A Guide to Multisector Labor Market Models 
 
Gary S. Fields * 
Cornell University 
gsf2@cornell.edu 
 
Paper Prepared for the World Bank  
Labor Market Conference, 
Washington, DC 
November 18-19, 2004 
 
 
I. Introduction 
This is a paper on labor markets. Why are labor markets important to economic 
development? Many individuals and institutions, including the World Bank and the 
regional development banks, seek “a world free of poverty.” Broadly speaking, those 
who are poor are poor because 1) they earn little from the work they do, 2) the societies 
in which they live are too poor to provide them with substantial goods and services by 
virtue of their citizenship or residency, and 3) the poor are not permitted to move to 
richer countries. Thus, anti-poverty efforts can be focused on 1) helping people as 
workers (defined broadly to include wage employees, informal employees, and the self-
employed in all ranges of the skill distribution), 2) helping people as citizens/residents 
through publicly-provided goods and services, and 3) striving for freer movement of 
labor from poor to rich countries. This paper is concerned with the first channel: helping 
improve labor market opportunities for workers. 
 Labor markets deserve a prominent place in anti-poverty efforts. Research studies 
using decomposition methodologies have shown that labor income inequality is as 
important or more important than all other income sources combined in explaining total 
income inequality; see Ayub (1977) for Pakistan, Fields (1979a) for Colombia, Fei, 
Ranis, and Kuo (1978, 1979) and Fields and Mitchell (1999) for Taiwan, and Shorrocks 
                                                 
*  I am grateful to Mabel Andalón, Lisa Dragoset, Robert Duval Hernández, and Maria Laura Sánchez 
Puerta for helpful comments on an earlier draft. 
 2
(1983) and Karoly and Burtless (1995) for the United States. The reason that labor 
income is so important is that most poor people in the developing world derive no 
significant income from sources other than their own labor: in the words of the 1990 
World Development Report, “the poor’s most abundant asset [is their] labor.” (World 
Bank, 1990, p. 3). What drives income inequality, therefore, is that some people earn 
very large amounts for their labor while a great many earn very little. Thus, it is therefore 
the inequality of labor incomes that accounts primarily for the inequality of total incomes. 
Labor income also plays a predominant role in income mobility research. In much 
of this literature, economic welfare is gauged by household income per capita (PCI) or 
household consumption per capita (PCC). Research on changing PCI in Indonesia, South 
Africa, Spain, and Venezuela has shown that household per-capita income changes are 
determined much more by changes in household income (the numerator) than by changes 
in number of household members (the denominator) and that changes in labor income far 
outweigh changes in other sorts of income (Fields et al., 2003). 
 This paper approaches labor markets through multisector modeling. Multisector 
labor market models start with the recognition that in many countries, the labor market 
consists of quite distinct segments that are linked with one another. Writing as one who 
has worked with multisector labor market models for many years (e.g., Fields 1972, 
1975, 1979b, 1989, 1997, forthcoming), I owe much of my intellectual heritage to the 
early dual labor market adherents and modelers (Roy, 1951; Lewis, 1954; Fei and Ranis, 
1964; Harris and Todaro, 1970; Doeringer and Piore, 1971; Stiglitz, 1971, 1982) as well 
as to more recent analysts (Bourguignon, 1990; Banerjee and Newman, 1993; Basu, 
1997; Stiglitz, 1999). 
 Multisector models explain phenomena such as these: why an increase in 
productivity might cause wages to fall; why an economic boom in one economic sector 
might produce rising wages in all sectors; why firms might be able to add workers at will 
without having to raise wages; why urban employment creation may result in more urban 
 3
unemployment; and why the solution to urban unemployment might be rural 
development. These phenomena simply do not make sense in a single market setting. 
To develop a framework/typology and lay out the main issues on multisector 
labor market modeling, the paper proceeds in four stages. The first main substantive 
section (Section II) presents the essence of labor market dualism. I maintain that labor 
markets often consist of quite distinct segments and that a useful and insightful analytical 
approach is to start with just two.  
The second main substantive section is on models of wages and employment in 
the formal economy. To be reviewed here are 1) the market-clearing labor market model 
and the presumed equilibrating forces, 2) above-market-clearing wages set institutionally, 
3) above-market-clearing wages set by efficiency wage considerations, and 4) above-
market-clearing wages set by worker behavior. 
 The third main substantive section is on wages and employment in the informal 
economy. This section presents three characterizations of informal sector labor markets: 
1) the informal economy as a free-entry sector that prospective workers enter only as a 
last resort, 2) the informal economy as a desirable sector that workers choose in 
preference to formal sector work, and 3) the informal economy with its own internal 
dualism, combining 1) and 2). 
 The fourth main substantive section is on intersectoral linkages. The models here 
are: 1) the integrated labor market model with full market clearing, 2) crowding models, 
and 3) the Harris-Todaro model. 
While the models presented here differ from one another in important respects, 
they all share certain common features of which the reader should be aware from the 
outset. First, firms in these models are assumed to be maximizing profits. This means that 
they hire workers, raise wages, and improve worker quality if and only if it is in their 
profit-maximizing interest to do so. Second, workers in these models are assumed to be 
maximizing utility. Especially in poor countries, in which large numbers of people value 
 4
additional goods greatly compared to leisure, the utility-maximization assumption may 
often be fruitfully replaced by an income-maximization assumption. Third, the notion of 
“market equilibrium” used in this paper needs to be clarified. “Market equilibrium” is a 
state toward which a market tends and, once there, it tends to stay. “Market-clearing” is a 
state in which the quantity of a good or service supplied equals the quantity demanded. 
Some of the equilibria we shall deal with in this paper are market-clearing and others are 
not. And finally, I concentrate on positive analysis in this paper and offer very little 
normative analysis. Much of what happens in labor markets is not pretty. My objective 
here is to make labor market outcomes understandable. 
  
II. The Essence of Labor Market Dualism 
 At the core of dualistic labor market models is the distinction between a sector 
that is alternatively called “formal,” “modern,” “industrial,” or “urban” and another that 
is alternatively called “informal,” “traditional,” “agricultural,” or “rural.” (At one point, I 
even called this latter one the “murky” sector.)  Throughout this paper, I shall use the 
formal/informal terminology.  
In my view, labor market dualism is a useful stylization of what has been called 
“labor market segmentation” or “labor market fragmentation.” Why have just two 
sectors? Basu (1997, pp. 151-2) put it well:  
 
The dual economy model of LDCs has had its demurrers. It has been 
pointed out that labor markets are often fragmented into more than two 
parts and also that dualism is not the distinguishing feature of 
underdevelopment because there are traits of it even in developed 
economies. These are not disturbing criticisms. It is unlikely that any of 
the initiators of the dual economy model would deny that the labor market 
may in reality be fragmented into more than two sectors. The assumption 
of duality is merely for analytical convenience. If fragmentation – 
irrespective of the number of parts – in itself causes some problems and 
we wish to examine these, then the simplest assumption to make is that of 
dualism. 
 5
Along similar lines, Dixit (1973) wrote earlier:  
 
The dual economy has, over the last decade, proved itself to be a useful 
conceptual framework for analyzing several problems of economic 
development. . . Dual economy models provide a significantly better 
description and understanding of the problems of development than any 
aggregate model, not because two sectors are better than one . . .  but 
because the sectoral division chosen reflects several vital social and 
economic distinctions in the type of economy being analyzed. 
Unfortunately, international practice has been quite ambiguous about the feature 
distinguishing the two sectors. The International Labour Organisation and the Economic 
Commission for Latin America and the Caribbean have defined the informal sector as the 
sum of non-professional self-employed, domestic workers, unpaid workers, and workers 
in enterprises employing five or fewer workers. In Brazil, the formal sector consists of 
workers who hold labor cards entitling them to various benefits and protections and the 
informal sector of those who do not.  In other contexts, the formal sector is distinguished 
according to whether the firm is registered with the government and pays taxes. Yet 
others equate the informal economy with drugs, prostitution, and other illegal activities. 
For alternative definitions and operationalizations, see ILO (2002) and Jhabvala, 
Sudarshan, and Unni (2003). 
The distinguishing feature that I find most appealing is that associated with the 
Nobel laureates Arthur Lewis (1954) and Simon Kuznets (1955) as well as other dual 
economy modelers. Lewis, in typical fashion, presented the difference between formal 
and informal employment in picturesque terms (p. 147):  
 
What we have is not one island of expanding capitalist employment, 
surrounded by a vast sea of subsistence workers, but rather a number of 
such tiny islands . . . We find a few industries highly capitalized, such as 
mining or electric power, side by side with the most primitive techniques; 
a few high class shops, surrounded by masses of old style traders; a few 
highly capitalized plantations, surrounded by a sea of peasants. But we 
find the same contrasts also outside their economic life . . . There is the 
same contrast even between people; between the few highly westernized, 
trousered, natives, educated in western universities, speaking western 
 6
languages, and glorying Beethoven, Mill, Marx, or Einstein, and the great 
mass of their countrymen who live in quite other worlds. 
For Lewis, the essence of dualism is the fact that workers earn different wages 
depending on the sector of the economy in which they are able to find work. Lewis wrote 
(p. 150): “Earnings in the subsistence sector set a floor to wages in the capitalist sector, 
but in practice wages have to be higher than this, and there is usually a gap of 30 per cent 
or more between capitalist wages and subsistence earnings.” Lewis explained that 
although part of the gap is “illusory” because of the higher cost of living in the capitalist 
sector, there remained a real wage gap due to a) the “psychological cost of transferring 
from the easy going way of life of the subsistence sector to the more regimented and 
urbanized environment of the capitalist sector,” b) the payoff to experience in the 
capitalist sector, and c) “workers in the capitalist sector acquiring tastes and a social 
prestige which have conventionally to be recognized by higher real wages.” 
Kuznets (1955) further developed the model of wage dualism and intersectoral 
shifts by exploring how various measures of income inequality would change as the high-
income sector comes to employ an increasing share of the population. All of the 
inequality measures used by Kuznets exhibited an inverted-U pattern, which later came to 
be known as the “Kuznets Curve.” Subsequent research examined inequality (Knight, 
1976; Robinson, 1976; Fields, 1979a; Anand and Kanbur, 1993) and poverty (Fields, 
1979a ; Anand and Kanbur, 1985) in the Lewis-Kuznets process of intersectoral shifts. 
 More recent writings on labor market dualism are grounded in human capital 
theory as developed by Schultz (1961, 1962), Becker (1962, 1964), and Mincer (1962, 
1974). The more modern labor market dualism literature stressed that for dualism to 
exist, different wages must be paid in different sectors to comparable workers. Many 
researchers have reported evidence of such dualism or segmentation; for one early 
compilation of evidence, see Fields (1980a). Lewis and Kuznets should not be faulted for 
 7
neglecting the human capital issue, because human capital theory had not yet been 
devised when they developed their dualistic development models. 
 The idea that different wages are paid to comparable workers has been 
incorporated, largely without question, into job search theory, which also did not exist in 
Lewis’s and Kuznets’s time. Since the late 1960’s and early 1970’s, a whole class of 
models has arisen in which a wide variety of wages exist in the labor market, and workers 
are presumed to search among employers for the best possible opportunities. See, for 
instance, the textbook treatments of job search in Ehrenberg and Smith (2003) and Cahuc 
and Zylberberg (2004) and also the work on equilibrium wage distributions by Stiglitz 
(1985) and Burdett and Mortensen (1998).  
 Dualistic labor market models have been criticized on a number of grounds. A 
particularly harsh critic is Rosenzweig (1988). Rosenzweig accepted that empirical 
studies often show that workers with given measured human capital characteristics have 
systematically different wages or earnings depending on the type of employment in 
which they are working. He questioned not the fact but the interpretation. In his words (p. 
756): “Do [these differentials] suggest barriers to mobility – non-competing groups – or 
do they merely reflect compensatory differentials, rewards for unmeasured skills or 
compensation for unmeasured differences in the disutility of the workplace?” 
Rosenzweig favored the latter set of explanations, and therefore found the dualistic labor 
market literature unconvincing. He wrote (p. 757): “It is the lack of a precise behavioral 
interpretation of the results that is the principal shortcoming of the dualistic labor market 
empirical studies.”  
 I would offer the following behavioral interpretation which Rosenzweig says is 
lacking. Consistent with job search theory described above, suppose that there is luck in 
the labor market, pure and simple. Both Rosenzweig and I were hired at Yale as junior 
faculty members within a year of one another. I truly believe that if we had not been hired 
at Yale, we would have had a less favorable professional environment and less time for 
 8
research, and so would not have left Yale for tenured positions at Cornell (Fields) and 
Minnesota (Rosenzweig). We therefore would have experienced much less career success 
than we did. The fact that there were job vacancies at Yale when we entered the labor 
market produced a better outcome throughout our careers than we ourselves with our own 
unmeasured skills would have received had those vacancies not existed. Putting it 
differently, I think labor markets are better characterized as being segmented in the sense 
of cumulative advantage and low-level traps (Nelson, 1966; Merton, 1968; Doeringer and 
Piore, 1971; Boudon, 1973; Meade, 1976) than as being unified in the sense that the next-
best employer is essentially indistinguishable from the current one. 
 Dualistic labor market models, and segmented labor market models more 
generally, have been criticized on other grounds as well. A recent Inter-American 
Development Bank report (IADB, 2003) put it thus: “According to [the dualistic view of 
the labor market], the formal and informal economies operated in segmented labor 
markets and there is limited mobility between the two. Nothing could be further from the 
truth . . . In a given six-month period, about 16 percent of workers in Mexico and 11 
percent of workers in Argentina move either in or out of an informal job.” Nonetheless, 
the fact is that most workers remain in the sector in which they began. Economic 
mobility is a very important phenomenon, and indeed I am devoting the bulk of my 
current research precisely to this issue, but I do so within the context of dualistic labor 
markets. 
 To conclude this section, in most settings, I find that it is far more useful to think 
of developing countries’ labor markets as being fragmented or segmented than to think of 
all workers and firms in a country participating in one single labor market. When 
possible, Occam’s Razor suggests limiting the analysis to two sectors. In my view, labor 
market dualism was a most useful starting point for analyzing some economies when it 
was first presented decades ago, and it remains a useful characterization of some 
 9
economies today. But when two sectors are simply not enough, three-sector or n-sector 
models can prove insightful, a point to be developed further in the sequel.  
We turn now to the formal sector labor market, the informal sector labor market, 
and the interactions between them. 
 
III. The Formal Sector Labor Market 
 In this section, I present four alternative models of wages and employment in the 
formal economy: the market-clearing labor market model, models with wages set above 
market-clearing levels for institutional reasons, models with wages set above market-
clearing levels for efficiency wage reasons, and models with wages set above market-
clearing levels because of supply-side considerations. 
 
 A.  The Market-Clearing Labor Market Model 
 The market-clearing labor market model is so well-known that it can be presented 
quite concisely. Figure 1 displays the three essential features. First, the amount of labor 
demanded is taken as a decreasing function of the wage, other things equal. The market 
labor demand curve slopes downward because of diminishing marginal revenue product 
of labor and the associated substitution and scale effects of a wage change. Second, the 
amount of labor supplied is taken as an increasing function of the wage, other things 
equal.  The market labor supply curve slopes upward because a higher wage induces 
workers to enter this labor market from other labor markets and induces non-workers to 
enter the labor force. And third, the wage is set by supply and demand in order to clear 
the market. 
 According to the market-clearing model, three equilibrating forces operate: 
behavior of firms, behavior of workers, and behavior of wages. In the model, firms are 
free to hire workers or not depending on what is in their profit-maximizing interest to do. 
If market conditions change, what is in their profit-maximizing interest to do will change 
 10
accordingly, and firms are free to act on these changes. Similarly, workers are free to 
supply their labor in any given labor market or not depending on what is in their utility-
maximizing interest to do. For them too, if market conditions change, what is in their 
utility-maximizing interest to do will change accordingly, and they (workers) are free to 
act on these changes. And finally, if supply and/or demand conditions change, real wages 
are free to rise or fall accordingly. (In this paper, wages should always be thought of in 
real terms.) 
 As elementary and familiar as this model is, I am often surprised by analysts’ 
failure to use it. Note well what this model says. The wage is determined by what the last 
employer is willing to pay in order to attract and employ a worker and by what the last 
worker requires in order to be attracted and employed. One common misperception is that 
the wage “should” vary with labor productivity, commonly measured as value added per 
worker. Nothing could be further from the truth. Suppose that half the formal sector 
workers are replaced by machines, keeping total output constant. “Productivity” 
approximately doubles (“approximately,” because it depends on the cost of the additional 
machines compared to the cost of the dismissed workers), but in the basic labor market 
models wages will not double. In fact, to the extent that wages change at all, they are 
likely to fall, because the total demand for labor is less after the productivity 
improvement than it was before. 
 Let us now consider three groups of models with different causal structures and 
different market outcomes. 
 
 B. Above-Market-Clearing Wages Set Institutionally 
An important class of models in the labor market literature holds that wages in the 
formal sector are set by a set of forces different from supply and demand. In the models I 
review in this section, the defining feature is heavy reliance on “institutional” forces. 
 11
(“Institutional” forces are those other than the profit-maximizing behavior of firms and 
the utility-maximizing behavior of workers.) 
I have found it useful to distinguish five institutional features that may be 
important in different settings (e.g., Fields and Wan, 1989; Fields, 1999). They are: 
minimum wages; trade unions; public sector pay policies; multinational corporations; and 
labor codes. More specifically: 
 Minimum wages aim to assure workers an "adequate" standard of living.  
Minimum wages have long been on the books throughout most of the developing world 
(Starr, 1981; World Bank, 1995; Inter-American Development Bank, 2003), but their 
effects differ. In some countries, such as Bangladesh, Côte d’Ivoire, and India, minimum 
wages are binding and enforced (World Bank, 1995, p. 75).  But in others, minimum 
wage laws make little difference. South Korea introduced a minimum wage system only 
in 1988, and the minimum wage levels have been set so that they prove to be a binding 
constraint for only about 2% of Korean workers (Park, 1991; Lee, 2002).  In the case of 
Taiwan, although a minimum wage law has been in force for decades, no company has 
ever been fined for violating the law (Chang, 1989). 
 Trade unions are often encouraged by government policy as a means of entitling 
workers to a "just" share of the fruits of their labor.  One early theory of trade union 
behavior is that unions have a variety of objectives, including both higher wages and 
greater union membership (Dunlop, 1944). More recent approaches have stressed that 
those already employed in unions (the “insiders”) may be more concerned about raising 
their wages than about increasing membership (Blanchard and Summers, 1986; Lindbeck 
and Snower, 1988). Indeed, unions have raised the wages of their members by as much as 
150% in Jamaica, 31% in Ghana, 24% in South Africa, and 20% in Malaysia (Tidrick, 
1975; World Bank, 1995, Table 12.2; Aidt and Tzannatos, 2002). By contrast, in South 
Korea and Taiwan, unions have been repressed and the union wage premium is at most 
two to three percent. (Lin, 1989; Park, 1991; Yoo, 1995). 
 12
 Public sector pay policies often result in substantially higher wages being paid to 
government workers than to their private sector counterparts.  Costa Rica is an example 
of this, and as a result, "everybody" tries to work for the government (Gindling, 1991).  
In East Asia, the public sector pays what it has to in order to compete with the private 
sector -- neither more nor less. 
 Multinational corporations frequently pay above-market wages in sub-Saharan 
Africa and elsewhere (Berg, 1969; Squire, 1981).  Wage levels and working conditions 
tend to be higher in export-oriented firms than in firms producing for the domestic market 
(ILO, 1998; Moran, 2002; Ghose, 2003). Although export-oriented multinationals offer 
higher wages and better working conditions partly for efficiency wage reasons (see 
subsection C below), they also do this because some governments have "encouraged" 
them to do so by not so subtly threatening expulsion or expropriation if they do not 
(Eaton and Gersovitz, 1984).   
Finally, labor codes in some countries regulate hiring and firing, impose payroll 
taxes on firms, and mandate that employers provide certain benefits to their workers.  
Panama had such a labor code, and it was estimated to have raised labor costs by 90% 
(Spinanger, 1985) before it was finally abandoned as unsustainable.  Likewise, Bolivian 
employment legislation raises labor costs by an estimated 90-110% (Bravo, 1995).  
Larger firms in India and Zimbabwe may not dismiss workers; employment levels have 
been found to be artificially high, and economic efficiency artificially low, as a result 
(Fallon and Lucas, 1991, 1993).   
Higher-than-market-clearing wages for institutional reasons in the formal sector 
are at the core of many economic models. They include the Keynesian macroeconomic 
model, Lewis’s classical development model, Harris and Todaro’s dualistic labor market 
model, and many others. Virtually without exception, economic models regard formal 
sector employment as being determined in a very neoclassical way: given the wage and 
the capital stock, employment is set according to the marginal revenue product of labor.  
 13
 Ample research has shown that labor demand elasticities are significantly 
negative; Hamermesh (1993) provides a comprehensive review of the empirical 
literature.  In South Africa, for example, various researchers have produced estimates of 
the wage elasticity of employment in that country’s formal sector (Bowles and Heintz, 
1996; Fields, Leibbrandt, and Wakeford, 2000). Most estimates range from -0.5 to -0.7. 
While these studies differ in terms of their precise estimates, what they agree on is that 1) 
the wage elasticity of employment is significantly greater than zero and 2) the wage 
elasticity of employment is significantly less than one.  
Given labor demand elasticities of such magnitudes, higher-than-market-clearing 
wages therefore would be expected to reduce formal sector employment in equilibrium. 
Unemployment will result in the economy unless all of the workers not employed in the 
formal sector take up employment in the informal sector. Whether they do or not is the 
subject of Section IV.  
Given these research findings, the five labor market interventions reviewed above 
need to be considered carefully. Their aims are laudatory - to raise earnings and reduce 
poverty - and they do indeed benefit the workers who are fortunate enough to work in 
covered sectors of the economy.  However, they appear to have had adverse employment 
and efficiency effects and to have contributed to the informalization of the economy, as 
employers evade the regulations by not engaging workers as regular employees or by not 
even appearing as official companies  (DeSoto, 1989; Turnham, 1993; Maloney, 2003).  
Helping formal sector workers may or may not be the best tool for fighting poverty in any 
given context. 
 
C.  Above-Market-Clearing Wages Set by Efficiency Wage Considerations 
An old and well-established idea that commands nearly universal agreement not 
only in economics but in human resource management is that a firm can raise its labor 
productivity by paying a higher wage. Credit is usually given to Leibenstein (1957) for 
 14
originating this idea in the economics literature. See also Stiglitz (1974, 1976), Bliss and 
Stern (1978), Akerlof and Yellen (1986), Dasgupta and Ray (1986), and Weiss (1990) for 
further developments. But it goes back much further than that to Henry Ford, who 
pioneered the radical practice a century ago of offering his workers $5 a day, which was 
twice the going wage at that time (Raff and Summers, 1987).  
Efficiency wage theory incorporates the proposition that higher wages can result 
in higher productivity but goes beyond it in a fundamentally important way. According to 
the core microeconomic model of firms, firms are trying to achieve higher profits, which 
may or may not be enhanced by higher productivity.  Thus, the basic postulate of 
efficiency wage theory is that profit-maximizing firms will pay higher-than-market-
clearing wages if and only if the gains in productivity from doing so outweigh the costs, 
so that profits are increased. In other words, it is not enough simply to maintain that 
paying a higher wage generates benefits. It must be that the benefits exceed the costs. 
Much that is written about “high road” labor relations practices ignores this fundamental 
truth; see, for example, Ulrich (1997) and Noe et al. (2000). 
Efficiency wage theory has also contributed usefully to analyzing the mechanisms 
by which productivity gains are realized. These fall into two major categories.  
One set of explanations is that higher wages enable firms to hire better-quality 
workers from a heterogeneous labor pool. They may, for example, hire workers who have 
more education and who for this reason are expected to be more productive. 
Alternatively, they may administer tests of potential job performance and hire those 
workers who perform the best on these tests. 
The other set of explanations is that higher wages induce workers of a given skill 
level to perform in a more productive manner. The mechanisms analyzed here include 
better nutrition, improved morale, reduced shirking, lower labor turnover, reduced 
absenteeism, and greater discretionary effort. 
 15
Where the efficiency wage models come out, then, is that wages remain above the 
market-clearing level because firms in the labor market find it in their profit-maximizing 
interest to keep wages above the market-clearing level. Put differently, a firm that is 
paying efficiency wages would hurt its profits if it lowered wages.  
As in the models reviewed in the last subsection, when wages are higher-than-
market-clearing for efficiency wage reasons, we also have unemployment as an 
equilibrium outcome. However, the unemployment that arises here occurs for a very 
different reason from the institutional wage case. In the efficiency wage models, it is 
firms that do not want to reduce wages, even though at least some of the unemployed 
would be willing to work for lower wages rather than remain jobless. This contrasts with 
the institutional wage case, in which it is employed workers who want the wage to remain 
where it is. 
 
D. Above-Market-Clearing Wages Set on the Supply Side 
Another explanation for wages remaining above the market-clearing level has 
been suggested and modeled by Bardhan and Rudra (1981), Drèze and Mukherjee (1989), 
Solow (1990), and Osmani (1991). Suppose that wages in a labor market are above the 
market-clearing level for some reason – for example, because the wage was set in the 
peak season and the economy is now in the slack season. According to the standard 
account of equilibrating forces in labor markets, unemployed workers would offer to 
work for lower wages rather than remain unemployed. However, if the labor market is a 
casual one in which hiring takes place afresh each day, the unemployed may behave 
differently. In the case where the demand for labor is inelastic, each of the unemployed 
knows that he or she will earn more on average over the course of many days if s/he does 
not undercut the established wage. Here, unlike the models in the earlier subsections, 
wages are kept above the market-clearing level by the behavior of the unemployed. In this 
 16
way, the wage remains above the market-clearing level, and unemployment persists as a 
result. 
 
E. Conclusions on Modeling the Formal Sector Labor Market 
In this section, I have discussed four models of formal sector labor markets. The 
one with which we are most familiar is the full market-clearing model with adjustments 
in labor demand, labor supply, and wage-setting free to take place. The market-clearing 
model makes many institutional assumptions, most importantly, that all of the standard 
equilibrating forces are free to operate. 
Three alternative models were then presented in which wages do not adjust. In the 
first of these models, wage rigidity arises from a variety of institutional forces that may 
include minimum wages, trade unions, government pay policy, pay policies of 
multinationals, and labor codes.  Firms would like to pay lower wages, and presumably 
would employ more labor if they could lower wages, but they cannot legally do so. 
Employment is lower in the presence of wage rigidity than it would be in its absence. 
 The second alternative model is one of efficiency wages. The essence of this 
model is that when a firm pays higher wages, it will either attract a better caliber of 
worker and/or it will induce existing workers to work more productively. In this model, 
some firms choose to pay higher wages, because the benefits to profits of doing so 
outweigh the costs. When efficiency wages are paid, the workers not in the high wage 
sector would like to be, but firms would not find it in their profit-maximizing interest to 
lower the wage in order to employ them. In the equilibrium that results, those not 
formally employed end up openly unemployed (i.e., with no work at all) and/or 
underemployed (i.e., working, but in jobs that pay lower wages than they themselves 
could earn were they formally employed). 
 Finally, there is a model in which wages are free to fall but workers do not bid 
wages down. This is because workers figure that they will come out ahead over the long 
 17
run if, when the demand for labor is inelastic, wages are kept high. Even when zero 
wages during periods of unemployment are factored in, the average wage over a longer 
period is higher if wages are kept high and not bid down. 
 We turn now to models of the informal sector labor market. 
 
IV. The Informal Sector Labor Market  
The crucial feature of labor market dualism described above is that the formal 
sector offers relatively attractive terms and conditions of employment while the informal 
sector offers relatively unattractive ones. This leads to the first characterization of the 
informal economy: workers prefer formal sector jobs and enter the informal sector only 
as a last resort. More recently, though, a different view has been put forth: that the 
informal economy is a desirable sector that workers choose in preference to formal sector 
work. A third view is that the informal economy has its own internal dualism, combining 
these two characterizations. A current resource on the informal economy is the ILO’s 
Informal Economy Resource Database, available at 
http://www.ilo.org/dyn/dwresources/iebrowse.home?p_lang=en . 
 
 A. The Informal Economy as a Free-Entry Sector of Last Resort 
Ample empirical research has shown that labor earnings in the informal sector are 
low, lower even than in the formal sector in a large number of countries. For example, 
Sudarshan and Unni (2003) see informal work as “a survival activity of the very poor,” 
noting that the dimensions of informal activity are large: 35-85 percent of non-
agricultural employment in Asia, 40-97 percent in Africa, and 30-75 percent in the Latin 
America-Caribbean region. Fields (1980a) offers a review of some of the early evidence 
on wages in different employment sectors, noting with Turnham (1971) and Squire 
(1981) that open unemployment is the tip of the proverbial iceberg: the greater part of the 
employment problem in developing countries consists of workers who earn so little when 
 18
they work that they and their families are poor. These are the “working poor,” and they 
are found disproportionately in the informal sector. 
In the cities of developing countries, we see large numbers of people 
engaged in work that earns them some cash each day or week. These include 
hoards of shoe shiners clustered in the town square, lottery ticket vendors 
seemingly every few feet, would-be construction workers clustered at a particular 
street corner awaiting the daily round-up, newspaper sellers approaching stopped 
cars at virtually every traffic light, and (sadly) groups of women, and sometimes 
men and children, gathered in the red light district. Lewis (1954, p. 141) referred 
to “the whole range of casual jobs – the workers on the docks, the young men 
who rush forward asking to carry your bag as you appear, the jobbing gardener, 
and the like. These occupations usually have a multiple of the number they need, 
each of them earning very small sums from occasional employment; frequently 
their number could be halved without reducing output in this sector.” (Emphasis 
added) 
Subsequent investigations into these people’s lives as well as casual empiricism 
led analysts to view these types of jobs as having free entry. In a pathbreaking ILO report 
on Kenya (1972, p. 6), the criteria defining the informal sector were:  
i) ease of entry; 
ii) reliance on indigenous resources; 
iii) family ownership of enterprises;  
iv) small scale of operation;  
v) labour-intensive and adapted technology; 
vi) skills acquired outside the formal school system; and 
vii) unregulated and competitive markets. 
The essence of free entry is that all who want a job can get one. (“Job” here is 
defined to include both self-employment and wage employment.) Barriers to entry into 
 19
such occupations are small or non-existent. In some contexts, primarily urban, all that 
would-be workers need to do is make a minimal investment in the product or service to 
be sold. In rural contexts, it is obligatory for the family or community to take back into 
the home those who find such work the best of a bad set of alternatives. One is reminded 
of Robert Frost’s immortal words in his poem “Death of the Hired Man”:  “Home is the 
place where, when you have to go there, they have to take you in.”  
The existence of free-entry employment opportunities in the informal sector helps 
explain why open unemployment rates in developing countries are comparable to those in 
developed countries, and often considerably lower (Turnham, 1971, 1993; World Bank, 
1995; ILO, 2003). The standard ILO definition of unemployment is a person who did no 
work for pay in the preceding week, not even for one hour. In poor countries lacking 
systems of unemployment insurance and cash assistance allowances, the great majority of 
poor people cannot afford to be without income for as long as a week. So to the extent 
that the poor can quickly find an opportunity to earn some cash in an informal job, they 
take it. Open unemployment in their economies is low as a result.  
Because of easy entry into economic activities of such kinds, a different wage 
determination process from the standard marginal productivity rule must be found. 
Income-sharing is a prominent feature of Lewis’s model and many subsequent ones (e.g., 
Fei and Ranis, 1964; Harberger, 1971; Fields, 1975). As modeled by Lewis, in the 
informal sector, the marginal product of labor is zero or low – in any event, below the 
average product of labor.  In his words (p. 142):  
 
Most businesses in underdeveloped countries employ a large number of 
‘messengers’, whose contribution is almost negligible; you see them 
sitting outside office doors, or hanging around in the courtyard. And even 
in the severest slump the agricultural or commercial employer is expected 
to keep his labour force somehow or other – it would be immoral to turn 
them out, for how would they eat, in countries where the only form of 
unemployment assistance is the charity of relatives? So it comes about that 
even in the sectors where people are working for wages, and above all the 
domestic sector, marginal productivity may be negligible or even zero.  
 20
One of the major criticisms of the Lewis model had to do with these features of 
labor surplus and income-sharing. Lewis (1972) later wrote, “Whether marginal 
productivity is zero or negligible is not at the core of fundamental importance to our 
analysis . . . This has led to an irrelevant and intemperate controversy.” Ranis 
(forthcoming) now labels as “unfortunate” the choice of the “labor surplus” term and 
dismisses this critique as a “red herring.” What matters, writes Ranis, is “that the 
marginal product is low, and sufficiently low to fall below the bargaining wage or income 
share.” 
 How, then, would we want to model informal sector wage determination today? 
Essentially, there are four tacks that might be taken, the first two for analytical simplicity 
and the second two for greater comprehensiveness.  
One is to assume that there is a fixed amount of income to be earned in the 
informal sector regardless of the number of people working in that sector - that is, the 
marginal product of labor is literally zero.  For example, there may be a fixed number of 
newspapers to be sold regardless of the number of newspaper vendors. How is the fixed 
income from newspaper vending to be divided? The easiest simplifying assumption here 
is full income-sharing among the informally employed, so that each earns the average 
product. The average product is not constant, though – it varies inversely with the number 
of people in the informal sector. This was the way that I modeled the urban informal 
wage in Fields (1975, 1989). 
 A second approach is to regard a part of the informal sector as facing, instead of 
zero marginal product, constant marginal product. The dual economy model developed 
by Harris and Todaro (1970) was formulated to fit the East African case, which they and 
others regarded as a land surplus economy at the time. Harris and Todaro assumed that 
anyone who wanted to work in agriculture could find a plot of land, cultivate it, and earn 
the marginal product from his or her efforts. Agricultural wages were equated to marginal 
product, not average product as in Lewis. If we can assume that the marginal worker and 
 21
the marginal land are as productive as preceding ones, a convenient simplifying 
assumption would be to regard the marginal product of labor in agriculture as constant. 
This was the way that the agricultural was first modeled in Fields (1975). Many others 
subsequently adopted this assumption in what has come to be called the simplified 
Harris-Todaro model (Fields, 1975; Anand and Joshi, 1979; Heady, 1981; Stiglitz, 1982; 
Sah and Stiglitz, 1985; Bell, 1991). 
 A third approach is intermediate between the first two: a positive but diminishing 
marginal product. Harberger (1971) put it thus: 
 
[This] variant associates disguised unemployment not just with low wages 
but with situations in which the marginal productivity of labour lies below 
the actual wages earned. . . There are a variety of activities to which this 
argument applies. A classic example is that of fishermen on a lake. The 
addition of more fishermen increases the total catch, but not 
proportionately, yet the last fisherman has an equal chance of making a 
given catch as the first. The expected catch is the same for all, and is equal 
to their average productivity. But, owing to the fact that the total catch 
does not increase in proportion to the number of fishermen, the marginal 
productivity of a fisherman is less than what he earns. 
Models with positive but variable marginal product are harder to work with than either of 
the two preceding ones. 
A fourth approach is to model a full demand system for agricultural and non-
agricultural products and workers. This was done by Bourguignon (1990). The equations 
of such systems are so complicated that they are best left to microsimulation and 
computable general equilibrium exercises. 
To conclude, the most common characterization of the informal sector is that it is 
an easy-entry sector that workers can enter to earn some cash in preference to earning 
nothing. An alternative view has been gaining popularity in recent years. Let us now turn 
to it. 
 
 22
B. The Informal Economy as a Desirable Sector 
A very different view of the informal sector also appears in the literature. It is the 
idea that a large number of those working in the informal sector are there voluntarily. 
This view has a long history in the literature (e.g., Hart, 1973; Balán, Browning, and 
Jelin, 1973). In fact, in 1990, I wrote this (Fields, 1990, p. 66):  
 
Many people are in informal activities by choice. When asked their 
reasons for doing what they were doing, many informal workers in Costa 
Rica gave the following answers most frequently: i) They feel they could 
make more money at the informal sector job they were doing than they 
could earn in the formal sector, or ii) Even though they made a little less 
money, they enjoyed their work more, because it allowed them to choose 
their own hours, to work in the open air, to talk to friends, etc.  
The choice approach to the informal sector has been developed more recently in a 
series of papers by William Maloney. A comprehensive summary of these arguments 
appears in Maloney (2003).  
As all labor economists know, workers choose among jobs and sectors on the 
basis of a package of characteristics. These include wages, benefits, the work 
environment, and so on. The variable denoted W on the vertical axis of a standard labor 
market diagram is ordinarily thought of as a shorthand for this package of benefits. And it 
is this package of characteristics which Maloney maintains are “roughly comparable” 
between informal self-employment and formal employment, at least in Mexico. 
Specifically, Maloney offers a number of reasons why workers might want to be in the 
informal sector: some can earn more (or at least hope to earn more) in informal self-
employment than they could earn in formal sector employment; they value the 
independence of self-employment; they would rather use the money that protections cost 
them for investing in their own small enterprises; they do not value protections such as 
health insurance which formal employment offers to them, in some cases because they 
already have these protections; and they don’t trust the government to deliver on 
promises such as future pension benefits. For any or all of these reasons, there may be a 
 23
sizeable numbers of workers who prefer informal self-employment to formal wage 
employment. 
One reason that self-employment is often seen as undesirable is that 
microenterprises exhibit very high rates of failure. Maloney responds to the 
precariousness argument thus (p. 77): 
 
Small firms will have higher costs, are likely to be informal, and will have 
very high failure rates. Though this corresponds exactly to the standard 
picture of the stagnant, precarious, unproductive, unprotected informal 
worker familiar in the literature, it is, in fact, the opposite. It emerges 
naturally from the workers trying their luck at entrepreneurship (risk-
taking), often failing, and not engaging in the formal institutions until they 
grow. In sum, there may be nothing pathological about informal self-
employment, and to recover the general sense of the word, nothing 
obviously less decent either. 
I agree with Maloney on this point, but I think he goes too far in one respect. He 
presents an integrated labor market model (pp. 68, 72) in which the total package of 
benefits is equalized between informal self-employment and formal wage employment.  
While this model might fit the choice between formal sector employment and informal 
self-employment for those who already have the option of working in the formal sector, 
this is a limited group of people. Rather, as I argued above, throughout the developing 
world, formal sector jobs appear to be far fewer in number than the number of people 
who want them. Thus, in my view, Maloney’s characterization applies to a subset of 
informal sector workers, but by no means all of them, nor probably even most. 
 In the end, Maloney sides with a view that I put forth in 1990: that the informal 
sector is itself heterogeneous. Let us now turn to that view. 
  
C. The Informal Economy with Its Own Internal Dualism 
Subsections A and B have put forward two polar views. One is that informal 
sector employment is worse than formal sector employment but superior to 
 24
unemployment. The other is that employment in the informal economy is preferred to 
formal sector employment. 
A way of combining these two polar views would be to regard the informal sector 
as having its own internal duality. On this synthesized approach, some informal activities 
are preferable to formal sector jobs and some are not. In fact, this is a view that I 
developed at length in Fields (1990), where I labeled the two segments “upper-tier” 
informal activities and “easy entry” ones. See also House (1984), Tokman (1987), 
Marcouiller et al. (1997), and Ranis and Stewart (1999).  
In fact, dualism within the informal sector is a view that Maloney has come to 
share. Summarizing the findings of Cunningham and Maloney (2001) for Mexico, 
Maloney writes (2003, p. 80): “The single distribution was rejected, supporting a two-tier 
view, but the share of the population found in the ‘lower’ tier was only 13 percent of the 
sample.” Perhaps most informal entrepreneurs are in the upper-tier in Mexico, but I doubt 
this is the case in India, Bolivia, or Kenya. 
Another way of modeling the duality of the informal sector is to specify two 
informal sectors that are geographically distinct. Todaro (1969) had three employment 
sectors – urban modern employment, urban traditional employment, and agricultural 
employment – but no unemployment. Harris and Todaro (1970) had urban modern 
employment, agricultural employment, and unemployment but no urban informal sector. 
In Fields (1975), I had three employment states – urban modern employment, an urban 
murky sector, and rural agricultural employment – plus unemployment.  
If the preceding sector distinctions are put together, we would have four 
employment states - employment in the formal sector, the upper-tier informal sector, the 
easy-entry sector, and rural agriculture – plus unemployment. Adding in rural off-farm 
employment – what is sometimes called the z-goods sector (Hymer and Resnick, 1969; 
Ranis and Stewart, 1993) – would introduce a fifth employment state. As far as I know, 
no analytical model has included all five employment states plus unemployment. I don’t 
 25
know why others haven’t built such models, but I do know why I haven’t: I find such 
models too complicated and intractable.  
Although six-state models have not been constructed, the literature offers a 
number of four-state models (consisting of three employment sectors plus 
unemployment). How the different states link to one another is open to alternative 
specifications. These are discussed in Section V.  
 
D. Conclusions on Modeling the Informal Sector Labor Market 
 Informal sector labor markets can be modeled in several alternative ways. One is 
as a free-entry sector, to which workers go when there is no other way to earn some cash. 
A second model of the informal sector is that it is a desirable sector, to which workers 
aspire. A third model combines the first two and allows for duality within the informal 
sector, which then consists both of an easy-entry component and an upper-tier 
component.   
 We turn now to links between formal and informal sector labor markets. 
 
V. Intersectoral Linkages in the Labor Market 
 This section is on intersectoral linkages. The models reviewed here are: 1) the 
integrated labor market model with full market clearing, 2) multisector models with no 
unemployment, and 3) the Harris-Todaro model, both in its original form and as 
extended. 
 
A. The Integrated Labor Market Model with Wage Equalization and No 
Unemployment 
The integrated labor market model, also called the unified labor market model, 
has as its distinguishing features that 1) each labor market clears, and 2) full intermarket 
 26
equilibrium is achieved through actual wage equalization. The model is explained with 
the aid of Figure 2. 
Suppose for ease of analysis that the economy has two sectors, here termed 
"manufacturing" and "agriculture."  All workers are identical, and so would be willing to 
work wherever the wage is higher, be it in manufacturing or in agriculture. The labor 
market is assumed to be an integrated one in the sense that the same wage prevails in 
both sectors of the economy for a given type of worker -- a realistic enough stylization is 
some settings, particular for countries in East Asia. For a model in which workers are not 
identical, but where the marginal worker earns the same regardless of whether s/he works 
in one sector or the other, see Roy (1951). 
 Let us start with a situation in which the demand for labor curve in the 
manufacturing sector, DM, is downward-sloping relative to origin OM, and likewise, the 
demand for labor curve in the agricultural sector, DA, is downward-sloping relative to 
origin OA.  The total labor supply is represented by the horizontal distance OM OA.   If 
the standard equilibrating forces in labor markets are free to operate, as is indeed the case 
in much of East Asia, wages would equalize across the two sectors at level W*.  At this 
wage, OM E workers would be demanded in the manufacturing sector, and OA E workers 
would be demanded in the agricultural sector. Furthermore, the total labor demanded in 
the two sectors combined would exactly equal the total labor supplied in the economy. In 
such an equilibrium, the marginal worker is indifferent between working the 
manufacturing sector or in the agricultural sector, because the two sectors pay the same 
wages. 
 Now suppose that economic growth takes place in the manufacturing sector.  
Because manufacturing firms need more workers to produce the extra output, the labor 
demand curve in the manufacturing sector shifts rightward to DM'.  Assuming no change 
in the agricultural product market, the agricultural employers' demand for labor curve 
would remain stationary at DA.  The labor market is now in disequilibrium, because at 
 27
the original wage W*, more labor is demanded than is supplied.  To resolve this 
disequilibrium, some manufacturing employers raise wages in order to retain existing 
workers and attract new ones, and agricultural employers raise wages to prevent their 
workers from leaving. The result is that the labor market equilibrates at a new common 
wage W' > W*.  Because of the sector-specific shift in labor demand, more of the 
country's workers are now in the manufacturing sector than before (OM E' rather than  
OM E) and fewer in agriculture (OA E' rather than OA E).   
 In the integrated labor market model, economic growth in one sector benefits 
workers in all sectors. Three groups of workers have been identified in this analysis:  (i)  
Those who had been working in manufacturing and now earn higher wages than before;  
(ii)  Those who are drawn by higher wages into manufacturing from agriculture; and (iii)  
Those who remain in agriculture and earn more than they did previously.  In this way, 
economic growth in a country's export sector reverberates throughout the labor market, 
benefiting those who produce manufactured goods and those who produce agricultural 
goods.   
 The extension of the integrated labor market model from two sectors to N sectors 
is immediate. 
 
B. Multisector Models with Wage Differentials and No Unemployment  
In contrast to the integrated labor market model just discussed, a number of 
multisector models are characterized by intersectoral wage differentials. Models with 
intersectoral wage differentials and no unemployment include the unlimited supply of 
labor model of Lewis (1954), the intersectoral shifts model of Kuznets (1955), the 
crowding model of Bergmann (1971), the minimum wage model with incomplete 
coverage of Welch (1974), and the modern sector enlargement model of Fields (1979b, 
1980b). As shown in Figure 3, crowding raises the wage of favored workers and lowers 
the wage of disfavored ones. 
 28
These models maintain labor market dualism in the sense that real wages are 
higher in the formal sector than in the informal sector. In this sense, they differ from the 
integrated labor market model described in the last subsection, in which wages are the 
same in the different sectors. The multisector models described in this section also 
maintain a particular kind of supply-side behavior: all workers not employed in the 
higher-wage formal sector are assumed to take up employment in the lower-wage 
informal sector. These models therefore exhibit no unemployment.  
Within this class of models, the most heralded version is the Nobel Prize-winning 
work of Lewis (1954). As discussed above, the novel feature of the Lewis model was that 
the modern sector faces an unlimited supply of labor at wages only somewhat higher than 
subsistence levels. It is this that makes the Lewis model “classical,” in contrast to a 
“neoclassical” model in which labor is scarce and has to be bid away from other uses. 
This feature of the classical model was later elaborated on by Ranis and Fei (1961), Fei 
and Ranis (1964), and Jorgenson (1967). 
The unlimited supply of labor to the modern sector is sometimes called an 
“infinitely elastic supply curve of labor,” but this designation is a misnomer. By 
definition, a supply curve tells the amount of a good or service that is forthcoming as a 
function of the relevant price. For it to be a proper function, there can be only one 
quantity for any price. That is, given the price of labor, the supply function delivers the 
unique quantity of labor available. Thus, in the Lewis model, when the formal sector 
wage is above the informal sector wage, the potential quantity of labor supplied to the 
formal sector is the entire labor force. However, because formal sector employers do not 
wish to employ all the workers who would like to work there at that wage, they (the 
employers) face an effectively unlimited supply of labor. Specifically, this means that no 
individual employer need raise the wage to attract additional labor, nor must employers 
as a whole within a substantial range. Indeed, there is a horizontal curve, but it is the 
 29
wage as a function of employment, not the supply of labor (defined in the standard way 
as the amount of labor supplied as a function of the wage). 
Over time, the process of savings, investment, capital formation, and economic 
growth highlighted in many growth models (both classical and neoclassical) shifts the 
marginal product of labor curve rightward. Workers respond to the increased demand for 
labor in the formal sector by taking up formal employment to the extent possible. 
Throughout a long range, the wage in the formal sector remains unchanged, because 
employers do not need to raise the wage to attract more labor. Ultimately, though, a 
turning point is reached once the supply of labor to the formal sector is no longer 
unlimited. 
Lewis’s characterization of intersectoral linkages thus generates two major 
predictions. The first is that as long as there exists a surplus of labor to the formal sector, 
economic growth would generate intersectoral shifts of employment but little or no 
increase in real wages. The second prediction is that once the unlimited supply of labor to 
the formal sector is exhausted and the turning point is reached, subsequent economic 
growth is marked by rising real wages economy-wide. 
The model proved to be remarkably prescient. Take the case of Taiwan, where 
manufactured exports were the engine of growth. Data on unemployment and real wages 
are displayed in Figure 4. At the time Lewis was writing, the open unemployment rate 
was 6.3%, higher than the generally agreed-upon level of full employment. In the next six 
years of Taiwan’s economic growth, unemployment fell to 4.3% and real wages in 
manufacturing rose by only 2% (total, not per year), consistent with excess labor 
continuing to be supplied relative to the amount demanded. But then, in the next decade 
(the 1960s), unemployment fell to 1.5% - a rate indicating severe labor shortages – and 
real wages shot up by 81%. Unemployment remained below 2% in the 1980s and 1990s, 
and real wages doubled again in each decade, not only in manufacturing but throughout 
the Taiwanese labor market.  
 30
The two phases predicted by Lewis appear clearly in the data for Taiwan: falling 
unemployment at essentially constant wages, then rapidly rising real wages at full or 
over-full employment. The dualistic model with intersectoral linkages tells a compelling 
story, and it did it before it happened. 
Despite the many insights of the original Lewis model and Fei and Ranis’s 
amplification of it, I find one feature of the model troublesome: the nature of the wage in 
the informal sector. Lewis used the term “subsistence wage.” If the wage is literally a 
subsistence wage, below which people cannot subsist, then it has a natural floor. But as 
these models have evolved, the informal sector wage does not take on the character of the 
minimum needed for survival. It is, rather, more of a basic wage, lower than the real 
wage received by formal sector workers. The question, then, is whether this wage is a 
constant low wage or whether it varies (inversely) with the number of people in the 
sector.  
The great majority of analysts regard production in the informal sector as subject 
to diminishing returns; see, for example, the Harberger quotation above. What 
diminishing returns in the informal sector implies is that when economic growth takes 
place and workers are drawn out of the informal sector into the formal sector, those who 
remain in the informal sector each receive a higher income than before; from my reading, 
this was first pointed out by Sen (1967). The informal sector wage should not remain 
constant. Indeed, the rising wage in the informal sector is a reason for the unlimited 
supply of labor to the formal sector to run out eventually: because the supply price of 
labor to the formal sector will have risen due to improved wage opportunities in the 
informal sector.  
In the dualistic labor market model with no unemployment, economic growth 
reduces poverty in two ways. One is the increase in wages and utility of those who are 
able to move from the informal to the formal sector. The other is the increase in wages of 
those who remain informal.  
 31
Thus, we see that in these models with wage dualism and no unemployment, as in 
the other multisector models, employment and wages in each sector of the economy are 
determined by labor market conditions in all sectors of the economy. Partial equilibrium 
analysis simply cannot explain what we see.  
 
C. Models with Wage Differentials and Unemployment: The Harris-Todaro 
Model and Extensions of It 
In 1970, a major alternative was developed in the context of East Africa. John 
Harris and Michael Todaro (1970) formulated a model in which, to be hired for a formal 
sector job, it was necessary to be physically present in the urban areas where the formal 
sector jobs are located. In the Harris-Todaro model, more workers search for formal 
sector jobs than are hired. Employers hire some of the searchers but not all of them. 
Those not hired end up unemployed ex post. Open unemployment, though a feature of the 
world, was not a feature of the models reviewed in the last two subsections. 
 In essence, the Harris-Todaro labor market operates as follows. Employers in the 
formal sector hire workers until the point where the marginal product of labor equals the 
wage W F . On the other hand, in the informal sector, there is assumed to be free entry; 
thus, all persons who wish to work in the informal sector may do so. Each person 
employed in the informal sector earns a wage WI < W F . 
 Workers are assumed to consider the mathematical expected wages from each of 
two search strategies: (1) Searching for a formal sector job, which pays a relatively high 
wage but runs the risk of unemployment, and (2) Taking an informal sector job, which 
offers a low wage with no risk of unemployment. Harris and Todaro’s insight was that 
workers would be expected to allocate themselves between formal sector and informal 
sector search strategies so that the expected wages from the two search strategies are 
equalized: E(WF) = E(WI). In the basic Harris-Todaro model, this equilibrium condition 
 32
E
becomes W FF =W I , where EF is employment in the formal sector and LF is the labor LF
force in the formal sector. Because W F >W I , it follows that 
EF <1,  i.e., the formal 
LF
sector labor force exceeds formal sector employment, and therefore a Harris-Todaro 
equilibrium is characterized by open unemployment. 
The Harris-Todaro model produced two powerful policy results. The first 
concerned a policy of formal sector employment creation to employ the unemployed 
(who, in the Harris-Todaro model, were all in urban areas, because that is where the 
formal sector jobs were assumed to be located). A policy of increasing formal sector 
employment by ∆EF= EF' - EF increases the formal sector labor force by 
W
∆E FF and W I
increases open unemployment by ∆E (W FF −1).  Thus, the solution to urban W I
unemployment is not urban employment creation.  
The second policy option considered was a policy of rural development. Suppose 
that such a program could increase the (rural) informal sector wage from W I  to W I '. 
From the H-T equilibrium condition, unemployment would then fall from  
UNEM = E (W FF −1) to UNEM' = E
W F
F ( −1).   Thus, in the Harris-Todaro model, the W I W I '
solution to urban unemployment is rural development.  
Soon after the model was published, the government of Kenya followed the 
Harris-Todaro precepts by putting into place an integrated rural development program. 
Indeed, unemployment in Kenya did indeed fall. For a more comprehensive welfare 
economic analysis of various policy options in the Harris-Todaro model, see Fields 
(forthcoming) 
Harris and Todaro’s fundamental contribution was to build a model with wage 
dualism and unemployment based on sound microfoundations. The fact that the model 
 33
remains part of our toolkit more than three decades later is a tribute to its basic insight 
and analytical power. 
At the same time, some of the assumptions of the Harris-Todaro model were 
judged to be too restrictive, and so the model was generalized in the years that followed 
to nest their specific formulation within a broader framework. Their model was first 
extended by Fields (1975). In that paper, I allowed for on-the-job search from rural 
agriculture, the existence of an urban informal sector, preferential hiring of the better-
educated, and employment fixity. The model has subsequently been extended and 
generalized to allow for duality within the rural sector, mobile capital, endogenous urban 
wage setting, risk-aversion, a system of demand for goods, and many other factors 
(Corden and Findlay, 1975; Calvo, 1978; Moene, 1988, 1992; Khan, 1989; Fields, 1989; 
Chakravarty and Dutta, 1990;  Bourguignon, 1990; Basu, 1997). 
To give the flavor of these extended Harris-Todaro models, consider the 
geographically distinct version of informal sector duality described in Section IV.C. 
Presumably, those people located nearer to where the formal sector jobs are stand a better 
chance of being hired for any given job vacancy. The fact that they do has implications 
for urban informal wages.  For strictly positive numbers of people to choose each of the 
three search strategies, the extended Harris-Todaro equilibrium requires that expected 
wages equalize across the three search strategies. If one group of informal sector workers 
has better on-the-job search opportunities than another, the labor market equilibrium must 
be one where the group with the better on-the-job search opportunities ends up with a 
lower wage in equilibrium. Viewed in this way, it is not surprising that some of the worst 
poverty in the developing world would be found in the urban areas: the urban poor 
consist at least in part of those who sought urban formal sector jobs but who were 
unlucky enough not to be hired for them. Of course, there is another reason for very low 
urban informal sector wages – landlessness – which the extended Harris-Todaro 
explanation complements. 
 34
 
 D. Conclusions on Modeling Intersectoral Linkages 
In this section, we have examined models of the linkages between the different 
sectors.  One model of intersectoral linkages is the multisector analog of the market-
clearing model. In this integrated labor market model, the two sectors are distinct but 
otherwise offer essentially identical wages and conditions of employment.  As a result, 
the marginal worker is indifferent as to which of the two sectors he or she wishes to work 
in.  
One alternative to the integrated labor market model is a model of wage dualism 
and no unemployment. In this model, all workers seek formal sector jobs. Those who are 
fortunate enough to be hired for such jobs take them.  All those who are not so fortunate 
then take up work in the informal sector. No one ends up unemployed in this model. 
The other principal alternative is a model of wage dualism with unemployment. In 
this model, it is also the case that all workers would like formal sector jobs and those who 
are offered such jobs take them. However, what differentiates this model from the 
preceding one is that some number of workers will find it advantageous to search for the 
high wage jobs and risk unemployment. Each available job will have more than one job-
seeker. Once hiring has taken place, the unlucky ones are found to be unemployed. 
 
VI. Conclusion 
 I shall now try to summarize the main points of this review and offer some 
concluding thoughts.  
First, multisector labor market models have proven very useful. When possible, 
Occam’s razor suggests limiting the analysis to two sectors. But when two sectors are 
simply not enough, three-sector or n-sector models can and have proved insightful. In 
these multisector models, employment and wages in each sector of the economy are 
 35
determined by labor market conditions in all sectors of the economy. Partial equilibrium 
analysis simply cannot explain what we see.  
Second, formal sector labor markets can be formulated in several alternative 
ways: in terms of market-clearing, in terms of institutional wage-setting, in terms of 
efficiency wages, and in terms of worker-side resistance to wage cuts.  
Third, informal sector labor markets can be modeled as a free-entry sector, as a 
desirable sector, or as having its own internal duality.  
Fourth, the linkages between the different sectors can be modeled in a number of 
ways: as an integrated labor market, as a model of wage dualism and no unemployment, 
and as a model of wage dualism with unemployment.  
What should we do with these components? Each of the three – formal sector 
labor market, informal sector labor market, and intermarket linkages – has three or four 
alternatives. Each combination is possible. Thus, even this relatively coarse 
categorization results in thirty-six different labor market models.  
No one would expect that the same model would fit East Africa and East Asia or 
South Africa and South Korea. Surely, the “correct” model is context-specific. Blending 
empirical observation and analytical modeling has yielded great advances. Yet, much 
more remains to be done. 
Coming up with the “correct” model matters for more than understanding; it 
matters for policy purposes as well. The same policy can have different effects in the 
different models. Take the policy of expanding employment in the formal sector. 
In all three models, as would be expected, labor moves from the informal sector to 
the formal sector to take up the available jobs. Those individuals who are able to make 
the move are better off in all three models. 
The models differ, however, in other respects. In the integrated labor market 
model, wages rise by equal amounts in the two sectors. All workers are better off by the 
same amount. 
 36
In the model with wage dualism and no unemployment, the wage level in the 
formal sector stays the same, but wages rise to some degree in the informal sector. Thus, 
the remaining informal sector workers benefit by what went on in the formal sector.  
Finally, in the model with wage dualism and unemployment, the informal sector 
wage may or may not rise. What will surely rise, though, is unemployment. In this model, 
unlike the other ones, formal sector employment creation produces both winners and 
losers.  
Thus, whether a policy of formal sector employment creation would be expected 
to have favorable labor market effects depends on which labor market model best fits a 
particular country’s institutional circumstances. Sound labor market policies require 
sound labor market models.  
 
 37
 
References 
 
Aidt, Toke and Zafiris Tzannatos (2002). Unions and Collective Bargaining: Economic 
Effects in a Global Environment. (Washington: World Bank). 
 
Akerlof, George and Janet Yellen, eds. (1986). Efficiency Wage Models of the Labor 
Market. (Cambridge: Cambridge University Press). 
 
Anand, Sudhir, and Vijay Joshi (1979) "Domestic Distortions, Income Distribution and 
the Theory of Optimum Subsidy," The Economic Journal, June. 
 
Anand, Sudhir and S.M.R. Kanbur (1985). “Poverty Under the Kuznets Process,” 
Economic Journal 95 (supp.): 42-50. 
 
Anand, Sudhir and S.M.R. Kanbur (1993). “The Kuznets Process and the Inequality-
Development Relationship,” Journal of Development Economics 40: 25-52. 
 
Ayub, Mahmood (1977).  Income Inequality in a Growth-Theoretic Context: The Case of 
Pakistan. Unpublished Ph.D. dissertation, Yale University.   
 
Balán, J., H.L. Browning, and E. Jelin (1973). Men in a Developing Society. (Austin, TX: 
Institute of Latin American Studies, University of Texas at Austin). 
 
Banerjee, Abhijit V. and Andrew F. Newman (1993). “Occupational Choice and the 
Process of Development,” Journal of Political Economy. 101: 274-298. 
 
Bardhan, P.K. and A. Rudra (1981). “Terms and Conditions of Labour Contracts in 
Agriculture: Results of a Survey in West Bengal 1979,” Oxford Bulletin of Economics 
and Statistics 89-111. 
 
Basu, Kaushik (1997). Analytical Development Economics. (Cambridge, MA: MIT 
Press). 
 
Becker, Gary S. (1962). “Investment in Human Capital,” Journal of Political Economy. 
LXX Supplement: 9-49. 
 
Becker, Gary S. (1964). Human Capital. (New York: Columbia University Press for the 
National Bureau of Economic Research). 
 
Bell, Clive (1991). "Regional Heterogeneity, Migration, and Shadow Prices," Journal of 
Public Economics, October. 
 
Berg, Elliott J. (1969). “Wage Structure in Less Developed Countries,” in A.D. Smith, 
ed., Wage Policy Issues in Economic Development. (London: Macmillan). 
 
 38
Bergmann, Barbara (1971). “The Effect on White Incomes of Discrimination in 
Employment,” Journal of Political Economy 79: 294-313. 
 
Blanchard, Olivier and Lawrence Summers (1986). “Hysteresis and the European 
Unemployment Problem,” NBER Macroeconomics Annual 15-78. 
 
Bliss, Christopher J. and Nicholas H. Stern (1978). “Productivity, Wages and Nutrition: 
1. The Theory; 2. Some Observations,” Journal of Development Economics 5: 363-398. 
 
Boudon, Raymond A. (1973). Mathematical Structures of Social Mobility. (Amsterdam: 
Elsevier). 
 
Bourguignon, François (1990). “Growth and Inequality in the Dual Model of 
Development: The Role of Demand Factors,” Review of Economic Studies 57: 215-228. 
 
Bowles, S. and Heintz, J. (1996) “Wages and Jobs in the South African Economy: An 
Econometric Investigation,” Department of Economics, University of Massachusetts, 
processed, January.  
 
Bravo, David (1995). “Cost of Labor Standards in Bolivia,” Harvard University, 
processed. 
 
Burdett, Kenneth and Dale Mortensen (1998). “Wage Differentials, Employer Size, and 
Unemployment,” International Economic Review 39: 257-273. 
 
Cahuc, Pierre and André Zylberberg (2004). Labor Economics. (Cambridge, MA: MIT 
Press). 
 
Calvo, Guillermo A. (1978). "Urban Unemployment and Wage Determination in LDC's: 
Trade Unions in the Harris-Todaro Model," International Economic Review, February. 
 
Chakravarty, Satya R. and Bhaskar Dutta (1990). "Migration and Welfare," European 
Journal of Political Economy. 
 
Chang, C.-H. (1989). “A Study on the Labor Market in Taiwan,” in Chung-Hua 
Institution for Economic Research, Conference on Labor and Economic Development, 
Taipei. 
 
Corden, Max and Ronald Findlay (1975). "Urban Unemployment, Intersectoral Capital 
Mobility and Development Policy," Economica, February. 
 
Cunningham, Wendy V. and William F. Maloney (2001). “Heterogeneity in the Mexican 
Micro-Enterprise Sector,” Economic Development and Cultural Change 50: 131-156. 
 
Dasgupta, Partha and Debraj Ray (1986). “Inequality as a Determinant of Malnutrition 
and Unemployment,” Economic Journal 96: 1011-1034. 
 39
 
DeSoto, Hernando (1989). The Other Path. (New York: Harper and Row). 
 
Dixit, Avinash K. (1973). “Models of Dual Economies” in James A. Mirrlees and 
Nicholas H. Stern, eds., Models of Economic Growth: Proceedings of a Conference Held 
by the International Economic Association at Jerusalem. (New York: International 
Economic Association). 
 
Doeringer, Peter B. and Michael J. Piore (1971). Internal Labor Markets and Manpower 
Analysis. (Lexington, MA: Heath). 
 
Drèze, Jean and Anandita Mukherjee (1989). “Labour Contracts in Rural India: Theories 
ande Evidence, “ in S. Chakravarty, ed., The Balance Between Industry and Agriculture 
in Economic Development. (London: Macmillan). 
 
Dunlop, John T. (1944). Wage Determination Under Trade Unions (New York: 
Macmillan). 
 
Eaton, Jonathan and Mark Gersovitz (1984). “A Theory of Expropriation and Deviations 
from Perfect Capital Mobility,” Economic Journal 94: 16-40. 
 
Ehrenberg, Ronald G. and Robert S. Smith (2003). Modern Labor Economics. (Boston: 
Addison Wesley). 
 
Fallon, Peter and Robert Lucas (1991). “The Impact of Changes in Job Security 
Regulations in India and Zimbabwe,” World Bank Economic Review. September. 
 
Fallon, Peter and Robert Lucas (1993). “Job Security Regulations and the Dynamic 
Demand for Industrial Labor in India and Zimbabwe,” Journal of Development 
Economics.  
 
Fei, John C.H. and Gustav Ranis (1964). Development of the Labor Surplus Economy. 
(Homewood, IL: Irwin). 
 
Fei, John C.H., Gustav Ranis, and Shirley W.Y. Kuo (1978). “Growth and Family 
Distribution of Income by Factor Components,” Quarterly Journal of Economics 92: 17-
53. 
 
Fei, John C.H., Gustav Ranis, and Shirley W.Y. Kuo (1979). Growth with Equity: The 
Taiwan Case. (Oxford: Oxford University Press). 
 
Fields, Gary S. (1972). "Private and Social Returns to Education in Labour Surplus 
Economies," Eastern Africa Economic Review, June. 
 
 40
Fields, Gary S. (1975). “Rural-Urban Migration, Urban Unemployment and 
Underemployment, and Job Search Activity in LDC’s,” Journal of Development 
Economics 2: 165-188. 
 
Fields, Gary S. (1979a). “Income Inequality in Urban Colombia: A Decomposition 
Analysis,” Review of Income and Wealth 25: 327-341. 
 
Fields, Gary S. (1979b). “A Welfare Economic Approach to Growth and Distribution in 
the Dual Economy, Quarterly Journal of Economics 372: 325-354.  
  
Fields, Gary S. (1980a). “Education and Income Distribution in Developing Countries: A 
Review of the Literature,” in Timothy King, ed., Education and Income. (Washington: 
World Bank).  
 
Fields, Gary S. (1980b). Poverty, Inequality, and Development. (New York: Cambridge 
University Press). 
 
Fields, Gary S. (1984). “Employment, Income Distribution and Economic Growth in 
Seven Small Open Economies,” The Economic Journal, 94: 74-83. 
 
Fields, Gary S. (1989). “On-the-Job Search in a Labor Market Model: Ex-Ante Choices 
and Ex-Post Outcomes,” Journal of Development Economics, 30: 159-178. 
 
Fields, Gary S. (1990). “Labour Market Modeling and the Urban Informal Sector: Theory 
and Evidence,” in David Turnham, Bernard Salomé, and Antoine Schwarz, eds., The 
Informal Sector Revisited. (Paris: Development Centre of the Organisation for Economic 
Co-Operation and Development). 
 
Fields, Gary S. (1994). “Changing Labor Market Conditions and Economic Development 
in Hong Kong, the Republic of Korea, Singapore, and Taiwan, China,” The World Bank 
Economic Review 8: 395-414. 
 
Fields, Gary S. (1997). “Wage Floors and Unemployment : A Two Sector Analysis,” 
Labour Economics 4: 85-92. 
 
Fields, Gary S. (1999). “Employment Generation and Poverty Alleviation in Developing 
Economies,” in Ulrich Hiemenz, ed., Growth and Competition in the New Global 
Economy. (Paris: OECD Development Centre). 
 
Fields, Gary S. (forthcoming). “A Welfare Economic Analysis of Labor Market Policies 
in the Harris-Todaro Model,” Journal of Development Economics. 
 
Fields, Gary S., Paul L. Cichello, Samuel Freije, Marta Menéndez, and David Newhouse 
(2003). “Household Income Dynamics: A Four-Country Story,” Journal of Development 
Studies 40: 30-54. 
 
 41
Fields, Gary S., Murray Leibbrandt, and Jeremy Wakeford (2000). Key Labour Market 
Elasticities. Report prepared for the Government of South Africa. 
 
Fields, Gary S. and Jennifer O’Hara Mitchell (1999). “Changing Income Inequality in 
Taiwan: A Decomposition Analysis,” in Gary Saxonhouse and T.N. Srinivasan, eds., 
Development, Duality, and the International Economic Regime: Essays in Honor of 
Gustav Ranis. (Ann Arbor, MI: University of Michigan Press). 
 
Fields, Gary S. and Henry Y. Wan, Jr. (1989). “Wage-Setting Institutions and Economic 
Growth,” World Development. 
 
Ghose, Ajit K. (2003). Jobs and Incomes in a Globalizing World. (Geneva: ILO). 
 
Gindling, T.H., Jr. (1991). “An Investigation into Labor Market Segmentation: The Case 
of San José, Costa Rica,” Economic Development and Cultural Change.  
 
Hamermesh, Daniel (1993). Labor Demand. (Princeton: Princeton University Press). 
 
Harberger, Arnold C. (1971). “On Measuring the Social Opportunity Cost of Labor,” 
International Labour Review 103: 559-579. 
 
Harris, John and Michael Todaro (1970). “Migration, Unemployment, and Development: 
A Two Sector Analysis,” American Economic Review, 40: 126-142.  
 
Hart, Keith (1973). “Informal Income Opportunities and Urban Employment in Ghana,” 
Journal of Modern African Studies. 11: 61-89. 
 
Heady, Christopher J. (1981). "Shadow Wages and Induced Migration," Oxford 
Economic Papers. 
 
House, William J. (1984). “Nairobi’s Informal Sector: Dynamic Entrepreneurs or Surplus 
Labor?” Economic Development and Cultural Change 32: 277-302. 
 
Hymer, Stephen and Stephen Resnick (1969). “A Model of an Agrarian Economy with 
Nonagricultural Activities, American Economic Review. 59: 493-506. 
 
Inter-American Development Bank (2003). Good Jobs Wanted. (Washington: Inter-
American Development Bank). 
 
International Labour Office (1972).  Employment, Incomes, and Equality: A Strategy for 
Increasing Productive Employment in Kenya. (Geneva: ILO). 
 
International Labour Office (1998). Labour and Social Issues Relating to Export 
Processing Zones. (Geneva: ILO). 
 
 42
International Labour Office (2002). Women and Men in the Informal Economy:            
A Statistical Picture. (Geneva: ILO). 
 
International Labour Office (2003).  Global Employment Trends. (Geneva: ILO). 
 
Jhabvala, Renana, Ratna M. Sudarshan, and Jeemol Unni (2003). Informal Economy 
Centrestage: New Structures of Employment. (New Delhi: Sage Publications). 
 
Jorgenson, Dale W. (1967). “Surplus Agricultural Labour and the Development of a Dual 
Economy,” Oxford Economic Papers 19: 288-312. 
 
Karoly, Lynn A. and Gary Burtless (1995). “The Effects of Rising Earnings Inequality on 
the Distribution of U.S. Income,” Demography, August, 379-405. 
 
Khan, M. Ali (1989). “The Harris-Todaro Model,” in J. Eatwell et al., The New Palgrave. 
(London: MacMillan). 
 
Knight, John B. (1976). “Explaining Income Distribution in Less Developed Countries: A 
Framework and an Agenda,” Bulletin of the Oxford Institute of Economics and Statistics. 
 
Kuznets, Simon (1955). “Economic Growth and Income Inequality,” American 
Economic Review 45: 1-28. 
 
Lee, Chang-Hee (2002). “The Minimum Wage,” Asian Labour Update 42. 
 
Leibenstein, Harvey (1957). “Underemployment in Backward Economies, Journal of 
Political Economy 65. 
 
Lewis, W. Arthur (1954). “Economic Development with Unlimited Supplies of Labour,” 
Manchester School, 22: 139-191. 
 
Lewis, W. Arthur (1972). “Reflections on Unlimited Labor,” in Luis DiMarco, ed., 
International Economics and Development: Essays in Honor of Raul Prebisch. (New 
York: Academic Press), 75-96. 
 
Lin, C.-C. (1989). “The Basic Labor Standards Law and Operation of Labor Market: 
Theory and Partial Empirical Results,” Academia Sinica. 
 
Lindbeck, Assar and Dennis Snower (1988). Insider-Outsider Theory of Employment and 
Unemployment. (Cambridge, MA: MIT Press). 
 
Maloney, William F. (2003). “Informal Self-Employment: Poverty Trap or Decent 
Alternative,” in Gary S. Fields and Guy Pfeffermann, eds., Pathways Out of Poverty. 
(Boston: Kluwer). 
 
 43
Marcouiller, Douglas, Veronica Ruiz de Castilla, and Christopher Woodruff (1997). 
“Formal Measures of the Informal Sector Wage Gap in Mexico, El Salvador, and Peru,” 
Economic Development and Cultural Change 45: 367-392.  
 
Meade, James A. (1976). The Just Economy. (London: Allen and Unwin). 
 
Merton, Robert K. (1968). “The Matthew Effect in Science,” Science 159: 56-63. 
 
Mincer, Jacob (1962). “On-the-Job Training: Costs, Returns, and Some Implications,” 
Journal of Political Economy. LXX Supplement: 50-79. 
 
Mincer, Jacob (1974).  Schooling, Experience, and Earnings. (New York: Columbia 
University Press for the National Bureau of Economic Research). 
 
Moene, Karl Ove (1988). "A Reformulation of the Harris-Todaro Mechanism with 
Endogenous Wages," Economics Letters. 
 
Moene, Karl Ove (1992). "Poverty and Landownership," American Economic Review, 
March. 
 
Moran, Theodore H. (2002). Beyond Sweatshops: Foreign Direct Investment and 
Globalization in Developing Countries. (Washington, DC: Brookings Institution). 
 
Nelson, Richard R. (1966). “A Theory of the Low-Level Equilibrium Trap in 
Underdeveloped Economies,” American Economic Review 46: 894-908. 
 
Noe, Raymond A., John R. Hollenbeck, Barry Gerhart, and Patrick M. Wright (2000). 
Human Resource Management. (Boston: Irwin McGraw-Hill). 
 
Osmani, S.R. (1991). “Wage Determination in Rural Labour Markets: The Theory of 
Implicit Co-Operation,” Journal of Development Economics 34: 3-23. 
 
Park, Young-Bum (1991). “Union/Non-Union Wage Differentials in the Korean 
Manufacturing Sector,” International Economic Journal. 
 
Raff, Daniel and Lawrence Summers (1987). “Did Henry Ford Pay Efficiency Wages?” 
Journal of Labor Economics. 
 
Ranis, Gustav (forthcoming). “Is Dualism Worth Revisiting?”  in Alain de Janvry and 
Ravi Kanbur, eds., Poverty, Inequality, and Development: Essays in Honor of Erik 
Thorbecke. 
 
Ranis, Gustav and John C. H. Fei (1961). “A Theory of Economic Development,” 
American Economic Review, 51: 533-565. 
 
Ranis, Gustav and Frances Stewart (1993). “Rural Nonagricultural Activities in 
Development: Theory and Applicatrion,” Journal of Development Economics 40: 75-101. 
 44
 
Ranis, Gustav and Frances Stewart (1999). “V-Goods and the Role of the Urban Informal 
Sector in Development,” Economic Development and Cultural Change 47: 259-288. 
 
Robinson, Sherman (1976). “A Note on the U Hypothesis Relating Income Inequality to 
Economic Development,” American Economic Review 437-440. 
 
Rosenzweig, Mark (1988). “Labor Markets in Low Income Countries,” in Hollis Chenery 
and T.N. Srinivasan, eds., Handbook of Development Economics, Volume 1. 
(Amsterdam: North Holland). 
 
Roy, A.D. (1951). “Some Thoughts on the Distribution of Earnings,” Oxford Economic 
Papers. 3: 135-146. 
 
Sah, Raaj Kumar and Joseph E. Stiglitz (1985). "The Social Cost of Labor and Project 
Evaluation: A General Approach," Journal of Public Economics. 
 
Schultz, T.W. (1961). “Investment in Human Capital,” American Economic Review. LI: 
1-17 
 
Schultz, T.W. (1962). “Reflections on Investment in Man,” Journal of Political Economy. 
LXX Supplement: 1-8. 
 
Sen, Amartya K. (1967). “Review of J.C.H. Fei and G. Ranis, Development of the Labor 
Surplus Economy: Theory and Policy,” The Economic Journal 77: 346-349. 
 
Shorrocks, Anthony F. (1983). “The Impact of Income Components on the Distribution 
of Family Incomes,” Quarterly Journal of Economics 98: 311-331. 
 
Solow, Robert (1990). The Labour Market as a Social Institution. (Oxford: Blackwell).  
 
Spinanger, Dean (1985). “The Labor Market in Panama: An Analysis of the Employment 
Impact of the Labor Code,” Paper presented at the Seminar on Employment Policy in 
Latin America, Panama. 
 
Squire, Lyn (1981). Employment Policy in Developing Countries (New York: Oxford 
University Press for the World Bank). 
 
Starr, Gerald (1981). Minimum Wage Fixing. (Geneva: ILO). 
 
Stiglitz, Joseph E. (1971). “Alternative Theories of Wage Determination and 
Unemployment in LDC’s,” University of Nairobi, processed. 
 
Stiglitz, Joseph E. (1974). “Alternative Theories of Wage Determination and 
Unemployment in LDCs” The Labour Turnover Model, Quarterly Journal of Economics 
88: 194-227. 
 45
 
Stiglitz, Joseph E. (1976). “The Efficiency Wage Hypothesis, Surplus Labor, and the 
Distribution of Labour in LDCs,” Oxford Economic Papers 28: 185-207. 
 
Stiglitz, Joseph E. (1982). “The Structure of Labor Markets and Shadow Prices in 
LDCs,” in Richard H. Sabot, ed., Migration and the Labor Market in Developing 
Countries. (Boulder, CO: Westview). 
 
Stiglitz, Joseph E. (1985). “Equilibrium Wage Distributions,” The Economic Journal. 95: 
595-618. 
 
Stiglitz, Joseph E. (1999). “Duality and Development: Some Reflections on Economic 
Policy,” in Gary Saxonhouse and T.N. Srinivasan, eds., Development, Duality, and the 
International Economic Regime: Essays in Honor of Gustav Ranis. (Ann Arbor, MI: 
University of Michigan Press). 
 
Tidrick, Gene (1975). “Wage Spillover and Unemployment in a Wage-Gap Economy: 
The Jamaican Case,” Economic Development and Cultural Change. 
 
Todaro, Michael P. (1969). “A Model of Labor Migration and Urban Unemployment in 
Less Developed Countries,” American Economic Review 39: 138-148. 
 
Tokman, Victor (1987). “El Sector Informal: Quince Años Despues,” PREALC. 
 
Turnham, David (1971). The Employment Problem in Less Developed Countries. (Paris: 
Development Centre of the Organisation for Economic Co-Operation and Development). 
 
Turnham, David (1993). Employment and Development: A New Review of Evidence. 
(Paris: Development Centre of the Organisation for Economic Co-Operation and 
Development). 
 
Ulrich, Dave (1997). Human Resource Champions. (Boston: Harvard Business School 
Press).  
 
Weiss, Andrew (1990). Efficiency Wages. (Princeton: Princeton University Press). 
 
Welch, Finis (1974). “Minimum Wage Legislation in the United States,” Economic 
Inquiry 12: 285-318. 
 
World Bank (1990). World Development Report 1990. (Washington: World Bank). 
 
World Bank (1995). World Development Report 1995. (Washington: World Bank). 
 
Yoo, Gyeongjoon (1995). An Analysis and Decomposition of Changing Labor Income 
Distribution in Korea. Unpublished Doctoral Dissertation, Cornell University. 
 46
Figure 1. 
The Standard Market-Clearing Labor Market Model. 
 
 
 
 
 
 
 
 Real Wage 
 
S 
 
 
 
 
 
 
 
W*D 
Q uantity  
L* Of Labor 
 47
 
Figure 2. 
The Integrated Labor Market Model:  
A Higher Demand for Labor in One Sector  
Raises Wages In all Sectors. 
 
 
 
 
Real Wage 
WM´ WA´ 
W * W * M A
DM´ 
DA DM 
OM O  E E´ A  
 
 
 48
Figure 3. 
In a Crowding Model, a Higher Wage in Sector A 
Reduces Employment in That Sector, 
Crowds Workers into Sector B, 
and Lowers the Sector B Wage. 
 
 
Labor Market Consequence of Crowding Workers Out of Sector A 
Real Wage 
SA´ SA 
WA´ 
WA 
DA 
EA´          EA Quantity of Labor 
Labor Market Consequence of Crowding Workers Into Sector B 
Real Wage SB SB´ 
WB 
WB´ 
DB 
EB          EB´ Quantity of Labor 
 
 49
 
 
Figure 4. 
Unemployment and Average Real Wages in Taiwan. 
 
8
6 6.3
4 4.3
2 1.5 1.3 1.7
0
1955 1963 1972 1981 1990
Year
Unemployment Rate
 
 
 
1000
800 805
600
400 400
200 183
100 102
0
1950 1960 1970 1980 1990 2000
Year
Real Wages in Manufacturing (1954 = 100)
 
 
Sources: Fields (1984, 1994). 
 
 
 
 
  
 
Real Wages Unemployment Rate,
Percentage