, but that product

market clearing is now given by:

(2.3) w1 =

where p1 is the world relative price of the primary good. Other conditions for

equilibrium include the trade balance equation

(2.4) p1wE1 + w(N—E1) = p1x1 +

where x. is domestic consumption of good i and

(2.5) =
g()
which is a result of our assumption of hornothetic tastes.

With the system of equations (1.6) and (2.3) to (2.5), we can solve for

E as a function of p1 to explore the impact of world prices on primary sector

employment and welfare. The result is:

dE
[(w1 -) + (wE -x
dp1
1 1
(26) r
dp

where W is the sum of the equivalent variations associated with a change in p.

The intuitive explanation of (2.6) is that a reduction in world prices has

two effects. The first term on the right side of (2.6) arises because primary
 —24--

sector size is constrained by a combination of the NSC and world prices; a lower

reduces primary sector employment, causing a welfare loss. In competitive

model where wages were equalized, this term would not arise. The second term

reflects the welfare gain from the lower foreign prices. It is equal to net

imports, (x1—wE1) times the price drop. It is clear that if the world price is

below but sufficiently close to the autarky price so that net imports are small,

the loss of primary job3 and their attached rents implies that welfare is

reduced by free trade. Furthermore, analysis that is substantively identical to

that of (2.1) and (2.2) shows that if world primary sector prices are below but

sufficiently close to autarky prices then free trade is pareto-inferior for

domestic workers; the gain to secondary workers in lower primary sector prices

is outweighed by the reduced opportunities for primary sector employment.

A second conclusion with clear implications for industrial policy debates

follows from comparing Figures 4a and 4b. The welfare gains per dollar of

revenue cost from a subsidy to the primary sector are greater in an open economy

that would have had an autarky price less than or equal to the free trade price

than they would be in a closed economy. This is because the demand for primary

sector output is perfectly elastic in the case of a small open economy.

Therefore, for any subsidy per primary sector employee more people will switch

to the primary sector, and marginal worker shifts will provide as much of an

income gain as inframarginal movements. Subsidies in a closed economy run into

diminishing returns. As primary sector output expands, it comes to be valued

less and less. In the open economy, when primary sector output can be exported,

there s no similar factor limiting the gains from subsidizing primary sector

output. 17

17Thjs proposition and the one in the next paragraph may be verified by
 /
PMC

s/
/
NSC

/ /
Wages

Wi

W -_
Einploynient

Figure 4

The Effects of Subsidies

a) autarky

PMC NSC

/
Wages

Employment

b) free trade
 —25-

An equally important implication of our model, which can be seen from

(2.6), is that nations should subsidize "winners" —- primary secthr products

where they have a comparative advantage. The greater is the world price

of primary sector output, the greater is the gain from increasing primary

sector employment, (w1—w). There is little advantage to be had from subsidizing

primary sector employment if the world price of primary sector output is close

to the secondary sector wage w. In that case, there is little loss from letting

workers export secondary sector output, and purchasing the

primary sector good from abroad. Where the world price is high, and the home

economy has a substantial comparative advantage, a subsidy yields the maximum

benefit.

The model presented here captures many of the prominent themes in the

industrial policy debate. There are good jobs and bad jobs even in equilibrium.

Good jobs have higher wages and higher average productivity. Value-added per

worker is higher in these jobs. Protecting sectors with good jobs can raise

total economic welfare. International competitiveness raises the stakes and

increases the payoff to primary sector subsidies. Previous models endeavoring

to illuminate the industrial policy debate have focused on the strategic

interactions of imperfectly competitive firms. Policies in some circumstances

will help local oligopolistic firms at the expense of foreign oligopolistic

firms. These seem less relevant to the issues debated than does our model which

focuses on the effects of industrial policies on the number of good jobs

calculating dE1/ds explicitly from equations (1.6) and (1.7) for the closed
economy and (1.6) and (2.3) for the open economy.
 -26-

available for workers.18

In models with competitive labor markets public interferences-in free trade

through protection or export subsidies are undesirable. Yet governments

regularly engage in these practices. The model presented here presents a

rationale for these actions. Obviously, an analysis of this kind taken alone

cannot justify any particular industrial policy. It does however provide a

basis for thinking about many of the issues raised by the industrial policy

debate.

III. Discrimination in the Dual Labor Market

The persistence of discrimination in competitive markets has been difficult

for economic theory to explain. As originally noted by Becker (1957) and

emphasized by Arrow (1973), one would expect discrimination to be eliminated by

competition. Firms that did not discriminate would have lower costs than firms

which did and would therefore grow tending to eliminate discrimination. With

perfect capital markets, capital would be put to its most profitable use by

ending up in the hands of non-discriminatory entrepreneurs. While enduring

discrimination can be rationalized by pointing to the tastes of customers, this

explanation seems implausible in many settings where discrimination is thought

to occur. Nor is discrimination easily explained by pointing to workers'

tastes, since firms comprised only of minority group members would have lower

costs than other firms. Furthermore, some evidence, Blau (1982) suggests that

men working in heavily female establishments actually receive lower wages

18The formal model that we have presented here does not include capital as a
factor of production. Adding capital as a factor does not change the
conclusions reached so far, but leads to another conclusion. Traditional
competitive models suggest that it is in the national interest to allow capital
to seek the highest return available either at home or abroad. Once wage
 —27—

ceteris paribus.19

Beyond their difficulties in explaining the persistence of dicrimination,

existing theories have a difficult time accommodating the phenomenon of

occupational segregation. Studies of discrimination, particularly sex

discrimination find that much of the discrimination takes the form of equal pay

for equal work but unequal work (e.g. Lloyd and Niemi (1979)). Doeringer and

Piore (1971) in their discussion of the dual labor market emphasize that members

of disadvantaged groups are confined to the secondary sector. Even granting the

existence of wage differentials arising from employer discrimination, it is hard

tosee why occupational segregation should result.

The dual labor market model developed here has as its central element wage

differentials that are unrelated to productivity differentials. It is natural

to suspect therefore that it can provide the basis for a theory of

discrimination. Indeed Yellen (1984) suggests that in an environment like the

one considered here employers can costlessly discriminate since there is excess

demand for primary sector jobs. This is not correct. If the equilibrium

allocation of workers in different groups who are otherwise identical differs,

firms can lower the wage offered to disadvantaged workers without fear that they

will shirk. As noted in Section 1, an increase in the fraction of workers

confined to the secondary sector reduces the wage that must be paid to satisfy

the no shirk condition. Competition will eliminate discrimination as firms hire

differentials are considered, this conclusion is no longer valid. Keeping

capital at home, and in the primary sector, may raise welfare by increasing
rents created by primary sector jobs.

19Phelps (1972) and others have suggested theories of statistical

discrimination. As emphasized by Aigner and Cain (1977) these models have the
disquieting implication that the difference in average wages between groups
exactly equals differences in productivity. Alternatively discrimination can be
based on differential measurement ability for different groups as in Lundberg
 -28-

labor as inexpensively as possible. Our model thus cannot explain

discrimination if it is assumed that there are no differences at al between

members of different groups.

However, our dual labor market model can rationalize discrimination if it

is assumed that there are group differences that are unrelated to productivity.

After showing how these differences can lead to occupational segregation and

wage discrimination, we investigate the welfare. effects of antidiscrimination

policies.

Turnover and Discrimination

A major characteristic of groups thought to be disadvantaged in the labor

market is that they have very high separation rate. For example, Poterba and

Summers (1984) estimate that the rate of labor force withdrawal is about 1

percent per year for males 25-59 but about 19 percent per year for women in the

same age bracket.2° Data in Marston (1976) suggest large age and race

differences 'in separation probabilities as well. These differences have

provided the basis for theories of occupational discrimination based on

considerations of human capital accumulation as for example in Mincer and

Polachek (1974). Here we present an alternative explanation of how higher

separation rates can lead to group differences in wages. In our formulation

experience has no effect on productivity. Unlike human capital based

explanations for wage inequality, our model suggests that anti-discrimination

and Startz (1982). This seems a rather weak reed on which to base a theory of
discrimination.

20Some empirical studies suggest that turnover rates do not differ across
groups after controlling for job characteristics. These findings are not
inconsistent with the analysis presented here. If as implied by our model, high
turnover workers are assigned by the market to certain types of jobs, then
 -29-

policies may well increase economic efficiency.

Taking account of the fact that there are two identifiable groups in the

population the basic equilibrium conditions require that workers be paid their

marginal products and that both men and women in the primary sector be induced

not to shirk. Thus equilibrium is characterized by the conditions:

(3.1) w = W9(Elm+Eif) =
Pl(Elm4Elf)
- ar +
ar(d1+q )Nm
(3.2) (w1-w) = d -d
2 1 'd2 -d1''E2m
'

-
(w1—w) =
ar +
ar(d1+qf)Nf
(3.2)
2

where the subscripts m and f are used to denote males and females, for

concreteness in the discussion. Note that competition insures that primary

sector men and women receive the same wages. However, comparing equations (3.2)

and (3.3) it is clear that a higher proportion of women will be confined to the

secondary sector if qf is less than Since women have a shorter horizon in

primary sector jobs, they must receive a greater inducement if they are not to

shirk. With equal wages, this can only occur if secondary sector women have a

smaller chance of moving to the primary sector than do secondary sector men.

These points are illustrated in Figure 5. As long as the NSC for women lies

above the NSC for men, there will be partial occupational segregation.

Figure 5 suggests an alternative theory of occupational discrimination. If

controlling for job characteristics is inappropriate. Moreover, an analysis

depends on workers' expected horizon in the labor force not on the current job.
Here the differences between groups are especially pronounced.
 I

NS Cf I

I
I
I
I
I
Wages I II

I
I
V — —J — —I — ———
I I ii
I I
I I
N E+E N+N
m mf
Nf

E
in

Employment

Figure 5

Discrimination in the Dual Labor Market

(drawn assuming Nf = N)
 -30-

groups differ in the utility they get from being in the secondary sector,

which may be taken to include home production, occupational segreg.ation will

result. The model captures a commonly observed aspect of firm behavior. Firms

prefer to give jobs to workers who "really need them" than to workers who gain

less surplus from holding them.

It is consistent with the observation emphasized by Blau (1982), marriage

raises the wages of men but reduces or has no effect on the earnings of women.

Marriage reduces male probability of employment separation but increases

females' separation probabilities. It is also plausible that marriage raises

the cost to men of losing a desirable job, but reduces the costs to women for

reasons relating to the analysis of Becker (1985). The differential effects of

marriage on men and women mitigates against theories of discrimination based on

employer prejudice.

Below, the welfare effects of anti-discrimination policies in this model

are considered. Before turning to this analysis it should be noted that our

framework suggests a number of other theories of discrimination. One such

theory might be based on the differential ability of different groups to enter

into efficient labor contracts in the primary sector. Population groups that

are liquidity constrained and therefore unable to accept low starting wages and

rising age-wage profiles will not be able to get primary sector jobs. The wage

they must be paid in order to get them not to shirk may well exceed the going

• primary sector wage. For example, historically some workers may not have been

able to afford apprenticeships.

An argument paralleling theories of statistical discrimination might run as

follows. Employers cannot accurately predict workers' productivity. Therefore

 -31-

they make use of a worker's demographic status in assessing likely productivity.

In a perfect labor market, workers in disadvantaged groups could be offered

primary sector jobs at low wages reflecting their low expected productivity, and

given a chance to prove themselves. In the environment considered here, there

would be no incentive for a firm to make such an offer since at low wages all

workers would shirk.

Anti-Discrimination policies

It is clear from the discussion in the preceding section that a policy of

subsidizing employment in the primary sector would continue to be desirable with

two classes of workers. A more interesting question involves a policy of

subsidizing the primary sector employment of disadvantaged workers, and taxing

the primary sector employment of advantaged workers in a balanced budget manner.

Equivalently, what are the normative properties of policies which require that

disadvantaged workers and advantaged workers be hired in some fixed proportions

in the primary sector? We examine these questions in the context of the model

presented above where discrimination arises from differences in separation

probabilities, and then consider other possible sources of discrimination.

We consider the effects of a tax at rate t on male employment in the

primary sector used finance a subsidy to women's employment. it follows that

the subsidy to women's employment is at rate ElM/ElF. In order to show that at

least a small tax is desirable, we proceed as follows. Since men and women have

the same utility function, a tax increase will be desirable if d(EIM+EIF)/dt > 0.

Substituting equation (3.1) into (3.2) and (3.3) and introducing the tax

into the latter two equations allows us to implictly differentiate (3.2) and

(3.3) to find d(EIM+EIF)/dt. After some tedious algebra this yields

 -32-

d(E1M+E1F) — ar 1 (d1+q)NE1 (d1+qf)NfE1f

— 1 — t(EIM+Eif)
dt
d2—d1 E1f (N—E1 )2 (Nf_E 1 f) if
(34)
+

Att=O (d1+q)N (d1+qf)Nf

NM E1M Nf_E1f
So

ar 1 (e1+q)N ElM E1f

d(E1M+Elf) = d2—d1 Ef NM_ElM [NM_ElM Nf-E1f1
—
(3.5) >0
dt
IAI

Equation (3.5) is clearly positive, because the ratio of male primary

participation, ElM/(NMElM), exceeds the ratio of female primary sector

participation at a tax rate of zero. Its implication is that a small subsidy

to women will increase welfare.

However, we can also calculate whether an optimal policy would produce a

large enough subsidy to drive the female primary sector participation rate up to

the male level. At the point where ElM/NM = we find

E1f/Nf

ar Nf t(E1M+Elf)
d -d
•
2
(-f) - 2

ii
2 1
(3.6) d(E1M+Eif) -
— (Nf_E1f) E1f
dt

Because qf > expression (3.6) must be negative. Thus, starting with t

positive to the point where the sectoral composition of employment is the same

for men and women, a reduction in the subsidy to the hiring of women will raise

welfare. A policy intermediate between the laissez faire outcome and the

elimination of apparent discrimination is optimal in the sense of maximizing the

 —33-

sum of utilities.

These results may be justified intuitively. It is clear fro,. Figure 5 that

in the absence of any subsidy the NSC for women is less steep than the NSC for

men. Hence a policy of taxing each man in the primary sector $1 and subsidizing

each woman in the primary sector by $E/Ef will bea net revenue raiser and will

increase total employment. If no revenues were raised, but all the proceeds were

used to subsidize the employment of women, the employment gains would be even

greater. On the other hand, if the share of men and women in the primary sector

is equalized, equations (3.2) and (3.3) imply that the NSC is flatter for men

thn for women. This is because a small increase in wages raises the present

value of a job by more for men than for women because of their longer horizon.

In this case, reducing the tax on men and the subsidy to women raises total

employment.

It should be clear that equivalent results hold for affirmative action

policies. Each possible choice of t is equivalent to a policy mandating that a

specific proportion of women be hired in the primary sector. If policy is

carried on through such quotas, the wages of women will rise above average

primary sector wages, and male wages will fall. A quota policy which reduces

male wages by t percent relative to average primary sector wages is exactly

equivalent to a t percent tax on male employment with revenues used to finance

subsidies to primary sector women.

An alternative policy which might be labelled "comparable worth" would

artificially compress the wage differential between the primary and secondary

sectors. This would be disastrous for both men and women. With. a compressed

wage differential between the two sectors, the no shirk condition would require
 -34-

reduced primary sector employment of both men and women.

These results pertain to a model where apparent discrimiriatiQn results from

differences in turnover probabilities across groups. We have also considered

the case where discimination results from one group getting more utility from a

secondary sector job than the other. In this case, the effect of a subsidy on

total welfare even starting at t = 0 is ambiguous. This is because a subsidy in

addition to raising total employment which increases welfare causes the

substitution of women for men in primary sector jobs. This reduces welfare

under the maintained assumption that women value primary sector jobs less than

men. We have not yet considered the welfare effects of affirmative action

policies in situations where discrimination results from group differences in

the ability to post bonds or is statistical. We suspect that it will be easy to

construct examples where it will be welfare enhancing.

Although primary sector jobs are rationed in our model, we have so far

assumed that no rent seeking behavior takes place. In a sense therefore the

allocation of workers to the primary sector is arbitrary. In the remainder of

the paper, we relax this assumption and provide a theory of unemployment.

IV. Unemployment in the Dual Labor Market

Most economists accept the idea of involuntary unemployment. Some workers

who would like to work at the going wage are unable to find work. Yet as many

critics have pointed out, the concept of involuntary unemployment involves an

essential ambiguity. What is meant by the going wage? The option of

self-employment is always open to the unemployed. And even in times of

stagnation, some jobs are available. We propose here a formulation which

 -35-

resolves these ambiguities. This is done by assuming that primary sector

employers only hire Workers directly from the pool of unemployed workers. This

generates "wait unemployment" as workers queue for primary sector jobs.

Imposing the requirement that workers must be unemployed before they can

accept primary sector employment, we have the equilibrium conditions:

(4.1) w1 = wg1(E1/E2)

(4.2) a = (p1)(d2d1)
d1
+ q + r

(d1+q)E1 (p1—1)
(4.3) w
N-E1—E2

recalling from before that the utility from being unemployed (having no income)

has been normalized to zero.

These equilibrium conditions are easily interperted by recognizing that in

equilibrium, secondary sector Workers must be indifferent to becoming unemployed

and unemployed workers must be indifferent to accepting secondary sector jobs.

Equation (4.1) simply determines the relative price of primary sector output.

Note that we can no longer write it as g(E1) because now some workers are

unemployed. Equation (4.2) holds that the utility gain from shirking in the

primary sector must equal the present value of holding a job in the primary

sector. This is most easily reckoned by considering the alternative of being

permanently confined to the secondary sector. Equation (4.3) holds that the

difference in instantaneous utility between holding a secondary sector job and

remaining unemployed equals the instantaneous probability of obtaining a primary

sector job times the present value of such a job. This formulation differs
 —36-

somewhat from that of Harris and Todaro (1970) who assume that workers must be

indifferent between secondary sector employment and locating in the primary

sector with proportional chances of being employed and unemployed.

The determination of equilibrium is depicted in Figure 6. Note that

equilibrium requires both that the PMC and NSC cross and that secondary sector

workers be indifferent to being unemployed. The unemployment in Figure 6

corresponds to" Keynesian unemployment" in three senses. First, it is

involuntary as workers are queuing for primary sector jobs, and are identical to

primary sector job holders but cannot bid down wages and get primary sector

jobs. It is voluntary only in the sense that all unemployment is voluntary --

some undesirable jobs are available.

Second, relative to the first best the level of unemployment is too high

and the level of primary sector employment is too low. As we discuss more

formally below, subsidies to employment or taxes on unemployment will increase

welfare.

Third, if productivity varies, unemployment, wages and aggregate

consumption will all move pro-cyclically as long as product demand is not too

elastic and/or the primary sector is not too small. This proposition may be

examined by substituting (4.1) into (4.2) and (4.3) and implicitly

differentiating the latter two equations with respect to w. We find:

(d1+q+r) N-E1-E2 - E1+E2

(p1—1)[
'1 '' E E
2
d(E 1+E2)
=
I 2
(4.4)
dw
p(d1+q+r)wN

(d1+q)E1E

The denominator of (4.4) is negative (because p < 0) so whether total

 PMC
I
NSC
I

I
Wi
I

Wages

N
N-E1—E2U E2

Employment

Figure 6

The Determination of Equilibrium Unemployment

 —37-

employment E1 + E2 rises depends on whether

(d1+q+r) N—E1-E2 (E1+E2)

+P <0
d +q E2
2
1 2

If p — (consumers have very inelastic relative demands for the two

goods), then and E2 both rise proportionately and N - - E2 falls. But if

0 so that relative prices will not change much, the primary sector will

expand in response to a positive productivity shock while constant relative

prices will imply the same ratio of N-E1-E2/E1 as before the shock also

unemployment will be increased.

If the primary sector already dominates the economy (E11E2 large) then an

increase in w will, by its reduction in the primary sector unemployment rate,

raise total employment. If, however, the economy is dominated by the secondary

sector, the principle effect of a shock will be to move workers into the primary

sector where the unemployment rate is higher.

The Cyclical Behavior of Labor Markets

The response of the economy considered here to productivity shocks

corresponds well to the cyclical behavior of actual labor markets as summarized

in Okun (1982). First, Okun noted that quits are pro-cyclical while employer

initiated separations move counter-cyclically. This is predicted by our model.

Positive productivity shocks lead to quits as workers leave the secondary

sector, while negative shocks lead to job loss as the primary sector contracts.

Second, Okun observed the shrinking of relative wage differentials at cyclical

peaks. In the model here, positive productivity shocks lead to equal absoi
 -38-

increases in wages in the two sectors. This corresponds to a narrowing of

relative wage differentials. Third, he emphasized the enduring nature of

attachments between workers and firms. As emphasized in Section 1 our

formulation provides an explanation for firms use of temporary layoffs and for

labor hoarding. Both serve to increase workers' horizon and therefore limit the

wage that must be paid to insure that workers do not shirk. A fourth

observation most sharply distinguishes our model firms from alternative accounts

of employment fluctuations.

The available evidence suggests that the vast majority of cyclical

variations in labor input reflect variations in the number of persons working

rather than the number of hours worked per person. Worksharing arrangements in

the face of downturns in demand are a rarity in the American economy. While it

might be argued that technological considerations make short hours infeasible as

a worksharing device, it is hard to see why employers who must reduce the size

of their work force cannot use rotating layoffs as a worksharing arrangement.

As Okun observes, the absence of worksharing antedates any incentives for full

layoffs provided by unemployment insurance and any important effects of

unionization.

These observations are troubling for many accounts of cyclical

fluctuations, it is hard to square discontinuous movements into and out of full

time employment with utility maximiation by a representative consumer.

Worksharing arrangements are a natural device for performing the insurance

function which is central to many implicit contract theories. And formulations

which focus on considerations of fairness and equity in determining labor market

practices would tend to predict the evolution of worksharing arrangements.

 —39-

The absence of worksharing arrangements is predicted by the model

considered here. Assume that a worker holding a fraction (1-z) of his time to

searching for a full time job, and assume that his search efficiency is 1-z

times that of a full time worker. Available evidence on job search, Clark and

Summers (1979), suggests that in fact part time primary sector workers could

probably search about as efficiently as the full time employed. Assuming that

dismissal rates are proportional to hours on the job (so that a nor-shirking

part-timer is less likely to be erroneously caught shirking than a non-shirking

full timer) but that quit rates are related to calendar time on the job, the

no-shirk wage per unit of time worked for a z-time employee, w, in an economy

where all other primary jobs are full-time is given by:

(1—z)E1 (1—z)E1
+ +
r + e (z N-E N-E
-
(4.5) w — w = [
(w1—w)
z(r+d1q)

The crucial insight is that both a full—timer and a part—time worker get

the same benefit from shirking for an hour, and each is as likely to be caught.

Therefore, to prevent the part-timer from shirking the employer must offer him a

job which is equally valued (has the same present value) as a full-time job.

Such a constraint necessarily implies that the part-time employees will require

higher hourly pay than the full-timers to prevent shirking.

An important feature of actual unemployment is that it is strongly serially

correlated. While we have not completed a full dynamic analysis of our model,

it is clear that temporary productivity shocks can give rise to serially

correlated movements in unemployment. Consider for example an increase in

productivity that is known to be temporary. It is apparent that primary sector

 -40-

employment must move continuously after the productivity shock is observed. If

a discontinuous movement were anticipated when productivity reverted to its

normal level, the instantaneous separation rate would be arbitrarily high and so

workers would shirk unless wages were infinite. It follows that a temporary

productivity shock would give rise to an immediate increase in primary sector

employment followed by a slow decline. Notice that primary sector wages would

rise more and employment less for a temporary productivity shock than for a

permanent productivity shock. This is because the knowledge that the shock was

temporary would reduce workers' horizons.

Stuctural Unemployment

While our model is able to account for a number of features of cyclical

fluctuations, it does not provide a satisfactory eplana'cion of the driving force

behind fluctuations. The work of the real business cycle school

notwithstanding, we find it implausible that productivity shocks are the primary

cause of cyclical fluctuuations. Aggregate demand, which cannot be defined in a

real model like ours with competitive product markets, seems to matter in

cyclical fluctuations. We return to this issue in the concluding section of the

paper. It may be more appropriate to think of our model as providing insights

into the determinants of "natural" or "structural" unemployment. Indeed our

model suggests some hypotheses about the dramatic increase in structural

unemployment observed in Europe over the last decade.

Consider first the effect of an increase in social insurance benefits.

Higher benefits may be thought of as corresponding to an increase in utility

from being unemployed a tax on wages. Raising unemployment benefits increases

 —41—

the incentive of secondary sector employees to become unemployed. This effect

is accentuated by the taxes necessary to finance insurance benefits. Reduced

secondary sector output leads to a reduced relative price for primary sector

output and lower primary sector wages. This in turn causes primary sector

output to contract, further increasing unemployment. The analysis parallels the

analysis of a negative productivity shock exactly except for the further

distortions caused by the taxes necessary to finance unemployment insurance.

Notice that the adverse effect of unemployment insurance on unemployment

occurs despite the maintained assumption that it does not affect workers' search

decisions and that its financing does not affect employers' decisions about

layoffs. Rather, its effect occurs primarily through firms' wage setting

behavior. This model provides a possible basis for suspicions about the adverse

effects of social insurance programs on macroeocnomic performance that is

consistent with the relatively modest effects typically found in microeconomic

studies.

A second implication of our analysis is that policies directed at

and reduce
increasing job security may substantially increase unemployment

primary sector employment. Such policies may be thought of as reducing d2, the

likelihood that a shirking worker will be detected and laid off. As we have

already seen, policies directed at worksharing may be counterproductive. If as

a result of worksharing arrangements, primary sector firms must raise their

• wages, primary sector employment will decrease and unemployment will rise.

Third, our model suggests that public or private policies directed at

compressing wage differentials may have deleterious effects. Equation (4.2),

the No Shirk condition, pins down the relationship between primary and secondary
 —42—

sector wages which must exist if primary workers are not to'shirk, as long as

secondary sector jobs are freely available. If this wage differertial is

reduced, primary sector firms will not be able to operate because the No Shirk

condition cannot be satisfied.

V. Conclusions

We believe that the analysis in this paper suggests the applicability

of efficiency wage models to a wide variety of problems in labor economics.

They can contribute to the understanding of many phenomena that are difficult to

acount for in other ways. And they have very different normative implications

than do other labor market models. Taking account of the effort elicitation

problem calls into question the general proposition that private labor market

arrangements are efficient. It also complicates the analysis of various labor

market policies.

A good example is provided by minimum wage laws. Taking account of the

effort elicitation problem introduces several new elements into the analysis of

minimum wages. First, minimum wages may interfere with efficient contracting

leading to unemployment and allocative distortions. If minimum wage laws

constrain firms so that they cannot offer age-wage profiles which are as steep

as they would like, they will have to raise the total level of compensation in

order to insure that workers do not shirk. Second, minimum wage laws have a

potentially beneficial effect which could conceivably offset the distortions

they create. Because they will in general create some unemployment they raise

the cost to being laid off from the primary sector. This will permit lower

primary sector wages and greater primary sector employment. Third, minimum wage
 -43-

laws may have a second potentially beneficial effect in raising productivity in

firms affected by the minimum. Such firms will be able to reduce.the resources

which they devote to the supervision of workers. If as we argue below

competitive equilibrium is characterized by excessive supervision, this may

enhance social welfare.

Beyond their application to various labor market issues, there are a number

of important directions for future research on efficiency wage models. We

consider them in increasing order of importance. First, we and other writers on

efficiency wages have made strong assumptions about the nature of compensation

arrangemerts. It is theoretically possible that compensation arrangements can

be devised which obviate the need for firms to offer more than the going wage in

order to insure that workers do not shirk. It would be valuable to model

explicitly the impediments to these arrangements created by the problems of

worker liquidity, moral hazard on the part of firms, risk aversion on the part

of employees, and inabilities of firms to make certain types of commitments.

The observation that many workers are paid more than their opportunity cost

requires rationalization.
firms'
A related question emphasized by Shapiro and Stiglitz (1984) concerns

choice of a supervision technology. It is easy to see that models of the type

discussed in this paper suggest that the architecture of economic organizations

will be flawed. There will be " too many chiefs and too few indians". Firms

will invest in supervision until the marginal effect of a dollar spent on

supervision equals the marginal effect of a dollar spent on higher wages in

reducing workers' incentive to shirk. Extra wages benefit workers but extra

monitoring does not so there will be more supervision than is socially

 -44-

efficient. Policies which encourage firms to give workers more responsibility

and to raise their wages will increase social welfare. Of Course_they may

reduce the welfare of those with human capital which is specialized towards

supervisory ability.

A second direction for future research concerns the sources of the linkages

between wages and productivity. In keeping with the work of others, we have

emphasized the effects of increases in the wages firms pay on the level of

effort that they are able to elicit from their workers. The mechanism we have

stressed is the role of the fear of job loss in inducing workers not to shirk.

This may not in fact be of central importance in wage setting. Textbooks on

personnel management, businessmen and union leaders all stress that equitable

and fair wage policies are critical to motivating workers. While standard

neoclassical economics has no role for income effects in the theory of the firm,

the question of how much firms can "afford to pay" plays a prominent role in

firm wage setting decisions and in collective bargaining. And the

determination of the relative wages paid to workers in different jobs is a major

preoccupation of actual wage setters.

Workers effort level may depend more on whether they feel their wage is

just than on its absolute level. Depending on how "justice norms" are formed,

results similar to those produced by our formulation may result. Even in the

face of unemployment or laborers willing to work at wages below current levels

firms will be reluctant to cut wages for fear of alienating current employees.

Inter-industry wage differentials may persist because they have become

established, and the costs of changing them for any one firm may exceed the

benefits. It is also clear that justice norms may influence the type of workers
 -45-

hired by firms.

An important direction for future theoretical work involves explaining

cyclical fluctuations in output, that are driven by aggregate demand shocks.

One promising avenue towards this objective involves the "menu costs" arguments

of Mankiw (1985) and Akerlof and Yellen (1985). These authors show that if

there is any cost to changing prices and if as in our model, the economy does

not attain the first best level of employment, expansionary policies will have

real effects and may raise welfare. Small nominal shocks will have real effects

because it will not be worthwhile to change wages or prices.

An alternative direction for elucidating cyclical fluctuations would

involve introducing monopolistic competition and increasing returns into the

analysis of the primary sector along the lines of Kiyotaki (1985). In such a

model, increases in current aggregate demand will have real effects, similar to

those of a productivity shock. The demand for primary sector output will

increase due to the increase in aggregate demand. Expansion of the primary

sector will reduce unemployment, the size of the secondary sector or both.

Finally, our formulation provides a channel through which monetary policy

can influence real output. If monetary policy can affect the interest rate, it
I
will effect the wages firms must pay in order to insure that their workers do

not shirk. Increases in real interest rates reduce the discounted value of

losing a job tending to increase workers' incentive to shirk. This leads to

higher wages which in turn reduce primary sector employment. Thus purely

monetary changes can affect the level of unemployment as long as they affect

interest rates. The empirical importance of this channel of monetary policy

seems questionable, however.

 -45-

Probably the most important direction for future research in this area

involves empirical testing of alternative hypotheses about wage differentials.

Using recently developed techniques and data sets, it should be possible to

determine the extent of variations in occupational and industrial wage

structures which cannot be attributed to unmeasured quality differences or

compensating differentials. It would also be valuable to relate measures of

worker productivity to their wages. At this point we cannot judge whether

Fordts experience in instituting the $5 day was aberrant. If not, there is a

strong presumption that many firms are or could profit from offering their

workers non-competitive wages. Opportunites for empirical work would be greatly

enhanced if satisfactory measures of monitoring costs could be constructed. A

final direction for empirical tests of these ideas would involve examining the

effects of the length of worker horizons on performance. Clearly such an

investigation would have to confront the endogeneity of workers horizons on

their jobs.

None of these empirical tests are likely to be decisive. Rationalizations

can be adduced for most any labor market outcome without resorting to models of

the type considered here. Ultimately the case for the utility of models based

on the effort elicitation problem, must rest on the common explanation they

provide for a very wide range of phenomena. Our investigation convinces us that

models based on the incentive effects of non-competitive wages provide the best

available formulation for understanding deviations of observed labor markets

from the competitive ideal.

 —47-

References

Aigner, D.J. and G.C. Cain, "Statistical Theories of Discrimination in Labor

Markets," Industrial and Labor Relations Review 30, 2 (January 1977),
pp. 175-187.

Akerlof, 0. and J. Yellen, "The Macroeconomic Consequences of Near Rational

Rule of Thumb Behavior," unpublished, 1984.

Arnold, H. L. and F.L. Favarote, FORD METHODS AND THE FORD SHOPS (New York)
1915.

Arrow, K., "The Theory of Discrimination," in 0. Ashenfelter and A. Rees,

Eds., DISCRIMINATION IN LABOR MARKETS (Princeton, New Jersey: Princeton
University Press) 1973 3-33.

Becker, G., THE ECONOMICS OF DISCRIMINATION (Chicago: University of Chicago

Press) 1957.

Becker, 0., "Human Capital, Effort and the Sexual Division of Labor," Journal
of Labor Economics, January 1985.

Becker, 0. and 6. Stigler, "Malfeasance and the Compensation of Enforcers,"

Journal of Legal Studies, 1973.

Bhagwhati, J. and T.N. Srinivasan, LECTURES ON INTERNATIONAL TRADE THEORY

(Cambridge, Mass.: MIT Press) 1983.

Blau, F., "Discrimination Against Women: Theory and Evidence," in LABOR

ECONOMICS: MODERN VIEWS (Boston: Martinus Nijhoff) 1983.

Bowles, S., "The Production Process in a Competitive Economy: Walrasian,

Neo-HobbeSiafl and Marxian Models," American Economic Review, March 1985.

Brown, C. and J. Medoff, "The Employer Size—Wage Effect," unpublished, 1984.

Brown, C. and J. Medoff, "Trade Unions in the Production Process," Journal

of Political Economy, 1978.

Calvo, 6., "Quasi-Walrasian Theories of Unemployment," American Economic

Review, May 1979, pp. 102-107.

Clark, K. and L. Summers, "Labor Market Dynamics and Unemployment: A

Reconsideration," Brookings Papers on Economic Activity 1979:1, pp. 14-60.

Doeringer, P.B. and M.J. Piore, INTERNAL LABOR MARKETS AND MANPOWER ANALYSIS
(Lexington, Mass.: D. C. Heath and Co.) 1971.

Dunlop, 3., "Industrial Relations and Economics: The Common Frontier of

Wage Determination," unpublished, 1985.
 -48-

Freeman, R. and J. Medoff, WHAT DO UNIONS DO? (New York: Basic Books) 1984.

Gintis, H. and 1. Ishikawa, "Wages, Work Discipline and MacroeconQmic

Equilibrium," unpublished (1984).

Hall, R., "The Rigidity of Wages and the Persistence of Unemployment,"

Brookins Papers on Economic Activity 1975:2, pp. 301-335.

Harris, J. and M. Todaro, "Migration, Unemployment and Development: A Two

Sector Analysis," American Economic Review, March 1970, pp. 126-143.

Jencks, C., INEQUALITY, 1972.

Lawrence, C. and R. Lawrence, "Relative Wages in U.S. Manufacturing: An

Endgame Interpretation," forthcoming in Brookings Papers on Economic
Activity 1985:1.

Lazear, E., "Agency, Earnings Profiles, Productivity, and Hours

Restrictions," American Economic Review, September 1981, 606-620.

Lee, J.R., "So Called Profit Sharing System in the Ford Plant," Annals of
the American Academy of Political and Social Science, May 1916,
pp. 297—310.

Lloyd, C. and B. Niemi, THE ECONOMICS OF SEX DIFFERENTIALS (New York: Columbia
University Press, 1979.

Lundberg, S. and R. Startz, "Private Discrimination and Social Intervention in

Competitive Labor Markets," American Economic Review, June 1983,
pp. 340—347.

Marston, S., "Employment Instability and High Unemployment Rates," Brookings

Papers on Economic Activity 1976:1, 169-203.

Medoff, J.L. and K.G. Abraham, "Are Those Paid More Really More Productive?
The Case of Experience," Journal of Human Resources 16, 2
(Spring 1981), 186-216.

Mincer, J., "Family Investments in Human Capital: Earnings of Women,"

Journal of Political Economy (March/April 1974) 82, 2 (Part 2), S76—5108.

Nelson, R., "Perspectives on the Productivity Slowdown," Journal of Economic

Literature, September 1981.

Nevins, Alan, FORD, THE TIMES, THE MAN, THE COMPANY (New York: Charles
Scribner Sons) 1984.

Okun, A., "Prices and Quantities: A Macro-economic Analysis," Brookings

Institution, 1982.

Phelps, E.S., "The Statistical Theory of Racism and Sexism," American

Economic Review, 62, 4, September 1972, 659-661
 —49-

poterba, 3. and 1. Summers, "Adjusting the Gross Flows Data: Implications

1984.
for Labor Market Dynamics," NBER Working Paper

Rock, Milton (ed.), Handbook of Wage and Salary,. AdminiStralLon, 1984.

Unemp3oyment as a Worker
Shapiro, C. and 3. StiglitZ, "Involuntary
Discipline Device," American Economic Reviw, June 1984.
OF SOCIO-ECONOMIC SUCCESS WITHIN
Taubman, P., KINOMETRICS: DETERMINANTS
AND BETWEEN FAMILIES (North Holland), 1977

Thurow, L., GENERATING INEQUALITY (New York: Basic Books) 1976

IN AMERICA, Task
u.s. Department of Health, Education and Welfare, WORKING
Force Report, 1973.

Weitzmari, M., THE SHARE ECONOMY (Boston: Harvard University Press) 1984.

American Economic Review,

Yellen, j., "Efficiency Wage Models of Unemployment,"
* vol. 74, May 1984, 200-208
</ref_section>