250 word forum post 424 week 2

Labor Market Rigidities and Informality in Colombia Mondragón-Vélez, Camilo; Peña, Ximena; Wills, Daniel . Economia ; Washington  Vol. 11, Iss. 1,  (Fall

2010): 65-101.

ProQuest document link

ABSTRACT  

Informality has been at the center of the economic debate in Colombia as a result of the high levels prevalent in

the country and its substantial increase during the 1990s. The informal sector includes a range of heterogeneous

activities, from unpaid labor to a number of unregulated salaried jobs. Informality is thought to have negative

implications, mainly through inferior working conditions, lack of formal health, unemployment, and old age

insurance, and low productivity levels for firms. This paper brings new elements to the study of informality in

Colombia and suggests directions for future research. The authors study the evolution of informality between 1984

and 2006 -- a period that includes both expansions and recessions, structural reforms of the labor market, and

significant variation in nonwage costs and the minimum wage. Informal workers are vulnerable, are frequently

uncovered by social security, have relatively low education, and on average earn lower wages than formal workers. FULL TEXT  

Informality has been at the center of the economic debate in Colombia as a result of the high levels prevalent in

the country and its substantial increase during the 1990s. The informal sector includes a range of heterogeneous

activities, from unpaid labor to a number of unregulated salaried jobs. Informality is thought to have negative

implications, mainly through inferior working conditions, lack of formal health, unemployment, and old age

insurance, and low productivity levels for firms. Alternative definitions of informality have been proposed in the

literature, each implying a different approach to this phenomenon.

The Colombian labor market is characterized by high nonwage costs and a high minimum wage relative to the

economy's level of productivity. Nonwage costs are costs faced by the employer and include health and pension

contributions, payroll taxes, and transportation (commuting) subsidies.1 These labor market rigidities imply that

the formal sector, where workers and employers comply with regulations, is less able to adjust to the business

cycle than the informal sector. Hence, economic policy originally designed to protect workers might actually be

worsening employment conditions by increasing informality.

This paper brings new elements to the study of informality in Colombia and suggests directions for future

research. We study the evolution of informality between 1 984 and 2006 - a period that includes both expansions

and recessions, structural reforms of the labor market, and significant variation in nonwage costs and the

minimum wage. By generating individual or city-level variation, we are able to disentangle the effects of nonwage

costs, the minimum wage, and the business cycle on informality. We begin our analysis by considering alternative

definitions of informality, two of which we adopt (primarily driven by data availability): the definition used by

Colombia's National Administrative Department of Statistics (DANE, for its initials in Spanish), which is based on

firm size and occupation, and a definition based on contribution to health insurance (as a proxy for compliance

with labor market regulations).

With regard to the empirical analysis, we first estimate the probability of being informal as a function of individual

characteristics, the business cycle, and labor market rigidities. Our results suggest that rises in nonwage costs

and the minimum wage are highly correlated with informal sector growth. Next, we look at the transitions between

sectors. On the one hand, we measure the transition flows between the formal and informal sectors using

transition matrices. These describe, for example, the proportion of job destruction in the formal sector that is

absorbed by the informal sector. On the other hand, we estimate the effect of labor market rigidities on the

likelihood of switching sectors (controlling for idiosyncratic characteristics and macroeconomic conditions), to

determine their role in the decision to make the transition. We find that labor market rigidities are important drivers

of the transition into informality, particularly for low-skilled workers. However, further research is needed to

understand the channels through which labor market rigidities affect the transition into the formal sector, in

particular for workers with high educational attainment.

One strand of the literature associates informality with labor market rigidities in Colombia. Núñez finds a positive

relation between informality and income taxes on labor revenue for the period 1 988-98. 2 Sánchez, Duque, and

Ruiz find that increases in labor market rigidities increase informality, unemployment, and its duration, based on

aggregate data.3 Using a firm panel from the industrial sector, Kugler and Kugler find that a 10 percent increase in

payroll taxes decreases formal employment between 4 and 5 percent.4 Santa Maria, Garcia, and Mujica use

individual data from Colombia's household survey; they find that the subsidized regime, financed through nonwage

costs, has increased the incentives to become informal, thus acting as a subsidy to informality.5 Our results

suggest that an increase of 10 percentage points in nonwage costs is associated with an increase of 5 to 8

percentage points in the size of the informal sector. Some authors characterize informal workers and study

informality from a segmentation perspective. Flórez finds that some informal workers are excluded from the

formal sector, while others opt out.6 Perry and others document two coexistent phenomena in the Latin American

region: exclusion from the formal sector as a result of labor market segmentation; and exit from the formal sector

as some workers find better conditions in informality.7 They show that whereas the self-employed choose to be so

in many Latin American countries, a sizeable fraction of the Colombian self-employed seem to be excluded from

the formal sector.8 Bernal reports that one-half of informal workers would accept a formal job for a wage equal to

or lower than the one they currently make.9 Along these lines, our results support the idea that low-skilled informal

workers are being excluded from the formal sector and suggest that high-skilled workers may also have exit

motives.

The paper is divided into five sections. The following section compares and documents the evolution of informality

across alternative definitions, and the paper then describes the evolution of labor market rigidities. Subsequent

sections estimate the effect of nonwage costs and the minimum wage on the size of the informal sector and the

probability of switching sectors and wages. The final section concludes.

The Evolution of Informality

This section describes the different definitions of informality that we consider and analyzes the extent to which

they coincide. We also explore the effect of the business cycle and relative wages on informality. Our study uses

data from the Colombian Household Survey, a repeated cross-section carried out by Colombia's National

Administrative Department of Statistics (DANE), for the period 1984-2006. The survey collects information on

demographic and socioeconomic characteristics of the population, such as gender, age, marital status, and

educational attainment, together with labor market characteristics for the population aged twelve or more

including occupation, job type, income, and industrial sector.

Based on data availability, our analysis focuses on Colombia's eleven main cities between 1984 and 2000 and the

thirteen main cities for the period 2001-06. In particular, we use the household survey's informality module, which

allows for several empirical definitions of informality as described in the next subsection. This module is available

every two years before 2001 and yearly thereafter. We use observations with a complete set of covariates and

restrict the sample to workers between fifteen and seventy years of age, who report working between sixteen and

eighty-four hours per week. The size of the weighted samples ranged from 1.7 million workers in 1984 to about 6.5

million in 2006.

Informality Definitions

The literature includes a number of alternative definitions of informality, usually dictated by data availability. This

section characterizes the evolution of informality using all definitions available from the Colombian Household

Survey, and it explores the extent to which definitions involving compliance (or noncompliance) with labor market

regulations coincide with DANE'S official definition related to firm size and occupation.

DANE defines informal workers as those who work in firms with ten or fewer employees; are unpaid family aids

and housekeepers; are self-employed (except for independent professionals and technicians); or are business

owners of firms with ten employees or fewer. In what follows, we refer to this definition as Firm Size and

Occupation, since it is largely driven by these two dimensions. This definition has been criticized in the literature

for not measuring the phenomenon directly.10 That is, it does not explicitly include any criteria related to labor

market regulations even though the negative aspects of informality have to do with lack of compliance.

We also consider three definitions related to social protection contribution that capture whether workers pay for

access to the benefits associated with formal employment. The first social protection definition has to do with

oldage insurance: we define informal workers as those who do not make pension contributions (and we call this

definition Pension). The next informality criterion is the lack of health insurance contributions (termed Health

below)." Finally, we consider the Health and Pension criterion that defines workers as formal if they contribute to

both health insurance and a pension fund and informal otherwise. This is the most comprehensive criterion and

thus implies the highest levels of informality.

Figure 1 presents the evolution of the informality rate for the period 1984-2006 across these alternative

definitions.12 According to Firm Size and Occupation, informality was stable at around 52 percent from 1984 to

1996, grew steadily between 1996 and 2001 to 56 percent, and remained at this level until the end of the period.

This increase is sizable and is at me center of the domestic debate. Under Pension, the percentage of informal

workers increased from 1998 to 2001, and then continually decreased until 2006. Informality as measured by

Health decreased between 1984 and 1996, then increased from 44 percent to 49 percent between 1 996 and 2000,

and finally decreased between 2001 and 2006. Informality is higher if measured through pension contributions as

compared to health contributions, suggesting either that workers value health more than old-age insurance or that

they have informal old-age insurance mechanisms such as family or friends. Informality measured by Health and

Pension closely follows the Pension criterion. The dotted line, measured on the right axis, is the urban

unemployment rate. The informality definitions that have to do with compliance seem to follow the cycle more

closely than the Firm Size and Occupation definition. However, according to all definitions, informality increased in

the second half of the 1 990s and peaked around 2000, after the deep economic recession of the late 1990s.

To What Extent do Alternative Definitions of Informality Coincide?

Despite the difference in levels, the evolution of informality measured by Pension and Health is very similar. When

we take a closer look at the composition of these groups, we find that those classified as informal under Health are

almost a subset of those classified as informal in Pension: on average, one percentage point of those considered

informal using the Health criterion are formal under Pension. Hence, given that Health captures the relevant

dynamics of informality measured by compliance and is available for a longer period than Pension, we use it as the

benchmark definition for compliance with labor market regulations in the remainder of the paper.

The levels of informality suggested by Health are lower than those measured by Firm Size and Occupation.

Moreover, their trends are not identical over time. This raises the question of whether these alternative definitions

are classifying the same workers as informal. To address this issue, we first present a Venn diagram portraying the

extent to which these definitions coincide using data from 2006 (figure 2).13 The diagram has three main

implications. First, the majority of informal workers under Firm Size and Occupation or Health (38.9 percent of

total workers) are informal under both definitions. This is because noncompliance with social security regulations

is a smallfirm phenomenon. Second, the Firm Size and Occupation definition captures the bulk of workers

considered informal under Health (90 percent of health informality), despite the fact that it does not include any

criteria regarding social security compliance. Third, the difference between the two definitions is still sizable, as

about one-fourth of all workers is classified as formal or informal depending on the definition used.

Differences in informality levels across firm size and occupations measured by Health are staggering, as displayed

in figure 3. Less than 20 percent of individuals working in firms with more than ten employees are informal,

compared with over 80 percent of those working alone. This reflects the fact that not only do small firms have a

harder time covering compliance costs than large firms, but they also find it is easier to stay below the

government's radar. Differences in informality measured by Health across occupations are also important: about

85 percent of the self-employed and household service workers (who are informal according to the Firm Size and

Occupation definition) do not contribute to health insurance, whereas few government employees are informal.

As mentioned above, the classification differences between Firm Size and Occupation and Health are substantial.

Figure 4 shows the evolution of the concurrence between the two definitions, that is, the information contained in

the Venn diagram over time. Panel A of the figure shows the evolution of the classification differences between

1984 and 2006. The percentage of workers classified as informal under Health and as formal under Firm Size and

Occupation were fairly stable for the period of study at around 7 percent. Workers classified as informal using Firm

Size and Occupation but formal using Health increased from 10 percent in 1984 to 17 percent in 2006. This may be

related to increasing enforcement efforts in regard to health insurance contributions. Panel B of figure 4 shows the

percentage of workers who are either informal or formal under both definitions. The shares range between 36

percent and 42 percent.14

Informality and the Business Cycle

The period under study includes both expansions and recessions, structural reforms in the labor market, and

significant variation in nonwage costs and the minimum wage. To study the effects of these movements on

informality, we begin by looking at the correlation between the size of the informal sector (across alternative

definitions) and the business cycle for the period 1984-2006. If informality is a disadvantaged sector of a

segmented labor market detached from formal activity, it should be countercyclical, expanding during downturns

to absorb displaced workers from the formal sector. If, on the other hand, the size of the informal sector is

procyclical, this would be suggestive of a micro-entrepreneurial sector, linked to the formal sector through the

provision of low-cost goods and services. Our estimates show that informality is positively correlated with

unemployment, regardless of the definition. Although we have few observations, the correlation is statistically

significant (and substantially higher) when measured by Firm Size and Occupation or by Health and Pension. This

implies that the size of the informal sector moves in the opposite direction of the business cycle, suggesting

segmentation in the labor market.15

Relative Wages: Returns to the Formal and Informal Sectors

The formal and informal sectors differ substantially in their associated levels of earnings.16 In this regard, the

informal sector is not attractive. As portrayed in figure 5 (panel A), informal workers earn, on average, 40-60

percent of what their formal peers earn, regardless of the definition of informality. Informal wages improved

relative to formal wages until 1994, but then deteriorated significantly up to the early 2000s - a time when both

nonwage costs and the minimum wage increased substantially and the country faced the deepest recession in

recent history. Informal wages recovered starting in 2002, but even after several years of high economic growth,

informal workers' relative earnings in 2006 were still lower than in the 1980s.

Relative returns also differ markedly across education groups (see figure 5, panel B), but the trends for each group

are similar to the trend observed for the total.17 Using the Health definition, informal high-skilled workers (those

with completed secondary education or more) were relatively better off than their low-skilled peers at the

beginning of the period. In the second half of the 1990s, the relative wages of the high-skilled group decreased

more than those of low-skilled individuals. The results are similar when we use the Firm Size and Occupation

definition.

Labor Market Rigidities and Informality

The Colombian labor market is characterized by relatively high nonwage costs and wage inflexibility associated

with the minimum wage.18 Such rigidities are thought to contribute to the high unemployment and informality

rates prevalent in the economy, mainly by limiting the formal sector's ability to adapt to economic cycles. The

coexistence of high levels of nonwage costs and a binding minimum wage reinforce each other's negative effects.

The literature includes extensive studies of die effect of payroll taxes and the minimum wage on the labor market.

An increase in nonwage costs makes labor relatively more expensive than capital. Hence, employers can either

shift production away from labor, destroying formal jobs, or try to pass through the additional cost to employees

via lower wages. However, the nominal downward rigidity imposed by a high and binding minimum wage implies

that formal employers cannot always completely pass on nonwage costs to workers via prices, which generates

further job destruction and pushes more workers into either informality or unemployment. Increases of the

minimum wage beyond price and productivity growth can also generate job destruction directly. If the minimum

wage rises in real terms, firms destroy formal jobs whose productivity levels lie between the old and new minimum

wage level.

Gruber shows that payroll taxes reduce labor demand and wages in equilibrium. ,9 Kugler and Kugler introduce

nonwage costs and a minimum wage into the Stiglitz-Shapiro model of efficiency wages; they find that nonwage

costs decrease formal employment when taxes are not completely shifted to workers via lower wages.20 This is

always the case when the minimum wage is binding. Albrecht, Navarro, and Vroman, in an extension of the

MortensenPissarides search and matching model, find that rises in nonwage costs increase the size of the

informal sector.21 Fortin, Marceau, and Savard suggest that higher taxes on labor and a higher minimum wage

increase the number of firms operating in the informal sector, decreasing the demand of informal labor at the

expense of decreasing vacancies for formal workers.22 Kristensen and Cunningham show that rises in the

minimum wage increase the relative cost of labor, which, given the presence of downward wage rigidities, makes

the formal sector unable to adjust via prices and forces it to do so via quantities.23 Card and Krueger review the

literature on minimum wages and suggest that an increase in the minimum wage increases wages and reduces

employment in the covered (formal) sector.24 The effect in the uncovered (informal) sector depends on the model

of labor supply chosen, but it is either negative or ambiguous on wages and either positive or ambiguous on

employment.

Colombia implemented very profound labor reforms in the early 1990s, as did other countries in Latin America. In

particular, Law 100 of 1993 structurally reformed the Colombian social security system, both in health insurance

and pensions. Two regimes were created in health insurance: contributive (namely, an employment-based

mandatory insurance system) and subsidized. Before 1993, health insurance contributions amounted to 8 percent

of wages, and the reform raised contributions to 12 percent.25 Similarly, pension contributions rose from 8 percent

to 13.5 percent of wages in a three-year period.26 Law 100 thus increased nonwage costs by 9.5 percentage

points.

In addition to the steep increase in nonwage costs, Bernal and others suggest that four aspects of the current

design of the Colombian social protection system, embedded in Law 100 of 1993, generate informality.27 First,

since social security benefits are multidimensional, workers who prefer partial coverage over full coverage may opt

out of the whole package and hence become informal. Second, there is a percentage of the population for whom

the quality of the services offered under the contributive and subsidized regimes is comparable. Since the

subsidized regime is free of charge, it is optimal for eligible workers in this population to remain in the subsidized

regime, which implies passing up formal jobs.28 Third, the current design does not allow for an easy transition

between regimes. Thus, workers in the subsidized regime may be reluctant to accept a formal job and enter the

contributive regime, since reentering the subsidized regime is time consuming. Finally, workers who are eligible for

the subsidized regime (that is, who are classified as poor) are also eligible for an array of social programs. By

accepting a formal job, these workers and their families gain access to the contributive health insurance regime

but may potentially loose other benefits.

The minimum wage in Colombia is high. A cross-country comparison reveals that the Colombian minimum wage is

the highest in Latin America, as measured by the ratio of the minimum wage to the median wage.29 In addition,

the minimum wage in Colombia is binding, and it indexes the whole wage distribution, since increases in the

minimum wage are adopted as a benchmark for wage increases in the whole economy.30 The effect is strongest

for wages close to the minimum, and it decreases for higher wages in the distribution. A sizeable fraction of the

workforce earns less than the minimum, especially in the informal sector. Hence, not only does the existence of a

minimum wage level generate wage inflexibilities, especially around the minimum, but its evolution can have

important effects on the dynamics of the labor market.

In this paper, nonwage costs include severance payments, health and pension contributions, payroll taxes, two

annual bonuses, vacation payments, and a transportation (commuting) subsidy.31 The latter is fixed by law at the

beginning of each year and applies exclusively to workers earning less than two times the minimum wage. In our

calculation of nonwage costs, this transportation subsidy is included as a percentage of the minimum wage.32 For

example, in 1993, 45.8 percentage points of an employee's earnings corresponded to nonwage costs, of which 7.0

were health insurance, 8.0 were pension contributions, 9.3 were severance payments, 9.0 were payroll taxes, 4.2

were vacation payments, and 8.3 were two annual bonuses. In addition, the transportation subsidy accounted for

9.3 percent of the minimum wage.33 Our measure of nonwage costs does not include severance payments for

unfair dismissals (as in Cárdenas and Bernal), because data are not available for estimating the probability of

being unfairly dismissed.34 We also do not include the work uniform subsidies mandated by law, given the

difficulty of determining their costs over time and applicability across sectors. Finally, the total nonwage costs we

include in our calculations is a proxy for the percentage of the wage employers would have to pay, regardless of

whether they actually pay them.

Figure 6 displays the informality rates under Firm Size and Occupation and Health, alongside the ratio of the

minimum wage to the median wage (MinW) and nonwage costs. Both nonwage costs and MinW display secular

increases during the period of study. The major increase in nonwage costs was due to Law 100 of 1993, which

substantially increased payroll taxes. MinW remained relatively stable at around 85 percent from the mid-1980s to

the early 1990s, decreased to 70 percent in the mid-1990s, escalated to over 90 percent by 2000, and remained at

this level until the end of the analysis period. This secular increase was the result of an inflation forecast error,

combined with legislation by the Constitutional Court.35 This combination of factors has kept the minimum wage

at a very high level since that time. As shown in figure 6, the increase in nonwage costs preceded the increase in

informality, while the increase in MinW coincided with it.

In sum, some analysts regard the increase in nonwage costs and the minimum wage as key drivers of the high

levels of unemployment and informality in Colombia.36 High nonwage costs increase the cost of formality for both

firms and workers, generating incentives for workers to avoid compliance either voluntarily (through exit from the

formal workforce, in response to labor supply behavior) or involuntarily (through exclusion from formal jobs,

related to labor demand behavior). Increases in the minimum wage could imply higher informality via exclusion

from the formal sector. The following sections explore the empirical validity of these hypotheses for Colombia over

the past twenty years.

The Size of the Informal Sector

We exploit the observed variation in MinW and nonwage costs over the period 1984-2006 to determine the effect

on the size of the informal sector. In these estimations we use the two alternative definitions of informality

discussed earlier (namely, Firm Size and Occupation and Health). We pool observations from all the quarters for

which the informality module is available in the period of interest to estimate the effects of labor market rigidities

on the probability of being informal, using a probit model and controlling for individual characteristics. We

estimate the following model:

INF^sub ict^ = ^sub 0^ + ^sub 1^MinW^sub ct^ + ^sub 2^NWC^sub it^ + ^sub 3^Income_pc^sub ict^ + ^sub 4^X^sub

ict^ + ^sub ict^,

where INF^sub ict^, is a dummy variable that takes the value of one if individual i residing in city c at time t is

classified as informal and zero otherwise, MinW^sub ct^ is the ratio of the (national) minimum wage to the median

wage in city c at time t, and NWC^sub it^, is the level of nonwage costs as a percentage of the individual's salary.37

NWC varies across individuals as well as over time. First, highly paid individuals have higher pension contributions

(as a share of their wages) in order to finance part of the country's subsidized regime. For instance, workers

earning more than four times the minimum wage have to contribute an additional percentage point of their

monthly earnings to the pension fund, and workers earning more than sixteen times the minimum wage contribute

up to two extra percentage points. Second, workers earning less than two times the minimum wage receive a

transportation (commuting) subsidy that is about 10 percent of the minimum wage.38 Income_pc^sub ict^ is real

household income per capita, which is a proxy for the business cycle, and X^sub ict^ is a vector of individual

controls that include age and age squared, educational attainment, gender, and marital status (defined as marriage

or cohabitation), in addition to industrial sector and city dummies.39 The regressions were estimated using

cluster-robust standard errors (clustered on year and city) to account for the fact that although our sample spans

over twenty-two years, we only have observations in fourteen periods, while MinW varies across years and cities

but not across individuals.

The estimated effects of NWC, MinW, and the business cycle on the size of the informal sector, measured through

the workers' probability of being informal, are shown in table 1 . The results suggest that labor market rigidities can

have sizeable effects on the rate of informality, as the increases observed in NWC and MinW during the period of

study are associated with a higher fraction of informal workers. For instance, an increase of 10 percentage points

in nonwage costs is associated with an increase in the probability of being informal of 5 to 8 percentage points,

depending on the definition of informality. To put this result in perspective, Kugler and Kugler estimate that a 10

percent increase in nonwage costs decreases formal employment by 4-5 percent.40 Similarly, the results obtained

when using the government's official definition of informality suggest that a rise of 20 percentage points in MinW

(as occurred between 1996 and 2001) implies an increase of 2 percentage points in the informality rate.41 The

economic cycle coefficient, when significant, implies countercyclicality (though it generally has second-order

effects).42

The magnitude and significance of the marginal effects of nonwage costs and the minimum wage on the total

sample vary with the definition of informality. While effects are higher when informality is measured as Firm Size

and Occupation, the coefficient of MinW is neither statistically nor economically significant when using Health. To

address these robustness issues, we explore potential differences across education groups. Specifically, we study

heterogeneous effects of the model on two different groups: low-skilled workers, which includes individuals with

less than completed secondary education; and high-skilled workers, which includes workers with completed

secondary education or more. The estimates in table 2 generally suggest significant differences in the effects of

MinW and NWC on low- and highskilled workers, which could be interpreted as a gradual segmentation of the labor

force. First, higher MinW and NWC imply higher informality for those within the low-skilled group, regardless of the

definition of informality. These results could be linked to the concept of exclusion: higher MinW and nonwage

costs negatively affect the supply of formal jobs for low-skilled individuals, leaving them with the choice of

informal paid jobs or involuntary self-employment.43 Since the vast majority of informality is captured within this

group, we consider these results the most relevant in regard to the effects of labor market rigidities on the size of

the informal sector.

To facilitate the interpretation of the results, we isolate the net effect for the high-skilled group and its significance

level (see table 3).M The effects of MinW and NWC on high-skilled workers are clearly different from the effects on

low-skilled workers. Increases in MinW seem to decrease the probability of being informal for high-skilled workers,

although the effect is only weakly significant under Health and not significant under Firm Size and Occupation.

Given that the Colombian minimum wage is high and binding, rises in real terms turn out to be a regressive policy

in terms of the size of the informal sector: it seems to protect high-skilled workers while pushing lowskilled

workers into informality or unemployment.

Rises in nonwage costs, on the other hand, increase the probability of being informal for both low- and high-skilled

workers. When informality is measured by Firm Size and Occupation, nonwage costs affect low- and highskilled

workers equally since the interaction coefficient between NWC and high-skilled workers is not significant (table 2).

When we use Health, we see that nonwage costs affect the probability of being informal more for highskilled

individuals than for low-skilled ones. Further research is needed to explain this phenomenon. Finally, when we use

the Health definition, the effect of the business cycle is countercyclical for all workers, but the effect is stronger for

low-skilled workers. When the economy is contracting, workers with a low education level are more likely to be

informal, which suggests exclusion. Overall, these results suggest a strong link between informal lowskilled

workers and exclusion.

Transitions between the Informal and Formal Sectors

In this section we study the flow of agents in the labor force between the formal and informal sectors. We started

by developing transition matrices for each of the available cross-sections during the period 1986 to 2006 and then

characterized each of the flows involving entry into or exit from the informal sector by estimating transition

probabilities as a function of demographics, occupation-specific characteristics, and other idiosyncratic labor-

history factors. Since our database is composed of repeated cross-sections, we build the transitions within twelve-

month periods using retrospective questions. We observe the sector to which a worker belongs at the time of the

survey; consider only workers who switched jobs in the previous year; and only use the Firm Size and Occupation

definition, because retrospective information on contributions to health insurance is not available. Our results

indicate that in 1998, for example, 62.3 percent of those who were formal in their previous job remained formal and

37.7 percent became informal. For workers whose previous job was in the informal sector, 80.3 percent remained

informal while 19.7 percent moved into the formal sector. These values indicate some persistence across sectors.

As a share of the workers who switched jobs the previous year, 29.7 percent of workers were formal and remained

formal, 18.0 percent were formal and became informal, 12.1 percent were informal and became formal, and the

remaining 40.2 percent were informal and remained informal. The differences in the flows between workers

entering the informal sector and those exiting from it are statistically significant for 1998.

The evolution of these transitions across time is plotted in figure 7. Panel A plots the share of workers who

remained in either the formal or informal sector from one year to the next. The graph shows that persistence in the

informal sector decreased at the beginning and end of the period and increased between 1994 and 2001, while the

opposite is true for persistence in the formal sector. Panel B, which illustrates the trend for workers who moved

between sectors, suggests that the expansion in informality was generated by greater inflows of workers into the

informal sector than outflows, starting in 1984. The differences in flows into and out of the informal sector were

sizeable between 1996 and 2002, after the steep increase in nonwage costs and the minimum wage. By the end of

the period, however, the flows into and out of informality became very similar, so the size of the informal sector

stabilized.

We now study the effect of labor market rigidities and unemployment on the probability of moving into and out of

the informal sector. We use an extended probit specification similar to the one described in the last section, which

includes the unemployment spell and a dummy variable for whether individuals changed their economic activity as

part of the covariates. The dependent variable in these regressions measures twelve-month transitions from the

formal to the informal sector and vice versa, and we again include only those individuals who changed jobs in the

past year.45 The reported estimates thus measure the effects on the transition probability across the formal and

informal sectors conditional on changing jobs. As before, these transitions can only be observed for the Firm Size

and Occupation definition; regressions were estimated using cluster-robust standard errors (clustered on year and

city); and results are only reported for the covariates of interest (namely, Min W, NWC, household income per

capita as a proxy for the business cycle, the unemployment spell, and industrial sector change).46

The first column of table 4 shows the estimates for the transition probability into the informal sector. These results

imply that labor market rigidities are important drivers of the transition into informality, consistent with the

findings of the previous section regarding the size of the informal sector. A rise of 10 percentage points in

nonwage costs increases the probability of transitioning into informality (given that the worker switched jobs) by

8.7 percentage points, whereas a 20 percentage point increase in the ratio of the minimum wage to the median

wage does so by 6 percentage points. Unemployment spell effects have the expected sign, but they are not

significant.47

In regard to the transition into formality (third column of table 4), the effects of nonwage costs, the unemployment

spell, and the business cycle have the expected signs (though with small economic significance for the last two).

The probability of transitioning into formality decreases with higher nonwage costs or an unemployment spell, as

expected. Again, the positive correlation of the transition into formality with changes in economic activity arises

from the nature of informal activities and the processes of exclusion associated with them. While increases in

MinW raise the probability of transiting into informality, the reverse is not true: MinW effects appear both

economically and statistically insignificant in the transition into the formal sector. To explore behavioral

differences and explain the asymmetric effect of MinW on the transition probabilities, we explore, as before, the

robustness of these results across different education groups. Finally, changes in economic activity are positively

correlated with both the transition into informality and the transition into formality. The former suggests that this

decision is not necessarily related to a career path, and in line with exclusion.

The first column of table 5 shows that the effects on the transition into the informal sector for the total sample are

mainly driven by the low-skilled group. The strongly positive and significant effects of MinW and nonwage costs on

the probability of transitioning into informality are fully consistent with their effects on the size of the informal

sector for this particular group. This constitutes additional support to the idea of a segmented labor market;

workers are excluded from the formal sector as the effects of the minimum wage and nonwage costs work mainly

through quantity adjustments in the demand for low-skilled labor. To facilitate the interpretation of the results in

table 5, we highlight the net effects for the high-skilled group in table 6.48 For high-skilled workers, the effect of

increases in MinW (that is, the sum of the low-skilled group and the interaction coefficients) is not statistically

significant. This implies that increases in MinW do not affect the probability of transitioning from the formal to the

informal sector for this group. This result is in line with the findings from the last section. Higher nonwage costs

increase the probability of transitioning from the formal to the informal sector for all workers, but especially for

high-skilled ones.

Regarding the transition from the informal to the formal sector, increases in MinW are not statistically significant

for any group. However, lower nonwage costs imply a higher transition into formality for the low-skilled group, as

expected; this is linked to quantity adjustments via the demand for labor. The total effect for high-skilled

individuals (that is, the sum of the low-skilled group and the interaction coefficients) is not statistically significant.

Consistent with our findings in the last section, these estimations strongly support the exclusion of the low-skilled

group. When looking at heterogeneous effects, changes in economic activity are again positively correlated with

both the transition into informality and the transition into formality.

Conclusion

Informal workers are vulnerable, are frequently uncovered by social security, have relatively low education, and on

average earn lower wages than formal workers. The secular increase in the size of the informal sector in Colombia

is highly correlated with increasing labor market rigidities, namely, the minimum wage and nonwage costs. These

increased rigidities have made the formal sector less able to adjust to economic cycles. The coexistence of high

nonwage costs and a high minimum wage implies that the formal sector must adjust to the economic cycle

through quantities rather than wages, cutting back on mostly low-skilled jobs and forcing those workers into the

informal sector. Because the minimum wage is not binding in the informal sector, a high percentage of informal

workers is paid below the minimum. As more low-skilled workers move into the informal sector, wages in that

sector drop overall.

Rigidities also have consequences for the relative sizes of the formal and informal sector, which have ultimately

triggered the documented increase in the latter. All in all, our results suggest that labor market rigidities affect

lowskilled workers the most. Further research is needed to understand the channels through which labor market

rigidities affect the transition into the formal sector and, in particular, the motivations of such transitions for

informal workers with high educational attainment.

In terms of policy design, this paper provides useful evidence on the costs that labor market rigidities imply for

vulnerable workers. These workers are precisely the targeted population that the wage policies aimed to protect in

the first place, yet the unintended consequences have resulted in exclusion from the formal sector for many.

However, our exposition of the costs associated with these rigidities needs to be balanced by an analysis of the

benefits these instruments provide before policy recommendations can be formulated. Future research should

include a comprehensive approach to address the complex social and economic challenges that informality

represents across the developing world.

Comment

Adriana Kugler: This paper examines the rise in informal sector employment in Colombia from 1996 to the early

2000s and the subsequent decrease in 2004. It is particularly interesting to study the case of Colombia because

informality in Colombia is characterized by exclusion as opposed to voluntary participation in the sector.1

This paper focuses on the potential role of nonwage labor costs (or payroll taxes) and minimum wages in

explaining informality in Colombia. The idea is that labor regulations make it costly for employers to hire in the

formal sector and induces employers to hire in the informal sector, where they are not subject to these costs. For

this story to be valid, however, both nonwage labor costs and minimum wages would have to be binding. Payroll

taxes are binding only as long as the worker does not pay for these in the form of lower wages. Kugler and Kugler

find evidence that only about a fifth of the cost of payroll taxes is passed on to workers and that this shift is

greater for nonproduction or skilled workers than for production or unskilled workers.2 Thus, while payroll taxes do

seem to impose a substantial cost on employers, this cost is greater when hiring unskilled workers than when

hiring skilled workers. As I explain below, this is at odds with some of the findings in the paper. In the case of

minimum wages, Maloney and Nunez find that minimum wages are high enough in Colombia to be binding.3

Moreover, for these factors to explain the rise in informality in Colombia, payroll taxes and minimum wages would

have to have been increasing during the rise in informality and decreasing starting in 2004. Nonwage labor costs

did, in fact, increase sharply between 1994 and 1996, but they remained high thereafter. Minimum wages, in turn,

increased sharply in 1998 and fell somewhat after 2004. Thus, while nonwage labor costs and minimum wages

somewhat follow the pattern of informality, the business cycle could very well explain the rise in informality in the

late 1990s and the subsequent fall in 2004. The empirical analysis does not do a satisfactory job of controlling for

business cycle effects.

Before examining the impact of nonwage labor costs and minimum wages on informality, the paper assesses

various definitions of informality, including the standard International Labor Organization (ILO) definition based on

firm size and occupation and a definition based on coverage. The coverage definition used in the paper defines

workers as formal if they make pension and health insurance contributions. The problem with this definition is that

the focus would have to be on whether the employer and not the worker pays for pension and health insurance, as

it is the added costs to the employer that induce firms to hire in the informal sector. Which definition is used for

the analysis matters because the coverage definition appears to be countercyclical, while the firm size and

occupation definition show a sustained increase in informality starting in 1996 (see figure 1).

The paper then reports regressions of these different definitions on individual demographic characteristics,

household income, a ratio of the minimum wage to the median wage in the city, and nonwage costs as a fraction

of the individual's salary. The analysis thus exploits variation in the minimum wage across cities and over time. In

addition, the nonwage labor costs vary at the individual level. The problem is that whether or not the person is

informal determines their wage, so the causality may run from informality to the ratio of nonwage costs to salary.

This complicates the interpretation of the estimates on the nonwage cost variable. It would have been much better

to exploit the legislated changes in pension and health contributions for different types of workers in this context.

The results show that higher minimum wages are positively associated with the firm size definition of informality

and that nonwage labor costs are positively associated with both definitions of informality. The results further

show that minimum wages are more strongly associated with informality for less-skilled workers than for skilled

workers, as one would expect. However, the analysis indicates a stronger association of nonwage labor costs with

informality for highly educated employees, which is at odds with the previous evidence on the impact of payroll

taxes. If, as Kugler and Kugler report, a higher fraction of payroll taxes can be passed on as lower wages for skilled

than for unskilled workers, then these costs should be more onerous for firms hiring unskilled workers, thereby

inducing greater informality for this group.4 This result may instead suggest reverse causality if formal activity is

more strongly associated with salaries for highly skilled workers.

The most interesting part of the paper explores transitions from formality to informality and from informality to

formality using a similar specification to the one used for the probability of being informal, but including an

indicator of whether the person changed sectors and the unemployment spell. The paper does not clearly explain

what data are used to follow these transitions, as panel data were only available for Colombia from 2000 to 2006.

It seems instead that the paper is using information on the past job for those who changed jobs. In this case, there

would be a selected sample, as only those who changed jobs during the past year would be included in the

analysis. This could generate positive biases if, for example, those who change jobs are more likely to be

discontent with their working conditions and to move toward jobs with better conditions.

The main results in terms of transitions show that the minimum wage increases transitions from formality to

informality, but not the other way around, and that minimum wages have a greater impact on less-skilled workers.

However, as with the probability of being informal, the analysis on transitions finds the odd result that nonwage

labor costs increase transitions from formal to informal and that the effect is greater for skilled workers. This is at

odds not only with previous analyses of payroll taxes in Colombia, but also with the description in the introduction

and conclusion, which states that "all in all, our results suggest that labor market rigidities affect low-skilled

workers the most."

As mentioned above, the patterns of informality are highly countercyclical, but so is the transition from the formal

to the informal sector. It seems particularly important in this context to control for business cycle measures. The

analysis claims to proxy for the business cycle with the household income variable, but per capita household

income is not a good measure of the business cycle since there are many permanent and transitory factors that

affect household income besides macroeconomic factors. In fact, the descriptive analysis at the beginning of the

paper shows a strong positive correlation between unemployment and informality. It is not clear, therefore,

whether what is being captured by the minimum wage measure is simply an effect of the business cycle on

informality. The regressions should have controlled for gross domestic product or gross state product, as well as

other contemporaneous changes in regulations, given the other structural reforms occurring in Colombia during

this period. These included a major labor market reform that reduced dismissal costs in Colombia in 1991, a major

trade reform that took place in 1991, and capital market deregulation in 1992.s While some of these changes, such

as the reduction in dismissal costs, may have contributed to reducing informality, others, such as trade opening,

may induce firms to move toward informality.6

Footnote

1 . Payroll taxes refer to employer contributions to finance public social services such as job training and

childcare.

Footnote

2. Núñez (2002).

3. Sánchez, Duque, and Ruiz (2009).

4. Kugler and Kugler (2009).

Footnote

5. Santa María, Garcia, and Mujica (2009).

6. Flórez (2002).

7. Perry and others (2007).

8. See, for example, Cunningham and Maloney (2001) for the case of Mexico.

9. Bernal (2009).

Footnote

10. See, for example, Flórez (2002).

Footnote

11. Informality behaves very differently when measured by health access instead of health contribution, mainly

because of the expansion of the subsidized regime that provides free health insurance to the poor. If an individual

has health insurance, the spouse, children, and parents are also covered. Therefore, covered spouses have no

incentive to contribute, as law-abiding families are double-taxed for health insurance. In this paper, we consider

workers who are covered by spousal or family insurance or who work but are covered by the subsidized regime as

informal.

12. The time series for each definition is presented according to data availability.

Footnote

13. Henley, Arabsheibani, and Carneiro (2006) present a similar analysis for the Brazilian case.

Footnote

14. Workers who are classified as informal under Firm Size and Occupation but as formal under Health can be

grouped in two categories: those who are relatively well educated, older, and wealthier and those who are unpaid

family aids, business owners, or household workers. Workers who are considered informal under Health but formal

under Firm Size and Occupation are relatively younger, less educated, and wealthier (see table Al in the appendix

for details).

Footnote

15. We get very similar results when we estimate the correlations with gross domestic product (GDP) growth.

16. There are two caveats to the analysis of returns to the two sectors. First, as discussed in the literature, some

workers voluntarily transit into the informal sector, while others simply cannot get a job in the formal sector. The

former effect implies that the selection of agents into sectors is not random. There is currently no good way of

adjusting for selection into the formal or informal sector given data availability, so the results may be biased.

Second, given the data structure, there is no way to disentangle the returns to the labor and capital components of

selfemployment and business ownership. Therefore, the returns reported by these categories may overstate

earnings as a return for their work.

17. Workers with less than eleven years of schooling are classified as low skilled, while workers with eleven or

more completed years of schooling are classified as high skilled. On average, 8.7 percent of the Colombian

workforce has completed tertiary education. Table A2 in the appendix displays the sample sizes and distributions

across sectors of these two education groups. This classification splits the Colombian workforce roughly in half:

53 percent of workers are low skilled, and 47 percent high skilled. When we compare the composition of the formal

and informal sectors, roughly one out of every three informal workers is high skilled, whereas two out of every

three formal workers are high skilled.

Footnote

18. Bernal and others (2009).

19. Gruber (1997).

20. Kugler and Kugler (2009).

21. Albrecht, Navarro, and Vroman (2009).

Footnote

22. Fortin, Marceau, and Savard (1997).

23. Kristensen and Cunningham (2006).

24. Card and Krueger (1995).

25. In the case of wage-earners the worker and employer share the burden (4 percent and 8 percent, respectively)

and the self-employed have to contribute the full amount. Nearly 10 percentage points of the contribution finance

the worker's insurance and the remainder contributes to the health access of the poor and unemployed in the

subsidized regime.

26. Flórez (2002); Santa María, Garcia, and Mujica (2009). Contributions for workers earning more than four times

the minimum wage rose from 8 percent to 14.5 percent.

27. Bernal and others (2009).

Footnote

28. One of the proclaimed recent successes of Colombian public policy is precisely the increase in coverage of the

subsidized regime.

29. Maloney and Nunez (2004). Since changes in the minimum wage should take into account changes in both

purchasing power and labor productivity, a good way to characterize its evolution over time is to use the ratio

between the minimum wage and the median wage as a proxy for productivity. This measure trivially controls for

inflation.

30. Maloney and Núñez (2004).

31. Payroll taxes finance public social services, which are paid solely by the employer and are locally referred to as

parafiscales. They are contributions to the following three organizations: the Instituto Colombiano de Bienestar

Familiar (ICBF), which aims to protect children and improve the welfare of Colombian families; the Servicio

Nacional de Aprendizaje (SENA), which invests in the social and technical development of Colombian workers

through training programs; and Cajas de Compensación Familiar (CCF), whose main function is to redistribute a

portion of the payroll of the covered work force through monetary subsidies and in-kind services to workers with

the lowest wages.

Footnote

32. We compared different ways to include the transportation subsidy, and the results were robust to the different

specifications tested. In particular, if the subsidy is included as the fixed amount mandated by law, it can become

an unrealistically high percentage of the salary for those workers earning less than the minimum wage. As a

robustness check, we repeated the estimations in the paper without the transportation subsidy in the nonwage

costs measure, and the main results did not change.

33. For a detailed description of nonwage costs by year and income, see table A3 in the appendix.

34. Cárdenas and Bernal (2003).

35. Increases in the minimum wage for the coming year in Colombia are negotiated by the end of the current year,

based on the projected inflation rate plus calculated increases in productivity. In December 1 998, the expected

inflation for 1999 was in the range of 15-17 percent, and the negotiated increase in the minimum wage was 16

percent. Over the course of the year, however, the country fell into the deepest recession since die 1930s, and the

realized inflation rate for 1999 was 9 percent. Consequently, the minimum wage increased 7 percent in real terms.

The following year, the Constitutional Court ruled that minimum wage increases could never be below the observed

inflation rate of the previous year (Constitutional Court Sentence N0 815/99, 20 October 1999).

Footnote

36. See, for instance, Bernal and others (2009).

Footnote

37. We performed a robustness exercise in which we used the real minimum wage by city, adjusting for differences

in the evolution of inflation levels across cities. The results are qualitatively and quantitatively very similar. The

estimated effects are higher when we eliminate the regional variation in MinW by using the national minimum-

median ratio.

38. As discussed earlier, the main findings of the paper are robust to the inclusion or exclusion of

transportation/commuting subsidies within nonwage costs.

Footnote

39. We use real household income per capita instead of regional GDP in order to generate variation across

individuals. Our main findings are robust to changes in the business cycle proxy, for example by using real total

household income, or to the exclusion of this variable from the estimation. When we use regional GDP, business

cycle effects are in general not statistically significant. A better proxy for business cycle would be the growth of

real household income per capita. However, because we are working with repeated cross-sections, it is impossible

to build such a measure.

40. Kugler and Kugler (2009).

Footnote

41. To the best of our knowledge, there are no other estimations of the effect of the minimum wage on the size of

the informal sector.

42. Regression estimates are only shown for the variables of interest; detailed and complete results for this

estimation are available on request. The estimated coefficients for the control variables indicate that informality

rates tend to be higher for women, young people, unmarried workers (though the effects are relatively small), and

workers with low education. There is also significant variation across regions and economic sectors. Gender and

marital status differences are not significant when using the Health definition of informality in the total sample.

Footnote

43. See Perry and others (2007). This is consistent with the findings of Ardagna and Lusardi (2010) and

Mondragón-Vélez and Peña (2010).

44. We used a chi-squared test to calculate the significance level and the delta method for the standard errors.

Footnote

45. When we include workers who had been on the same job for over a year (by defining their previous sector to be

the same as their current one), the magnitude of the coefficients of interest (MinW and NWC) falls dramatically,

although the sign and significance of the probability of transitioning into informality do not change. For the

estimation of the transition into formality, the significance of the effects changes, as well.

46. The remaining covariates have the expected sign and significance levels. The full set of results are available on

request.

Footnote

47. The probability of transition into informality is expected to increase with higher nonwage costs or MinW as

firms decrease the supply of formal jobs in response to larger hiring costs; decrease with the business cycle as the

supply of formal job opportunities is pro-cyclical; and increase with the unemployment spell as the odds of getting

a job in the formal sector are hurt by a long unemployment spell.

Footnote

48. Again, the significance level was calculated using a Chi- test; the standard errors were computed using the

delta method.

Footnote

1. Perry and others (2007). See, for example, Cunningham and Maloney (2001) for an analysis of voluntary

participation in the informal sector in Mexico.

2. Kugler and Kugler (2009).

3. Maloney and Núñez (2004).

Footnote

4. Kugler and Kugler (2009).

5. See Kugler (1999, 2004) on the labor market reform and Eslava and others (2004, 2010) on the capital market

deregulation.

6. See, for example, Goldberg and Pavcnik (2003); Eslava and others (2009, 2010).

References

References

Albrecht, James, Lucas Navarro, and Susan Vroman. 2009. "The Effects of Labor Market Policies in an Economy

with an Informal Sector." Economic Journal 19(539): 1105-29.

Ardagna, Silvia, and Annamaria Lusardi. 2010. "Explaining International Differences in Entrepreneurship: The Role

of Individual Characteristics and Regulatory Constraints." In International Differences in Entrepreneurship, edited

by Joshua Lerner and Antoinette Schoar, pp. 17-62. University of Chicago Press and National Bureau of Economic

Research.

Bernal, Raquel. 2009. "The Informal Labor Market in Colombia: Identification and Characterization." Desarrollo y

Sociedad63 (March): 145-208.

Bernal, Raquel, and others. 2009. "Desarrollo económico: retos y políticas públicas." Working Paper 5269. Bogotá:

Universidad de los Andes, CEDE.

Card, David, and Alan B. Krueger. 1995. Myth and Measurement: The New Economics of the Minimum Wage.

Princeton University Press.

Cárdenas, Mauricio, and Raquel Bernal. 2003. "Determinants of Labor Demand in Colombia: 1976-1996." Working

Paper 10077. Cambridge, Mass.: National Bureau of Economic Research.

Cunningham, Wendy V., and William F. Maloney. 2001. "Heterogeneity among Mexico's Microenterprises: An

Application of Factor and Cluster Analysis." Economic Development and Cultural Change 50(1): 131-56.

Eslava, Marcela, and others. 2004. "The Effect of Structural Reforms on Productivity and Profitability Enhancing

Reallocation: Evidence from Colombia." Journal of Development Economics 75(2): 333-71.

_____. 2009. "Trade Reforms and Market Selection: Evidence from Manufacturing Firms in Colombia." Working

Paper 14935. Cambridge, Mass.: National Bureau of Economic Research.

___ .2010. "Factor Adjustments after Deregulation: Panel Evidence from Colombian Plants." Review of Economics

andStatistics 92(2): 378-91.

Flórez, Carmen Elisa. 2002. "The Function of the Urban Informal Sector in Employment: Evidence from Colombia

1984-2000." Working Paper 6883. Bogotá: Universidad de los Andes, CEDE.

Fortin, Bernard, Nicolas Marceau, and Luc Savard. 1997. "Taxation, Wage Controls, and the Informal Sector."

Journal of Public Economies 66(2): 293-312.

Goldberg, Pinelopi and Nina Pavcnik. 2003. "The Response of the Informal Sector to Trade Liberalization." Journal

of Development Economics 72(2): 463-96.

Gruber, Jonathan. 1997. "The Incidence of Payroll Taxation: Evidence for Chile." Journal of Labor Economics 15(3,

part 2): S72-S101.

Henley, Andrew, G. Reza Arabsheibani, and Francisco G. Carneiro. 2006. "On Defining and Measuring the Informal

Sector." Discussion Paper 2473. Bonn: Institute for the Study of Labor (IZA).

Kristensen, Nicolai, and Wendy V. Cunningham. 2006. "Do Minimum Wages in Latin America and the Caribbean

Matter? Evidence from 19 countries." Policy Research Working Paper 3870. Washington: World Bank.

Kugler, Adriana. 1999. "The Impact of Firing Costs on Turnover and Unemployment: Evidence from the Colombian

Labor Market Reform." International Tax and Public Finance Journal 6(3): 389-410.

______. 2004. "The Effect of Job Security Regulations on Labor Market Flexibility: Evidence from the Colombian

Labor Market Reform." In Law and Employ ment: Lessons from Latin America and the Caribbean, edited by James

J. Heckman and Carmen Pages. University of Chicago Press.

Kugler, Adriana, and Maurice Kugler. 2009. "The Labor Market Effects of Payroll Taxes in Developing Countries;

Evidence from Colombia." Economic Development and Cultural Change 57(2): 335-58.

Maloney, William F., and Jairo A. Núñez. 2004. "Measuring the Impact of Minimum Wages: Evidence from Latin

America." In Law and Employment: Lessons from Latin America and the Caribbean, edited by James J. Heckman

and Carmen Pages, pp. 109-30. University of Chicago Press and National Bureau of Economic Research.

Mondragón-Vélez, Camilo, and Ximena Peña. 2010. "Business Ownership and Self-Employment in Developing

Economies: The Colombian Case." In International Differences in Entrepreneurship , edited by Joshua Lerner and

Antoinette Schoar, pp. 89-127. University of Chicago Press and National Bureau of Economic Research.

Núñez, Jairo A. 2002. "Empleo informal y evasión fiscal en Colombia." Archivos de Economía Paper 210. Bogotá:

Departamento Nacional de Planeación.

Perry, Guillermo E., and others. 2007. Informality : Exit andExclusion. Washington: World Bank.

Sánchez, Fabio, Valentina Duque, and Mauricio Ruíz. 2009. "Costos laborales y no laborales y su impacto sobre el

desempleo, la duración del desempleo y la informalidad en Colombia, 1980-2007." Working Paper 11. Bogotá:

Universidad de los Andes, CEDE.

Santa María, Mauricio, Fabián García, and Ana Virginia Mujica. 2009. "El mercado laboral y la reforma a la salud en

Colombia: incentivos, preferencias y algunas paradojas." Bogotá: Fedesarrollo.

AuthorAffiliation

Mondragón-Vélez is with the International Finance Corporation; Peña and Wills are with the Universidad de Los

Andes.

We would like to thank Santiago Saavedra for excellent research assistance in the early stages of this project, as

well as Alejandro Badel, Raquel Bernal, Alberto Carrasquilla, Adriana Kugler, William Maloney, and Mauricio Santa

Maria for useful suggestions. DETAILS

Subject: Studies; Regression analysis; Labor market; Minimum wage; Economic history;

Economic reform

Location: Colombia

Classification: 9173: Latin America; 9130: Experiment/theoretical treatment; 6400: Employee

benefits &compensation; 1120: Economic policy &planning

Publication title: Economia; Washington

Volume: 11

Issue: 1

LINKS Check for full text in other resources

Database copyright  2018 ProQuest LLC. All rights reserved. Terms and Conditions Contact ProQuest

Pages: 65-101

Number of pages: 37

Publication year: 2010

Publication date: Fall 2010

Publisher: Brookings Institution Press

Place of publication: Washington

Country of publication: United States, Washington

Publication subject: Business And Economics--Domestic Commerce

ISSN: 15297470

Source type: Scholarly Journals

Language of publication: English

Document type: Feature

Document feature: Graphs Diagrams Tables References

ProQuest document ID: 857256952

Document URL: https://search.proquest.com/docview/857256952?accountid=8289

Copyright: Copyright Brookings Institution Press Fall 2010

Last updated: 2015-10-24

Database: ProQuest Central