Child labour and the Economic recession of 1999 in Colombia

República de Colombia
Departamento Nacional de Planeación
Dirección de Estudios Económicos
ARCHIVOS DE ECONOMÍA
Child labour and the Economic recession of 1999 in
Colombia
(A dissertation submitted to the University of Bristol in accordance with the requirements of the
degree of MSc. Economics and Econometrics in the Faculty of Social Sciences.Department of
Economics)
Claudia Marcela UMAÑA APONTE
Documento 246
16 de Enero de 2004.
La serie ARCHIVOS DE ECONOMIA es un medio de la Dirección de Estudios Económicos, no es un órgano
oficial del Departamento Nacional de Planeación. Sus documentos son de carácter provisional, de
responsabilidad exclusiva de sus autores y sus contenidos no comprometen a la institución.
Child Labour and the Economic Recession of 1999 in Colombia
Claudia Marcela UMAÑA APONTE♣
Abstract
This paper examines the relationship between economic growth, households’
income, child labour and school attendance in Colombia. It also analyses the impact
of the economic recession of 1999 on child labour and education. An important
contribution of this paper is the use of micro and macro data in the estimation of the
empirical models, since they allow analysing micro and macro sources of child
labour. The results show that both households’ earnings and economic growth are
relevant for the families’ decision-making process regarding education or labour
activities of their children. I find that, in Colombia, child labour is contra-cyclical and
education is pro-cyclical to economic growth. Therefore, higher levels of social
income increase the children’s welfare providing them with more education and less
economic responsibilities. This implies that the main reason why children work in
Colombia is poverty. The late nineties’ economic crisis impact on child labour and
education was a sharp increase of children in the labour force and a slight decrease
of school attendance.
♣
A dissertation submitted to the University of Bristol in accordance with the requirements of the
degree of MSc. Economics and Econometrics in the Faculty of Social Sciences.
Department of Economics
September, 2003.
Author’s declaration
I declare that the work in this dissertation was carried out in accordance with the
Regulations of the University of Bristol. The work is original except where indicated
by special reference in the text and no part of the dissertation has been submitted for
any other degree.
Any views expressed in the dissertation are those of the author and in no way
represent those of the University of Bristol. The dissertation has not been presented
to any other university for examination either in the United Kingdom or overseas.
SIGNED:
DATE:
2
Table of Contents
ABSTRACT ........................................................................................................... 1
AUTHOR’S DECLARATION ................................................................................. 2
TABLE OF CONTENTS ........................................................................................ 3
1.
INTRODUCTION ............................................................................................ 5
2.
THE COLOMBIAN CONTEXT........................................................................ 7
A.
B.
GENERAL FEATURES ..................................................................................... 7
DESCRIPTIVE STATISTICS FOR CHILD LABOUR IN COLOMBIA ............................ 10
i.
Participation rates .................................................................................. 10
ii. Type of activity....................................................................................... 11
iii. School attendance and educational level .............................................. 12
3.
DATA DESCRIPTION .................................................................................. 13
4.
OBJECTIVES ............................................................................................... 15
5.
LITERATURE SURVEY ............................................................................... 15
6.
METHODOLOGY AND RESULTS ............................................................... 21
A.
B.
7.
GRAPHICAL EXAMINATION ............................................................................ 21
MODELS SPECIFICATION AND RESULTS .......................................................... 21
i.
Time effects ........................................................................................... 22
ii. Micro vs macro effects ........................................................................... 24
iii. Quadratic GDP ...................................................................................... 27
CONCLUSIONS ........................................................................................... 28
REFERENCES .................................................................................................... 29
List of tables
Table 1. Participation rates by gender.................................................................. 10
Table 2. Child labour by economic activity ........................................................... 11
Table 3. Chidren who work and attend school ..................................................... 12
Table 4. Years of education completed by working children ................................ 13
Table 5. Number of observations for 12-17 year-old children in NHS .................. 14
Table 6. 1993’s Colombian census ...................................................................... 14
Table 7. Graphical examination results ................................................................ 31
Table 8. Participation Rates Results .................................................................... 32
Table 9. Variables Description ............................................................................. 34
Table 10. Results Model 1. .................................................................................. 35
Table 11. Results Model 2 ................................................................................... 38
Table 12. Large differences in earnings due to education.................................... 39
Table 13. Results Model 3 ................................................................................... 40
Table 14. Results Model 4 ................................................................................... 41
3
List of graphs
Graph 1. Colombia’s Economic Growth ................................................................. 8
Graph 2. Urban unemployment rate....................................................................... 8
Graph 3. Unemployment Rate by gender............................................................... 8
Graph 4. Unemployment Rate by age.................................................................... 8
Graph 5. Participation Rate by gender ................................................................... 9
Graph 6. Participation Rate by age ........................................................................ 9
Graph 7. Employed children participation rate ..................................................... 11
Graph 8. Fitted Employed Children Participation Rate......................................... 32
Graph 9. Fitted Employed Children Participation Rate vs GDP .......................... 33
Graph 10. Fitted Employed Children Participation Rate vs. Quarters .................. 33
Graph 11. Probability of Child Labour vs. GDP percapita .................................... 37
Graph 12. Probability of School Attendance vs. GDP percapita .......................... 37
Graph 13. Collapsed fitted values vs. GDP percapita .......................................... 41
Graph 14. Collapsed fitted values vs. Quarters.................................................... 42
4
1. Introduction
After more than 90 years of stable and positive growth, Colombia’s economy sharply
dropped in late nineties collapsing to the most severe economic crisis in 1999. The
social effects of this crisis, perhaps, were most severe than the economic ones.
Great part of the population lost their jobs, their houses and their economic
opportunities, reducing their standards of living. The total unemployment rate rapidly
increased and poverty indicators showed how people were becoming poorer.
One of the most worrying effects of the crisis was the increasing proportion of young
people entering the labour market and, perhaps, dropping out of school. This study
aims to analyse the impact of this economic recession on the evolution of child labour
in Colombia. This topic is of the particular interest since the future of a country lies on
the child and young population.
The economic growth theory has demonstrated that one of the main ways to reach
high levels of development is the accumulation of human capital. This accumulation
is the result of a long way which entails providing children both a high quality
education and a good environment for their growing up process. Colombia is a
developing country which requires improving the standards of living and opportunities
for the people. Therefore, if we look after our children and provide them with the
conditions for a successful development, we can reach those objectives.
For the previous reasons, child labour is highly related to economic growth and
development. More educated and healthy children today will assure more productive,
imaginative, creative and ingenious adults tomorrow, who will contribute to boost the
economic growth and development. On the contrary, child labour can be a barrier for
this objective. Although, it provides more experience for a specific task, can reduce
the opportunity that a child exploits his skills and acquires new abilities. Even if
children combine work and school, the time spent working could reduce their
understanding and learning capacity in comparison with full-time students.
Moreover, Bhalotra and Tzannatos (2002) have found that child labour has micro and
macroeconomic adverse effects. On the one hand, it can be the source of a
households poverty trap: “less educated parents tend to send their children to work
and these children grow up to be less educated, who then send their children to
work. Since education is a strong predictor of income, this is a mechanism that
generates… a persistence of poverty from one generation to another” (p. 4). On the
other hand, they affirm that economic growth theories have demonstrated that “the
social return to human capital investment exceeds the private return” (p.5), thus, that
family’s persistence of poverty also has macroeconomic effects. Since they are not
investing enough in their children the social return to education is not sufficient to
accelerate the economic growth and countries remain underdeveloped.
5
These previous arguments are of particular interest in Colombia. A recent study of
United Nations and two of the main Colombian governmental institutions1 reveals
that Colombian households are diminishing their expenditures (investment) on their
children education as a consequence of unemployment and the economic crisis. In
addition, they have found that this situation is not only occurring in the poorest
households but also in the richest ones. Recently, the ICBF’s executive manager
declares that “a [Colombian] child, who regularly works, has 70% more probability of
never returning to school” (El Tiempo, 9 of July of 2003). For these reasons, both to
understand the reasons why children work and to analyse the consequences of
1999’s recession on child labour in Colombia, is fundamental to propose policies
which can contribute to alleviate it.
Child labour studies have mostly used either micro or macro data to corroborate their
hypothesis. Those which have only used macro data, in general, have analyse the
relation either income-child labour or trade-child labour in a cross country framework.
However, Bhalotra and Tzannatos (2002) indicate that aggregate evidence is
imprecise in identifying whether the changes in child labour were the result of
individual-household decisions, macroeconomic policies or structural social changes.
Other studies only utilize micro data to investigate the factors which determine child
labour and its relation with other variables, e.g. income shocks. Nevertheless, in this
case macroeconomic variables, such as GDP, inflation and governmental policies
can influence families’ decisions through expectations, uncertainty or risk aversion. In
fact, parent’s perception about the general economic behaviour can motivate or
prevent them to send their children to work. Such as Bhalotra and Tzannatos (2002)
observed “Poor households are likely to be particularly risk-averse and they may
employ [their] children ex ante to diversify their sources of income”. In order to control
all these previous effects, this study makes use of micro and macro data.
The macro data consist of the Colombian Gross Domestic Product (GDP) percapita,
quarterly from 1985 to 2000. One of the main questions of this paper is to identify if
child labour is pro-cyclical or anti-cyclical to GDP. The micro data comes from the
Colombian National Households Survey (NHS) which is conducted quarterly over the
same period of time. It allows identifying children’s, parent’s and households’
characteristics as well as their incomes and environment features.
The rest of this paper will attempt to investigate the causes and consequences of
child labour in Colombia and its behaviour before, during and after recession. The
next section will describe the policy context of 1999’s economic recession and the
main characteristics of urban child labour in Colombia. The third explains in detail the
database used for estimations. The main objectives of this paper are summarised in
the fourth section. The fifth will review the relevant studies related to the hypotheses
of this paper. Methodology and results are contained in the sixth section. Finally, the
conclusions are stated.
See United Nations, Colombian Economic Planning Ministry (DNP) and Colombian Family Welfare
Institute (ICBF) (2003).
1
6
2. The Colombian Context
a. General Features
Historically the Colombian economy showed a strong and steady trend of growth.
Nevertheless, from 1998 the economy began a downward trend, which in 1999 gave
rise to the worst recession in its economic history (graph 1), having had severe
effects on the whole population’s welfare. This situation was the result of some
factors such as a fiscal decentralisation, the economic liberalisation of 1993 and an
unfavourable external economic situation.
The decentralisation resulted in a disproportionate increase in government
expenditures and the economic liberalisation generated unbalanced problems
between agents’ debits and credits. According to Cardenas (2001), in early 1990 the
Colombian Government formulated a package of political an economic structural
reforms. These reforms were focused on establishing a new Constitution and
opening the economy to new markets, following a process of economic liberalisation.
Firstly, the constitutional change restructured the political order with the purpose of
creating a high decentralization level. In other words, Central government attempted
to assign more responsibilities to territories. The objective was transferring more
money to municipalities, which knew their own problems much better, in order to
provide all the public services (especially education and health) that people required.
This fiscal decentralization resulted in a disproportionate increase in government
expenditures.
Additionally, Cardenas (2001, p.2) states: “…This trend was reinforced by a
deliberately expansionary fiscal management throughout most of the decade. In fact,
the central government’s deficit rose to more than 8% of GDP in 1999 from nearly
zero in 1991”. As a result of this fiscal deficit, market interest rates dramatically rose
since financing necessities of the government were increasing the interest rates of
the internal public bonds.
Secondly, another series of legal changes in the beginning of the decade drastically
modified the economic regimes related to trade, foreign exchange and foreign
investment. The economic liberalisation of 1993 generated excessive households’
and companies’ expenditures, greater than their incomes, which turned into microdeficits that individuals could not sustain, mainly because of the high interest rates.
Finally, the external economic situation was very unfavourable for Colombia. Ecuador
and Venezuela, its main trade and business partners, were also facing a deep
recession. In 1999, their economies fell 8.0% and 7.5% respectively.
7
The 1999 crisis brought critical consequences not only on the economic context but
also on the social one. One of the most severe effects on the population welfare was
a dramatic increase of the unemployment rate, which reached the highest levels in its
history getting to the peak in 2000 (graph 2).
Graph 1
Colombia’s Economic Growth
1979 – 2002 (First Quarter)
Graph 2
Urban unemployment rate
8%
21
6%
19
4%
17
2%
15
0%
Source: Dane - NPD – Colombia
III
III
00 I
III
99 I
III
98 I
94 I
97 I
7
III
-8%
III
9
96 I
-6%
III
11
79 I
80 I
81 I
82 I
83 I
84 I
85 I
86 I
87 I
88 I
89 I
90 I
91 I
92 I
93 I
94 I
95 I
96 I
97 I
98 I
99 I
00 I
01 I
02 I
-4%
95 I
13
-2%
Source: Dane – NPD. Colombia
The increase in the unemployment rate dramatically affected the whole population
(Graph 3 and 4). Nevertheless, young people have always suffered more from this
fact than the rest of the people. As it can be seen in Graph 4, from 1991 the youngest
population’s unemployment rate has been not only the highest but also the one with
the highest growth, especially between 1999 and 2000. During this late period, the
unemployment rates of people older than 18 years were decreasing, while the one
for people between 12 and 17 years was still growing.
Graph 4
Unemployment Rate by age
Graph 3
Unemployment Rate by gender
(People over 12 years old)
25
23
21
Males
47
Female
42
12-17
30-39
18-29
40-70
37
19
17
%15
13
11
32
% 27
22
17
12
9
7
5
7
Source: Dane
91 I
III
92 I
III
93 I
III
94 I
III
95 I
III
96 I
III
97 I
III
98 I
III
99 I
III
00 I
III
91 I
III
92 I
III
93 I
III
94 I
III
95 I
III
96 I
III
97 I
III
98 I
III
99 I
III
00 I
III
2
Source: NHS (ENH) – Author’s calculations
8
Between 1999 and 2002, the Colombian Government attempted to explain the
causes of this quick increase in the unemployment rate. It affirmed that, there were
two main types of factors: structural changes in the Colombian labour market and
those produced by the current economic situation (DNP, 2002 p.23)
On the one hand, the structural changes were mainly originated from the economic
liberalisation of 1993. Since then, the productive sector, in order to increase its
competitiveness, substituted low skill labour demand for high skill labour and labour
for physical capital demand. For this reason, young people with low educational
levels faced a higher unemployment rate (graph 4). In addition, the women labour
supply had changed in the last two decades, increasing the participation of women in
the labour market. For instance, the Colombian Economic Planning Ministry (DNP,
2002) states, “in 1982 one out of two women entered the labour market, nowadays,
approximately two out of three women do” (p.23). This increase in the participation of
female population was, in part, because they have reached better educational levels
and got more opportunities within the labour market.
On the other hand, the economic recession caused an abrupt drop in households’
income, reducing their wealth and welfare. Consequently, young people and women,
who were not working or interested in getting a job before of the crisis, entered the
labour force in order to preserve or recuperate their standards of living. This implied
an increase in women’s and young people’s participation rates in the labour market
(graphs 5 and 6)
Graph 5
Participation Rate by gender
78
Men
77
Women
Graph 6
Participation Rate by age
57
19
18
55
17
75
%
74
53
%
51
16
73
49
72
47
71
45
Source: Dane
18-29
40-49
92.0
87.0
%15
82.0
%
14
77.0
13
72.0
12
11
67.0
96 I
II
III
IV
97 I
II
III
IV
98 I
II
III
IV
99 I
II
III
IV
00 I
II
III
IV
95 I
II
III
IV
96 I
II
III
IV
97 I
II
III
IV
98 I
II
III
IV
99 I
II
III
IV
00 I
II
III
76
12-17
30-39
Source: NHS (ENH) – Author’s calculations
Table 1 summarize the participation rates2 of men, women and children in Colombia.
As it can be seen, the women’s participation rate in the labour market sharply
increases by 13.3% points from 1985 to 1999, which implies in average a growth of
0.9% points per year. Nevertheless, between 1999 and 2000, growth rate doubled,
reaching 2% points in only one year. Boys’ and girl’s participation rates decrease
Participation rates are calculated as the sum of employed plus unemployed people divided by
population older than 12 years who are those that can work.
2
9
since 1985, but, as in the women case, in the last two years of the nineties they
increased by 1.8% points each, possibly indicating the effects of recession.
Table 1
Participation rates2 by gender
Male
Adult Male
Female
Adult Female
Boys (12-17)
Girls (12-17)
1985
1995
1999
2000
73.0
84.1
41.6
47.0
20.1
13.9
74.5
85.3
47.5
53.5
18.9
13.0
73.8
84.4
54.0
60.3
16.4
12.7
73.9
84.1
56.1
62.2
18.2
14.5
Source: NHS (ENH) – Author’s calculations
This paper focuses on studying the consequences of the economic recession on the
labour market for people younger than 17 years old, in order to investigate the
validity of those previous government explanations. Although in Colombia the
population older than 12 years is considered part of the labour force, the basic
education is completed after 11 years of schooling (9 of them compulsory) and the
official enrolment age is 6 years old in public education. This implies that a child must
finish the basic education at a minimum age of 17 years. For this reason, this study
considers as child labour that one which is performed by the population under 17
years old. The following section will describe the characteristics of urban child labour
in Colombia in some detail.
b. Descriptive statistics for child labour in Colombia
In order to analyse the causes and consequences of the economic deterioration on
the households’ wealth and children’s welfare, it is fundamental to understand the
main features of urban child labour in Colombia. Bhalotra and Tzannatos (2002, p.7)
have indicated the existence of some stylised facts for child labour. This section
attempts to follow some of these stylised facts in order to describe the evolution of
Colombian child labour market in urban areas for 1985, 1995, 1999 and 2000. These
years were chosen because they capture the consequences of both the 1993’s
economic liberalization and the recession of 1999, which can be compared with the
conditions faced in 1985, before these changes ocurred. Nevertheless, for the
empirical analysis that will be performed in the coming sections, the complete sample
period, 1985 – 2000, will be utilized.
i.
Participation rates
Bhalotra and Tzannatos (2002, p.8) affirm that the participation rates of children have
been declined everywhere since 1950. Graphically, in Colombia the participation rate
10
of employed children has considerably decreased since 19923 (Graph 7),
corroborating the trend identified by Bhalotra and Tzannatos.
Graph 7
Employed children participation rate4
85 I
86 I
87 I
88 I
89 I
90 I
91 I
92 I
93 I
94 I
95 I
96 I
97 I
98 I
99 I
00 I
17
16
15
14
13
%12
11
10
9
8
7
Source: NHS (ENH) – Author’s calculations
ii. Type of activity
Since the data used in this paper comes entirely from the seven principal urban
areas, the principal activity performed by children in this case is not agriculture, as
Bhalotra and Tzannatos (2000) have identified as a stylised fact5. Nevertheless,
some specific activities in which they are involved can be identified.
Urban child labour in Colombia is concentrated in three principal activities:
commerce, industry and services (table 2).
Table 2. Child labour by economic activity
Economic Sector
Agriculture, Forestry, Fishing and Hunting
Mining
Industry
Electricity, gas and water
Construction
Commerce, restaurants and hotels
Transport, storage and comunication
Financial sector
Personal services
No information available
TOTAL
1985
%
2,072
1.4
310
0.2
25,174 17.4
48
0.0
9,574
6.6
37,826 26.1
3,688
2.5
1,178
0.8
64,846 44.8
37
0.0
144,752 100.0
1995
%
1,417
0.7
371
0.2
37,455 19.8
515
0.3
19,257 10.2
58,073 30.7
6,729
3.6
3,438
1.8
61,635 32.6
228
0.1
189,118 100.0
1999
%
1,404
1.0
43
0.0
21,217 15.5
578
0.4
4,978
3.6
48,400 35.3
4,445
3.2
2,945
2.1
52,893 38.6
154
0.1
137,056 100.0
2000
%
1,095
0.8
292
0.2
23,474 16.2
34
0.0
6,660
4.6
52,514 36.1
4,211
2.9
2,717
1.9
54,233 37.3
49
0.0
145,277 100.0
Source: NHS (ENH) – Author’s calculations
The principal activity that children perform is personal services to households. In
1999 and 2000, 31% and 29.6% of them, respectively, were working in this activity.
There is not data available between 1950 and 1985
Calculated as the ratio of employed children to 12-17 years old population.
5 This fact is a result of the sample data and does not imply that no children are working in rural areas.
3
4
11
The second activity is retail trade; the percentage of children in this activity was
28.3% in 1999 and 28.7% in 2000.
There have not been large changes from 1985 to 2000 in the type of economic
activities in which children are involved. The most important features in this respect
have been that a less number of children are working in personal services but more
of them are working in commerce.
iii. School attendance and educational level
Work and school are not perfect substitutes for children. On the contrary, Bhalotra
and Tzannatos (2000) have found that many children combine work and school.
Colombia is not the exception of the rule. As it is illustrated in table 3, 37.7% of
employed children attended the school in 2000. The proportion of children who
combine both activities has increased by 10% points from 1985. Nevertheless,
nowadays a considerable proportion of working children do not attend the school in
Colombia remaining as unskilled labour force in the future. This represents a warning
for the country since, as it was stated above, low levels of human capital are the
source of a poverty vicious circle which delays the economic growth and
development.
Table 3. Chidren who work and attend school
year
School attendance
1985 Yes
No
Total
1995 Yes
No
Total
1999 Yes
No
Total
2000 Yes
No
Total
40,299
104,453
144,752
67,197
121,921
189,118
48,946
88,110
137,056
54,786
90,492
145,277
%
27.8
72.2
100.0
35.5
64.5
100.0
35.7
64.3
100.0
37.7
62.3
100.0
Source: NHS (ENH) – Author’s calculations
Although, these statistics show that there has been a high proportion of children not
attending the school in the last twenty years, it is of interest to know if some of them
have achieved a certain level of education. Table 4 shows the number of years that
employed children have completed. Primary and secondary education in Colombia
consist of 5 and 6 years respectively, therefore the basic education includes 11
years, although only 9 of them are compulsory6. In 2000, 11.4% of urban working
children completed the secondary school, and 28.8% achieved and passed the
compulsory basic education. In 1985, only 1.6% could finalize the secondary school
and 7.9% the basic education.
The “Education General Law” (Ley General de Educación) established as compulsory nine years of
education since 1995. Before that Law, only primary education (5 years) was obligatory.
6
12
Since the sample consists of 12-17 year-old children, some of them have not
reached the required age to complete the nine compulsory years of basic education,
but all of them must have achieved at least 6 years. In 1985, only 37.9% completed
those six years. This percentage increased in 2000 by 28% points, i.e. 66% of 12-17
children had completed at least 6 years of education. Although after 16 years there
was an important improvement on the level of education of the working children, it is
still worrying that 34% of these children have not achieved the level of education
corresponding to their age. It is of special interest that between 1999 and 2000, years
of recession, the proportion of children who completed the secondary school
decreased by 1% point.
Table 4. Years of education completed by working children
years
1985
%
1995
%
1999
%
2000
%
2.9
4,201
2.2
2,915
2.1
3,429
2.4
3,535
2.4
1,862
1.0
1,154
0.8
1,926
1.3
Primary Education
4,179
1
2
9,788
6.8
7,064
3.7
4,240
3.1
3,260
2.2
3
15,364
10.6
10,482
5.5
6,551
4.8
6,652
4.6
4
17,069
11.8
12,556
6.6
8,298
6.1
7,483
5.2
5
40,053
27.7
40,288
21.3
26,661
19.5
26,650
18.3
Univ. Secondary Education
0
6
19,990
13.8
27,417
14.5
14,594
10.6
18,716
12.9
7
14,720
10.2
25,710
13.6
16,912
12.3
17,795
12.2
8
8,607
5.9
18,949
10.0
17,264
12.6
17,469
12.0
9
6,483
4.5
14,552
7.7
11,512
8.4
14,754
10.2
10
2,684
1.9
9,453
5.0
9,700
7.1
10,590
7.3
11
2,259
1.6
15,464
8.2
16,115
11.8
15,946
11.0
12
22
0.0
518
0.3
604
0.4
520
0.4
13
51
0.0
244
0.2
76
0.1
n.a
552
0.3
294
0.2
13
0.0
189,118
100
137,056
100
145,277
100
Total
144,752
100
Source: NHS (ENH) – Author’s calculations
3. Data description
The relation economic growth – child labour in Colombia in this paper is examined
using micro and macro data. The micro data is obtained from the National Household
Survey or NHS (Encuesta Nacional de Hogares or ENH) of Colombia, performed by
DANE7. It consists of a cross sectional quarterly data from 1985 to 2000. During this
period, in average, the total survey consists of 18,677 observations for households
and 82,895 observations for individuals each quarter. The section regarding
population aged 12-17 years contains, in average, 8,260.7 observations for children
and 5,377.1 per household per quarter. The complete sample details between 1985
and 2000 are in table 5.
7
National Department of Statistics (Departamento Administrativo Nacional de Estadística)
13
Table 5
Number of observations for 12-17 year-old children in NHS
1985 – 2000
TOTAL Employed Unemployed
Total Children
Total Households
Children per household
Children per total households
528683
344135
1.5
1.5
63424
45352
1.4
0.2
23296
17975
1.3
0.1
Children's activity
Participating in
Students Houseworking Other activity
the labour force
86720
390411
27162
24390
63327
246982
18184
15642
1.4
1.6
1.5
1.6
0.3
1.1
0.1
0.1
Source: NHS (ENH) – Author’s calculations
The survey contains urban information from the seven principal Colombian cities:
Bogotá D.C.8, Barranquilla, Cali, Medellín, Bucaramanga, Manizales and Pasto.
Rural information is not included. In 1993, DANE performed the last population
census in Colombia which included 1,059 cities in total. The seven cities used in this
study corresponded to 43% of the urban census in 1993. Table 6 details the
population distribution in those cities in 1993. GDP data by city in Colombia is not
available, but Galvis (2001) found that the 80% of the economic activity in Colombia
is concentrated in the area between Bogotá, Cali, Medellín and Bucaramanga.
Nevertheless, because of its population and location in the ocean Atlantic coast,
Barranquilla is one of the main ports of Colombia and its economic activity is of vital
importance for the country.
Table 6
1993’s Colombian census
Population by seven principal cities
Cities
Bogotá D. C.
Barranquilla
Bucaramanga
Cali
Manizales
Medellín
Pasto
Total 7 cities
Total Colombia
Total
4,931,796
990,547
410,065
1,641,498
303,136
1,551,160
261,368
10,089,570
23,514,070
Males
2,334,338
469,069
192,245
775,326
142,434
714,382
121,095
4,748,889
11,211,708
Females
2,597,458
521,478
217,820
866,172
160,702
836,778
140,273
5,340,681
12,302,362
Total
21.0
4.2
1.7
7.0
1.3
6.6
1.1
42.9
100.0
Participation %
Males
20.8
4.2
1.7
6.9
1.3
6.4
1.1
42.4
100.0
Females
21.1
4.2
1.8
7.0
1.3
6.8
1.1
43.4
100.0
Source: DANE - 1993 Census
The survey is composed of approximately 150 variables which allow to identify the
characteristics of the individuals and households and the environment that they face.
The variables chosen in this paper comprise information about:
Household information: size and income.
Household head and spouse information: gender, age, years of education and
income
Children characteristics: gender, age, school attendance, number of years of
education achieved, type of activity which they perform, income per month
and occupation by economic sector.
Identification variables for city and strata. Strata variable is an urban
categorization according to the people's standard of living. It is a social and
8
Bogotá D.C is the Capital of Colombia
14
economic classification, and divides the cities by different zones or areas, i.e.
it is also a geographical organization. The lowest stratum (1) represents the
poorest people and the highest (6) the richest ones in each city.
Income variables were deflated with the quarterly consumer price index. The base
year is 1994, thus all incomes are in Colombian pesos of the fourth quarter of 1994.
One Sterling Pound corresponds to approximately 1,288.2 Colombian pesos in 1994
(december) and 4,574 Colombian pesos in August 2003.
Macro data involves the quarterly Colombian real gross domestic product (GDP)
percapita over the same period.
4. Objectives
This paper attempts to achieve two central objectives: examining the main reasons
why children work in Colombia and evaluating the impact of the 1999 economic
decline on labour and educational opportunities of children.
In order to accomplish those two general objectives it is important to set out some
specific areas of analysis. Those specific objectives will focus on:
The significance of micro and macro variables of income on explaining child
labour in Colombia. Are both of them relevant for the household’s decisionmaking process regarding education or working activities for their children?
The relation economic growth – child labour. Is child labour pro-cyclical or
contra-cyclical to GDP percapita?
The effects of children and parents’ characteristics on child labour. How age,
gender and educational level generate more or less probability of participating
in the labour force.
The coming sections will be concentrated on providing the framework to respond and
achieve those general and specific objectives.
5.
Literature survey
Child labour has been a topic of great concern and controversy around the world.
During the last ten years, an increasing number of researchers and institutions have
been interested in understanding the causes and consequences of this problem in
order to analyse and propose efficient policies which can alleviate it.
15
The scope of research on child labour has covered so many fields that can be ranged
from individuals’ decisions to governmental laws and policies and from countries’
internal structure to foreign trade or external relations between them.
This section attempts to summarise the main general findings regarding the causes
and consequences of child labour and to review the specific studies that are essential
to the objective of this paper.
The common belief has been that poverty is the main and only reason for children to
work. Bhalotra and Tzannatos (2002) have identified that there are more causes than
poverty or subsistence constraints why children work. According to them child labour
can be the result of supply and demand decisions.
On the one hand, either household’s features or market characteristics, which
household faces, can influence the supply decisions. Among the former, poverty
constraints and agency are the principal factors. Firstly, if the household faces very
low standards of living (or poverty), parents can be interested in sending their
children to work instead to school because the family needs that additional income.
Moreover, the opportunity cost of education is too high for this type of households.
Secondly, agency plays an important role if parents, relatives or adult people who
look after children are not altruistic enough. In that case, those adults value more
their benefits from child labour than the children’s benefits from performing activities
related to childhood such as education or recreation.
Among the latter, incentives and imperfections in markets can determine child labour
decisions. Parents can be interested in sending their children to work whether the
returns to education are too low, or those to experience are high enough. In this
case, the labour opportunities which households face motivate to perform economic
tasks rather than educational or recreational activities. Alternatively, credit market
imperfections can reinforce poverty constraints factors. When households cannot
easily have access to credit, are obligated to use their children’s labour, in order to
achieve better living conditions.
On the other hand, wage discrimination and technology influence the demand side
decisions. In some countries, imperfections in the labour market permit firms to be
more attracted to employ children than adults. Those cases arise, when children can
perform the same activity as adults but their salaries are lower, thus, reducing the
firm’s costs. This type of discrimination would not occur in a competitive labour
market, where the wages would make equal for both children and adults since their
marginal productivity is the same. Regarding technology issues, industries employ
children because they have qualities or attributes which make their work more
efficient, compared with adults, increasing their productivity. Nevertheless, it has
been historically proved that technological advances can eliminate those firm’s
incentives since physical capital can substitute those children’s attributes.
Some studies have investigated the relation between GDP and child labour as this
study attempts to do. Nevertheless, most of them have focused on cross-country
16
evidence and the impact of globalisation and trade on the basis of anti-liberalisation
arguments which claim that open markets stimulate child labour. For this reason,
these studies only make use of macro data. Krueger (1996) find a negative relation
between GDP percapita and children participation rates in several countries,
concluding that a country’s wealth can explain 80% of the worldwide cross-country
variation in child labour.
Dehejia and Gatti (2002) explore the relation between child labour and borrowing
constraints using aggregated data. Credit constrains are measured as the private
credit by deposit money banks to GDP for each country. Taking into consideration
that, in general, poverty is the main source of child labour; they include GDP
percapita in the model to control income effects. Their results showed that the impact
of GDP percapita, on children participation rates, is much larger than the one of the
credit constraints.
Becchetti and Tovato (2002) try to explain how macro economic variables (e.g GDP)
can affect a social indicator such as child labour and focus their study on the effects
of globalisation and trade openness. Using an intra-household bargaining model on a
panel of emerging countries9, they test the hypothesis that “unskilled workers or raw
material producers… have, by definition, low bargaining power and therefore do not
benefit of the increased demand unless they exit the low skill trap” (p. 7). The
dependent variable is the participation rate of children aged 10-14 years regressed
on: the GDP percapita, transfers (aid percapita), trade openness, parental education,
quality of child education, rural life and labour market conditions. The main finding in
this study is that globalisation has positive effects on child labour reducing its
incidence, but those benefits of globalisation “may be mitigated if poor skills…
prevent emerging countries from climbing up the value chain of tradable goods” (p.
21).
Cigno, Rosati and Guarcello (2002) evaluate the effect of economic liberalisation on
child labour using the trade ratio as a measure of trade openness. Trade ratio is the
sum of exports plus imports divided by GDP. In the estimations, they used controls
for income, using GDP percapita, health expenditure and the skill composition of
each country. Their main result is that globalisation increases child labour if, and only
if, raises the wage rate of uneducated relative to educated workers. Moreover, they
found that although there is a negative association between income and child labour,
for countries with similar levels of GDP percapita, there are very large variations on
their extent of child labour. Therefore, not only the GDP has an impact on child
labour but also income distributions and governmental policies on health and
education can have effects on child labour.
There have been some other studies concerned not only about a general relation
GDP-child labour, but also about the impact of income shocks on households and
children labour supply. These papers are relevant for this work since the Colombian
economic recession of 1999 represent a negative income shock for households.
9
Approximately 80 countries of Latin America, Asia and Africa were included
17
Nevertheless, those empirical works focused on the use of income variables at a
micro level without allowing for macro income shocks. Beegle, Dehejia and Gatti
(2003) and Ferraz (2003) are examples of these analysis.
Beegle, Dehejia and Gatti (2003) examine how transitory income shocks can
increase children labour supply if households have limit access to credit in Tanzania.
Using a panel data set (micro data), they relate the hours that children work in a
week to the value of crop loss due to different calamities10 (income shocks) and the
value of households’ durable goods11 not connected to child labour (access to credit),
controlling by individual, household and community characteristics, and fixed effects
(households, time and survey round). Their results reveal two important facts. First,
income shocks tend to increase the hours worked by children and second, the
magnitude of this impact is much less (a half) for households which own more
durable goods. Therefore, child labour is used as a financial source when
households face reductions in their incomes.
Ferraz (2003) utilizes household’s unemployment periods generated from economic
recessions in Brazil as an exogenous negative income shock to families. He bases
his hypothesis on the added worker effect12 to relate those income shocks to
increases on child labour. Through a model of intertemporal labour supply and a
panel database, he obtains two main findings. Firstly, in short periods of time (one
month), the hours worked by children do not change when the household faces an
income reduction. Thus, increases in child labour supply are only due to increases in
children wages. Nevertheless, for longer periods, e.g. one year, the effects of wages
lose significance, then increases in children’s labour supply only respond with
changes in household income. Hence, he suggests that “at least in the short term,
children are insured against shocks in household income” (p. 13) however “long
lasting crisis can have pervasive effects on children by inducing them to work longer
hours to cope with family shocks” (p.15).
I am aware of only two empirical works which use micro and macro data to
investigate the impact of macroeconomic variables on households (micro) decisions
or behaviour. Ray (2000) carries out a study comparing the sources and
characteristics of child labour in Peru and Pakistan. He uses micro data to prove the
validity of Basu and Van’s luxury and substitution hypotheses13 in those both
countries, finding that neither of them is valid in Pakistan and only the substitution
one holds in Peru. Macro data on infrastructure and goods prices is used as an
instrument to relate child labour to commodity markets and to control fixed
Insects, rodents and fire.
Radios, bicycles, fans, lamps and pots. These type of access can be used as financial source in case of
negative income shocks. It excludes cash holdings, business and land, which can be related to child
labour.
12 Labour supply response of wives (or other family members) to their husbands’ (household head) job
losses.
13 Ray defines those hypotheses as follows: “Luxury Hypothesis: a family will send the children to the
labour market only if the family’s income from non-child labour sources drops very low” and “Substitution
hypothesis: child labour and adult labour are substitutes” (p. 5)
10
11
18
community effects. Moreover, the macro information was useful for Ray to propose
policies which contribute to alleviate child labour. He concludes that improving social
infrastructure and living conditions can contribute to reduce the probability that a child
works and to increase school enrolments. Therefore, “community infrastructure
variables provide effective policy instruments in influencing child participation in
employment and in schooling” (Ray, p. 17)
Edmonds and Pavcnik (2002) develop an investigation on the impact of changes in
the rice price in Vietnam on economic activities of children. They are interested in
exploring the effects of economic globalisation on child labour based on arguments of
market integration opponents. They have affirmed that the free flow of commodities
from developing countries to developed ones increases the economic opportunities
of children to work and consequently stimulate child labour in those countries.
Rice is one of the most important commodities in Vietnam. Not only, Vietnamese
families base their diet on it, but also in rural areas, people produce and export rice at
high scale. Moreover, a high proportion of Vietnamese children work in the
agricultural sector. For this reason, rice prices have a high influence on both
households’ budget and labour market opportunities for children. Edmonds and
Pavcnik use a panel data of two rounds (1992-1993 and 1997-1998) of the Vietnam
Living Standars Survey (VLSS). That survey allows knowing the individual,
households and community characteristics; the individuals’ economic activity;
households’ expenditure and wages rates. On the other hand, 1993 and 1998 rice
prices regard the macro data and are available for each community in both rounds.
Therefore, the data allows for regional and intertemporal changes in both the relative
price of rice and child labour.
They use linear probability (logit) model to estimate the probability that a child works.
In a first stage, they regress a binary variable which equals one if a child j in a
commune i at time t works on the natural logarithm of the real price of rice, child’s
characteristics, season and year controls, and commune fixed effects. They found
that the increases on relative rice prices result in a significant reduction of child
labour. Nevertheless, they realised that the benefits (damages) of price increments
depends on the household’s contact to rice economic activities. Producers gain from
price rises but consumers lose because they have to pay more for rice. For this
reason, in a second stage they classify the households in three groups: (rural)
producers, (urban) consumers and those (mixed areas) who are not directly
producers but perform related activities to rice production.
Through this differentiation, the authors found that prices have a positive effect on
child labour only for producers and those children related to rice-agricultural
activities. Since the sample was not big enough for urban and mixed communities,
the analysis remains only for rural households. Then, they divide the sample in two
groups: households where all children work and those where no children work and
estimate logit models using them as dependent variables. They find that increases in
prices not only reduce the probability that a child works but also that all children of
19
the same household work, as well as increases the probability that no children of the
same household work.
In a third stage Edmonds and Pavcnik explore the possibility of spurious regression
between child labour and rice prices allowing for household fixed effects instead of
commune fixed effects, interacting region dummies with time and including
community’s accessibility. Those estimations did not show evidence of spurious
regression. Additionally, they take into account infrastructure improvements in the
model as a possibility of omitted heterogeneity which could produce spurious
regression. The possible infrastructure improvements were in: roads, electricity,
irrigation, schools, health or other. Those estimations generated similar results to
those previously obtained, confirming their robustness.
The fourth stage consisted of considering idiosyncratic shocks to either rice prices or
child labour which could be another source of a spurious relation between them. In
order to do that, the change (declining) in child labour (∆c) was regressed on the
change in rice price (∆p) using the instrumental variables model (2SLS). This
estimation was done including controls for region, commune accessibility and
improvements in either schooling or total infrastructure. The results were consistent
with those obtained previously: an increase in rice prices declines the incidence of
child labour in Vietnamese rural areas. Thus, no evidence of spurious regression was
found due to idiosyncratic shocks.
In the further stages, the authors analyse the mechanism through which rice prices
affect child labour. Firstly, they include differences in landholdings to capture the
impact of prices on households consistent with their heterogeneity: “land is an
important input into rice production, and households with greater production capacity
should benefit more from price increases” (p. 20) and subsequently, control for land
redistribution. The results showed that children who work in households, which own
greater areas of land, are the main beneficiaries of the rice prices increase.
Secondly, they examine the effect of wages. Since, rice is the main agricultural
commodity in Vietnam, its price increases must raise the marginal productivity of
labour and therefore the wages in agriculture. This increase enlarges the incentives
to work instead of studying for children, so the effect should be an increase in child
labour supply. Allowing for this effect, the author included wages variables in the
model. They found that child wages do encourage children to work but increments in
adult wages counteract this effect, i.e. adult wages are related to less child labour.
The total effect is consistent with a negative relation between rice prices and child
labour.
Finally, they consider children gender and age differences concluding that “rice price
increases are associated with the largest declines in child labour for the age and
gender groups that have the highest participation rates” (Edmonds and Pavcnik, p.
29).
20
6. Methodology and results
a. Graphical examination
A first step in this study is a graphical examination of a child labour–GDP relation.
This examination has two objectives, firstly to check if there were significant changes
on the level of children participation rates during the economic recession of 1999
and, secondly, to contrast GDP and child labour.
Initially, the participation rate of employed children14 was regressed on time
dummies, which identified each year of the sample between 1986 and 200015, and
city dummies, using OLS estimations. Table 7 contains the results of this estimation.
The estimated coefficients are significant since 1996 and show steady reductions in
the employed children participation with respect to 1985’s rates. All dummies for
cities are significant. Graph 8 shows the fitted values of this participation rate
obtained from the last regression. It is evident that there was a dramatic level change
on the children participation rate in late nineties which in part can be a consequence
of the economic behaviour of this period. Nevertheless, other factors, different form
economic ones, are also included in time dummies and could have had effects on
households’ decisions causing these changes.
The next step consisted on regressing, by OLS16, the participation rates of employed
children on the natural logarithm of total GDP percapita (lnGDPpc) and its square,
calculating, then, the fitted values and plotting them against the lnGDPpc. These
results are showed in graph 9. The relation GDP-child labour is strongly negative. As
the GDP percapita grows the participation rate of employed children decreases which
could reflect that poverty is the main reason why children work, because they need to
contribute to household’s income. Graph 10 present the evolution of the fitted values
on time (quarterly data from 1985 to 2000)17. It confirms the change in participation
rates levels found in the first results of this graphical examination (graph 8).
b. Models specification and results
The incidence of the GDP on child labour in Colombia is examined estimating the
probability that a child performs activities related to either labour market or education.
The estimations will be obtained using probit models since they allow estimating the
probability of occurrence of an event when the dependent variable is binary,
This is an aggregate (population) variable calculated as the ratio of employed children to 12-17 years
old population.
15 1985 was the dropped year.
16 Table 8 shows the estimated coefficients.
17 There are 64 quarters in the sample. Stata identifies quarterly dates with numbers beginning with
100 for the first quarter of 1985.
14
21
assuming that the error term is normally distributed. The general specification of a
probit model is:
y* = α ' X + u
y=1
y=0
if the event occurs
if not
Where:
y* → is the dependent variable
α → is the vector of parameters
X → is the matrix of independent variables
u → error term – normally distributed
The choice process of explanatory variables (X matrix) attempt to follow three main
objectives: first, the variables must identify the household and environment
characteristics that children face; second, not to be endogenous and third, capture
micro, macro and time effects. In total 46 variables were used but not all of them
entered in all of the estimations. Table 9 contains the description of these variables.
Three sets of alternative estimations will be done to analyse the incidence of GDP
percapita on child labour in Colombia. The first step will be to estimate a model with
time dummies which will identify the changes on probabilities of child labour and
schooling over the sample period. The second step consists of understanding the
general reasons why children work and analysing micro against macro effects.
Finally, the third step will allow for a quadratic relation of GDP-child labour.
i.
Time effects
Given that the subject of interest is the urban children labour force behaviour, under
the argument that more children enter to the labour market, during the recession
period, two models regarding this area were defined. The binary dependent variable
for each of them reflects whether the child is employed or attending school. The
complete specification of the model for a child i in stratum s in city c at quarter q
at year t is:
yiscqt * = α1Tt + α 2 ln( HHHY )iscqt + α 3 ln( HHHS )iscqt + β i ' X isctq + δ s + λc + u (1)
Where y* is either the probability that a child works or attends the school18, Tt are
time dummies which equal one for each year on the sample and zero otherwise, the
omitted year is 1985. Time variables are included to allow for economic growth and
crisis’ effect. The GDP percapita is not included in the model since is collinear to the
time dummy variables. ln(HHHY) and ln(HHHS) are the natural logarithm of
household head’s and spouse’s income respectively. X i describes the child’s,
household head’s and spouses’ characteristics19, and the household size20. The
y = 1 if the individual reports either being employed or attending the school, y = 0 otherwise.
Household head and his spouse are the child’s parents in 81.9% of the sample, but they can also be
relatives different from parents (13.2%).
18
19
22
individuals’ characteristics are: gender, age, parents’ education and parents’
education square. Children’s education is omitted because of is endogeneity. δ s
and λ c are dummies which control for unobservable environment characteristics
and identify strata and cities, i.e. stratum and city fixed effects. Finally, u is the error
term normally distributed.
When a child enters the labour market one of two events can occur: first, they
dropped out education or, second, began to combine school and work. If the first
event occurs, the effect of the GDP will be a reduction in school attendance, but if the
second one takes place, the latter will not be affected. It is of a big interest in this
study to examine these effects; one of the main concerns of the Colombian
government was the children’s dropping out of school as an economic crisis result.
For this reason, school attendance is also a binary dependent variable in the model.
Table 10 contains the estimated coefficients of this model. I plotted them with the
GDP percapita growth, in graph 11, in order to compare the evolution of child labour
and economic progress over the sample period. The graph shows a contra-cyclical
pattern between economic growth and employed children until 1995. From 1996, it
seems that there was a change in this pattern and a pro-cyclical behaviour arises.
This is of particular interest, since some Colombian economists have affirmed that
the first signs of recession appeared in 1996. DNP (2002) states that in 1996 the
Colombian economy initiated a negative trend that the government attempted to
prevent through short run polices in 1997-1998, which generated an economic
bubble. This bubble turned into the worst recession in 1999 because increased the
market interest rates, promoted a monetary attack and caused a crisis in the financial
sector.
The recession seems to generate a decrease in the probability that children work in
Colombia since 1996. During this period firms faced a profound reduction of their
profits which force them to reduce their costs. In order to do so, firms decide to lay
their employees off. This situation not only increased the unemployment rate but also
reduced the labour demand. For this reason, possibly, the firms’ demand for any
labour force had dropped so much that the possibility to get a job for a child 12-17
during the crisis was too low. On the contrary, the probability of school attendance
remained approximately constant from 1994, showing a slightly increase during the
recession period 1998-2000 (graph 12). This could have been the result of families’
decision to avoid children’s idleness. Since the economy was not allowing them to
work, it was better to send them to school rather than remain at home doing nothing.
Moreover, public21 basic education in Colombia is free, thus the cost of education is
low and the opportunity cost of education had decrease as a consequence of the few
economic opportunities for children.
20
21
Although household size can be endogenous, this variable was included in all of the estimations.
That one provided by the government
23
ii. Micro vs macro effects
The previous model is one of the alternatives to analyse the effects of GDP on child
labour and school attendance. Nevertheless, it has the disadvantage that time
dummies not only reflect the behaviour of the economy but also can include external
effects such as weather, social conducts, government policies and other factors. For
this reason, another way to evaluate the effects of GDP on child labour is including
the GDP percapita as an explanatory variable.
On the other hand, the Colombian government affirmed that one of the causes of
high child’s unemployment in late nineties was the reduction of households’ income
(DNP, 2002), which is a micro variable. Therefore, it is of the interest of this study to
include micro income effects, in addition to GDP per capita, as explanatory variables,
regardless of its endogeneity. This allows analysing whether the decisions of the
households are determined only as response of their private situation, or take into
account their perception of the whole economic situation. The people’s expectations,
uncertainty and risk aversion are very important in economics. Several times, agents
are influenced by their environment or government policies as much as their personal
situation.
The Colombian Households Survey gives several types of household’s income
variables: labour income, self-employment income, income from other sources, total
income, household head’s and spouse’s total income. I had to decide which could be
the less endogenous. Any classification of the household’s income (labour, self
employment and other sources) includes the child’s income because they mainly
work in commerce activities which sometimes are related to informal sector and
considered as self employment in the survey. In addition, children work in the
personal services sector as employees, therefore, both households’ labour and nonlabour income contains child’s income. For these reasons, their endogeneity is more
critical than taking into account only the parent’s income. These two variables were
included in the second set of estimations. The specification of this model is:
y iscqt * = α 1 ln(GDPpc) qt + α 2 ln( HHHY ) qt + α 3 ln( HHHS ) qt + β i ' X isctq + δ s + λc + u (2)
Where y* is either the probability that a child works or attends the school22,
ln(GDPpc) is the natural logarithm of GDP percapita, ln(HHHY) and ln(HHHS) are the
natural logarithm of household head’s and spouse’s income, respectively; and the
rest of the variables keep the previous description.
Table 11 provides the results of this model. Column 1 presents the estimates of
α 1 , α 2 and α 3 for employed children and column 2 for school attendance. α̂ 1 reveals
that child labour is contra-cyclical and school attendance pro-cyclical to economic
growth. This implies that increases in economic productivity reduces the probability of
children to work and stimulates their education. A 10 percent increase in GDP
22
y = 1 if the individual reports being employed or attending the school, y = 0 otherwise.
24
percapita is associated with a 0.57 percentage point decline in child labour and 0.79
percentage point increase in school attendance.
These results are crucial on explaining child labour. On the one hand, it could be
affirmed that poverty is the principal reason why children work. When economy
grows and people are better off, children do not need to work in order to contribute to
support household’s expenditures. On the other hand, economic growth encourages
children to study. As GDP increases, the probability to attend school is higher.
Those both previous results are highly related to the opportunity cost and returns of
education and experience. Since children mainly work because of poverty
constraints, the opportunity cost of education is too high at low levels of GDP and the
returns to work and experience overcome the future returns to education.
Nevertheless, since parents send their children to school when family does not need
that additional income, a signal that the returns to education are high23 in Colombia
arise. In fact, Blom and Hansen (2002) have studied the returns to education in
Colombia and found large differences in salaries by level of education. Table 12 has
been taken from Bloms’ and Hansen’s study and shows that wages for tertiary
education are approximately three times those of primary and secondary school.
Blom and Hansen have also found that the returns of tertiary and postgraduate
education have increased considerably from 1980 to 2000, while those to primary
and secondary school have remained constant over the same period. In late nineties,
the returns to university education were two times those in 1980 while the returns to
secondary school have only decreased in two decades by 0.8 percent points.
Consequently, when families have been able to overcome poverty constraints or
satisfied a certain level of subsistence needs24, those high levels of returns to
education generate incentives to send children to school instead of work.
The estimated coefficients of α 2 and α 3 confirm the families’ poverty constraints as
the main cause of child labour. Micro income effects are significant and reveal that
higher levels of household’s head income have a positive effect on child labour,
reducing it, and increase the educational opportunities of children. An increase of
10% in father’s earnings reduces the probability that a child work by 0.05% and
increases the probability of studying by 0.19%. Mother’s earnings are not significant
for child labour, however also encourage child’s education.
In conclusion, both macro an micro variables are relevant for the household’s making
choice process regarding education and working activities of their children and reflect
a negative relation income-child labour which arises from poverty constraints and a
positive relation school-income, due possibly to high returns to education.
Or the opportunity cost of education is low due to high future expected incomes.
Bhalotra (2003) have found evidence that the boy’s income in Pakistan is necessary to reach the
subsistence expenditures of the households. This implies that if a household cross this “subsistence
frontier” boy’s child labour is not necessary and they can perform activities related to childhood.
23
24
25
The model includes controls for children and parents characteristics. These variables
are also important for the objective of explaining the causes of child labour. As it can
be seen in table 11, the differences by gender indicate that boys tend to work and
attend school more than girls do. Children’s age increases their probability of
participating in the labour force and decreases the one of being student and
household size has negative effects on child labour and education.
With respect to parent’s characteristics, household head’s and spouse’s gender are
not significant for child’s educational and working choices. On the contrary, as it was
expected, parents’ age and education reduces the probability that children work and
increases their education.
Since I have included micro income variables in both previous models, it is relevant
to know if the macro income variable is significant for the incidence of child labour by
itself without allowing for micro effects. Moreover, the endogeneity of parents’ income
can be a source of spurious correlation between GDP percapita and child labour. For
this reason a third model will be estimated taking into account only macro effects.
The model specification is now:
y iscqt * = α 1 ln(GDPpc) qt + β i ' X isctq + δ s + λ c + u (3)
Where y* is either the probability that a child works or attends the school25,
ln(GDPpc) are the natural logarithm of GDP percapita, and the rest of the variables
keep the initial description.
Table 13 presents the estimates of equation (3). α̂ 1 is significant for both child labour
and school attendance, and the magnitude of the coefficients are not altered
considerably, specially for child labour. This effect is associated to households’
expectations, uncertainty and risk aversion. As it was indicated previously in this
paper, Bhalotra and Tzannatos (2002) observed that the poorer the household the
more risk aversion and, hence, more probability to send children to work in order to
diversify its sources of income. If the economy, as a whole, faces a negative income
shock, families are motivated to employ their children as a way to insure them
against the risk of a income loss and as a way to smooth their consumption26. This
implies that expectations and uncertainty on economic behaviour lead families to use
child labour because of their risk aversion.
The exclusion of micro effects slightly weakens the relation GDP-school attendance.
Nevertheless, the impact of GDP on school attendance is still high and significant,
suggesting that the economic growth promote higher educational levels possibly
through generating higher salaries for more educated people, i.e. increasing the
returns to education and decreasing its opportunity cost.
y = 1 if the individual reports being employed or attending the school, y = 0 otherwise.
Some studies have shown that this effect is deeper if families face credit constraints. See for example
Dehejia and Gatti (2002) and Beegle, Dehejia and Gatti (2003).
25
26
26
iii. Quadratic GDP
Finally, it is of interest to evaluate a quadratic relation child labour and GDP percapita
instead of a simple linear one. This could be done including in the models (2) and (3)
the square of ln(GDPpc). Nevertheless, this was not possible to perform for the probit
models. Due to the data has in average 8,260 observations (individuals) by quarter
and the GDP is available only quarterly, the GDP and its square generated
collinearity because of the lack of enough observations. For this reason, it was
necessary to collapse the data by quarter and use the linear regression technique
(OLS) to be able to do these estimations. The quarterly collapsed values for the
binary variables which identify employment and school attendance were regressed
on the lnGDPpc and its squared. The new model is given by the equation:
y iscqt * = α 1 ln(GDPpc) qt + α 2 [ln(GDPpc) qt ] 2 + C + u
(4)
Where y* is the mean27 of the binary variables stated above, i.e. either the
participation rate of employed children or the proportion of children attending the
school; ln(GDPpc) is the mean of the natural logarithm of GDP percapita;
[ln(GDPpc)]2 is the mean of the natural logarithm of GDP percapita squared; C is the
constant and u is the error term which contains the omitted variables.
The results are reported in table 14. The results for employed children and school
attendance confirm those obtained above. At low levels of GDP percapita more
children enter to work as part of the Colombian labour force. Nevertheless, as the
GDP grows their incentives to work are reduced. The turning point28 of this quadratic
∂y *
relation, calculated as the derivative
= 0 , is at a ln(GDPpc) value equal
∂ ln(GDPpc)
to 12.8. During the recession period (1999-2000) the ln(GDPpc) was in average
12.99 which is on the right hand side of the maximum, i.e. more children were
entering to the labour force as a result of poverty constraints.
In the same way, at low levels of GDP percapita, school attendance increases but
when the GDP is too high, the economic opportunities generated by this growth
create incentives for children to work instead of study, consequently, decreasing the
school attendance. The turning point, in this case, occurs at a ln(GDPpc) value of
13.06. Since, in this case, the value of the natural logarithm of GDP during recession
is less than the turning point, the economic reduction possibly caused a decline in
children’s school attendance.
Graph 13 plots the fitted values obtained from equation (4) against ln(GDPpc). It
confirms the negative relation GDP-child labour and the positive relation GDPeducation. The higher the economic growth the fewer children’s incentives to work
“Collapse” command in Stata calculates the mean of the binary variables which have values y = 1 if
the individual reports being employed or attending the school, y = 0 otherwise.
28 The turning point is either the maximum or minimum of the function, where the slope, i.e. the
marginal value, is equal to zero.
27
27
and the higher school attendance. Graph 14 also plots the fitted values of model 4 on
the sample period29. It shows how in the recession period 1998-2000 (quarters 152163) more children could have entered labour market. It is important to notice that
school attendance, during the same quarters, remains approximately constant,
possibly as a result of the children’s combination job-school.
7. Conclusions
This paper has examined the link between micro and macro income, child labour and
educational opportunities for children in Colombia and has analysed the impact on
the activities performed by children during the economic recession of 1999. It has
described the economic and political context which led to that crisis and the extent
and main features of child labour in Colombia.
An important contribution of this paper is the use of micro and macro data in the
estimations since they allow analysing micro and macro reasons of why children
work. I have found that both households’ earnings and economic growth influence
families’ decisions regarding child labour and education. The economic behaviour
generates expectations and uncertainty which through household’s risk aversion
affect positively or negatively the likelihood to send children to work or school.
Through empirical results, I have shown that child labour is contra-cyclical and
education is pro-cyclical to Colombian economic growth. Therefore, higher levels of
social income increase the children’s welfare providing them with more education
and less economic responsibilities. This implies that children in Colombia work,
mainly, as a result of poverty constrains. The late nineties’ economic crisis impact on
child labour and education was a sharp increase of children involved in the labour
force and a slight decrease of school attendance.
The new challenge for the government after 1999’s recession is to generate
incentives to families and children to be full time students. The principal strategy to
achieve this objective is alleviating people’s poverty constraints. In general, the
economic growth required to reduce poverty is very difficult to reach in the short run;
hence, it is necessary to apply other policies which can rapidly contribute to this
purpose. Some studies have demonstrated that limited access to credit and an
inequitable income distribution make stronger the effects of poverty constraints.
Dehejia and Gatti (2002) have proved that credit constrains are an important source
of child labour since families, which cannot borrow money, employ children as a way
to smooth their consumption. Cigno, Rosati and Guarcello (2002) found that a group
of countries with similar levels of GDP percapita show very large variations in child
labour. Therefore, not only the GDP percapita can affect children’s labour
There are 64 quarters in the sample. Stata identifies quarterly dates with numbers beginning with
100 for the first quarter of 1985.
29
28
participation but also income distributions and governmental policies on health and
education can influence household’s labour decision.
From this previous analysis, two main areas for further research arise regarding the
effects of households’ access to credit and income distribution in Colombia on child
labour. Another important area of further research is how the economic liberalization
of 1993 affected the incidence of child labour. This area requires additional
examination, especially because it could contribute to define other types of strategies
for the government such as educational policies. Some studies30 have found that
globalisation can diminish child labour, if and only if the country owns a high level of
human capital when the economic liberalisation occurs. Hence, if this was not the
Colombian case, educational policies must focus on generating higher levels of
human capital.
References
Bhalotra, S. (2003). “Is Child Work Necessary?”. University 1`of Bristol. Discussion
Paper 03/554.
Bhalotra, S. and Tzannatos, Z. (2002) “What Have We Learnt?”. The World Bank
Social Protection Discussion Paper, December 2002
Becchetti, L. and Tovato, G. (2002). “The Child Labour–GDP percapita Relationship:
is there something beyond it?”. Vergata University, CEIS. Departmental
Working Papers No. 171.
Beegle,K., Dehejia, R. and Gatti, R. (2003). “Child Labour, Income Shocks and
Access to Credit”. Mimeo.
Blom, A. and Hansen, T. (2002) “Economic Perspectives of Tertiary Education; The
Case of Colombia”. World Bank, Department of Human Development, LCSHD
Paper Series. July.
Cardenas, M. (2001). “Economic growth in Colombia: A reversal of "Fortune"?”.
Harvard University, Centre for International Development. Working paper # 83
Cigno, A., Rosati, F. and Guarcello, L. (2002). “Globalisation and Child Labour”. In
Harnessing Globalisation for Children: A report to UNICEF edited by Giovanni
Andrea Cornia. Chapter 6.
30
See for example: Becchetti, L. and Tovato, G. (2002) and Cigno, Rosati and Guarcello (2002)
29
Dehejia, R. and Gatti, R (2002). “Child Labor: The Role of Income Variability and
Access to Credit across Countries”. NBER Working Paper Series. Working
paper No. 9018. June.
DNP (2002). “Balance macroeconómico 1998-2002 y Perspectivas para 2003”.
Bogotá: National Planning Department. Documento Conpes No. 3181.
Edmonds, E. and Pavcnik, N. (2002). “Does Globalization Increase Child LAbour?
Evidence From Vietnam”. NBER Working Paper Series. Working paper No.
8760. February.
El Tiempo, 9 of July of 2003. “Familias Colombianas están invirtiendo menos dinero
en la educacion de sus hijos”. Newspaper Article.
Ferraz, C. (2003). “The impact of income Shocks on Child Labour: Evidence from
Brazil”. ARE, University of California – Berkeley. Mimeo.
Galvis, L. (2001). “La Topografía Economica de Colombia”. Banco de la República,
Cartagena. Documentos de trabajo sobre economía Regional No. 22.
Krueger, A. (1996). “Observations On International Labor Standars and Trade”.
NBER Working Paper Series. Working paper No. 5632. June
Ray, R. (2000). “Analysis of Child Labour in Peru and Pakistan: A Comparative
Study”. Journal of Population Economics, 13, 3 – 19.
United Nations, DNP and ICBF (2003). “Familias Colombianas: estrategias frente al
riesgo”. Bogotá.
30
Table 7. Graphical examination results
(1)
ECPR 1
T86
-0.002
(0.44)
T87
0.008
(1.53)
T88
0.004
(0.91)
T89
0.007
(1.51)
T90
0.001
(0.15)
T91
-0.001
(0.25)
T92
0.014
(2.83)**
T93
0.002
(0.43)
T94
-0.005
(1.08)
T95
-0.007
(1.43)
T96
-0.016
(3.16)**
T97
-0.020
(3.97)**
T98
-0.024
(4.79)**
T99
-0.035
(7.21)**
T00
-0.028
(5.72)**
C1
-0.062
(18.94)**
C2
0.038
(11.57)**
C3
-0.038
(11.79)**
C4
-0.048
(14.76)**
C5
-0.058
(17.93)**
C6
-0.019
(5.69)**
Constant
0.159
(38.85)**
Observations
448
R-squared
0.80
Absolute value of t statistics in parentheses
* significant at 5%; ** significant at 1%
1. ECPR: Employed children participation rate
31
Graph 8. Fitted Employed Children Participation Rate
Employed children
Fitted values
.145823
.096382
1985
1987
1989
1991
1993
year
1995
Table 8. Participation Rates Results
(1)
ECPR
lnGDPpc
9.550
(1.22)
lnGDPpc2
-0.373
(1.23)
Constant
-61.005
(1.20)
Observations
448
R-squared
0.04
Absolute value of t statistics in parentheses
* significant at 5%; ** significant at 1%
32
1997
1999
Graph 9. Fitted Employed Children Participation Rate vs GDP
Employed children
Fitted values
.135645
.108159
12.7907
lnGDPpc
13.0753
Graph 10. Fitted Employed Children Participation Rate vs. Quarters
Employed children
Fitted values
.135645
.108159
100 104 108 112 116 120 124 128 132 136 140 144 148 152 156 160163
qdate
33
Table 9. Variables Description
Variable
name
year
quarter
deflact
ide004
ide010
HHsize
HHHmale
HHHage
HHHeduc
HHSfemale
HHSage
HHSeduc
edu001
educ
ftr001
boy
age
HHHeduc2
HHHY
lnHHHY
HHSeduc2
HHSY
lnHHSY
empch
STch
HWch
GDPpc
lnGDPpc
lnGDPpc2
C1
C2
C3
C4
C5
C6
C7
S1
S2
S3
S4
S5
S6
Description
Obs
528683
528683
528683
528683
528683
528676
528676
528650
527218
401393
401388
400731
528676
528373
528671
528683
528683
527218
528683
490136
400731
528683
297699
528683
528683
528683
528683
528683
528683
528683
528683
528683
528683
528683
528683
528683
528683
528683
528683
528683
528683
528683
stratum
city
household size
household head's gender
household head's age
years of education of household head
spouse's gender
spouse's age
years of education of spouse
school attendance
child's years of education achieved
child's activity
boy
age
years of education of household head^2
household head's total income
ln (household head's total income)
years of education of spouse^2
spouse's total income
ln (spouse's total income)
dummy variable for employed children
dummy variable for students
dummy variable for houseworking children
GDP percapita
ln(GDP percapita)
ln(GDP percapita) squared
dummy variable for city 1
dummy variable for city 2
dummy variable for city 3
dummy variable for city 4
dummy variable for city 5
dummy variable for city 6
dummy variable for city 7
dummy variable for stratum 1
dummy variable for stratum 2
dummy variable for stratum 3
dummy variable for stratum 4
dummy variable for stratum 5
dummy variable for stratum 6
34
Mean
1992.072
2.478932
0.8603962
2.822625
3.756213
5.955599
0.7788229
45.87326
6.824037
0.978739
40.86862
6.64293
0.8122517
6.511031
3.850718
0.485009
14.54764
65.16997
335066.3
12.34551
59.48941
65400000
14.87416
0.119966
0.7384595
0.0513767
419137.1
12.94265
167.5189
0.174057
0.1135558
0.2186774
0.0693856
0.2077937
0.1332122
0.0833184
0.0809029
0.2681872
0.4629182
0.1350601
0.0412364
0.0116951
Std. Dev.
Min
Max
4.604451
1985
2000
1.130251
1
4
0.7086073 0.1158517 2.370848
0.9826498
1
6
1.915854
1
7
2.277456
1
43
0.4150399
0
1
11.01939
10
98
4.313065
0
21
0.1442532
0
1
9.11643
12
98
3.919302
0
21
0.3905114
0
1
2.336885
0
18
1.531438
1
9
0.4997757
0
1
1.70852
12
17
74.53658
0
441
588866.6
0 27600000
0.9388687 -0.8632479 17.13322
64.22209
0
441
166000000
0 8.63E+08
3.692526 0.3328151 20.57612
0.3249221
0
1
0.439474
0
1
0.2207651
0
1
33877.51
358854 477015.8
0.081448
12.79067
13.0753
2.107149
163.6012 170.9636
0.3791589
0
1
0.3172712
0
1
0.4133496
0
1
0.254109
0
1
0.4057287
0
1
0.3398042
0
1
0.2763631
0
1
0.2726862
0
1
0.443016
0
1
0.4986235
0
1
0.3417881
0
1
0.1988368
0
1
0.1075097
0
1
Table 10. Results Model 1.
T86
T87
T88
T89
T90
T91
T92
T93
T94
T95
T96
T97
T98
T99
T00
boy
(1)
(2)
empch
school attendance
-0.001
0.002
(0.25)
(0.36)
0.006
-0.003
(1.41)
(0.55)
0.012
0.001
(2.60)**
(0.24)
0.013
0.001
(2.86)**
(0.26)
0.007
0.017
(1.50)
(3.59)**
0.008
0.013
(1.80)
(2.74)**
0.023
0.004
(5.14)**
(0.75)
0.013
0.003
(2.89)**
(0.74)
0.005
0.021
(1.08)
(4.60)**
0.005
0.019
(1.12)
(4.31)**
-0.014
0.021
(3.30)**
(4.59)**
-0.009
0.014
(2.19)*
(3.15)**
-0.015
0.019
(3.64)**
(4.12)**
-0.029
0.023
(7.14)**
(5.05)**
-0.022
0.027
(5.23)**
(5.89)**
0.022
0.032
(14.92)**
(19.23)**
age
0.047
-0.062
(97.77)**
(111.01)**
HHsize
0.005
-0.009
(11.18)**
(20.02)**
HHHmale
-0.022
0.013
(1.59)
(0.77)
HHHage
-0.001
0.002
(9.23)**
(12.08)**
-0.010
0.015
HHHeduc
(14.20)**
(20.16)**
HHHeduc2
0.000
-0.001
(12.22)**
(16.67)**
HHSfemale
-0.001
0.013
(0.07)
(0.81)
continuates
35
Table 10. Continuation
HHSage
HHSeduc2
lnHHHY
(1)
(2)
empch
school attendance
-0.003
0.003
(19.90)**
(18.88)**
(19.89)**
(29.65)**
0.001
-0.001
(17.82)**
(24.63)**
-0.008
0.020
(6.96)**
(16.89)**
lnHHSY
-0.001
0.006
(1.15)
(6.27)**
S1
-0.031
-0.020
(5.20)**
(2.46)*
S2
S3
-0.038
0.000
(6.74)**
(0.05)
-0.050
0.032
(8.58)**
(4.45)**
S4
-0.033
0.028
(6.15)**
(4.13)**
S5
-0.013
0.003
(2.09)*
(0.40)
C1
-0.057
0.057
(21.97)**
(19.21)**
0.025
-0.007
(8.24)**
(1.96)*
-0.030
0.042
C2
C3
C4
C5
C6
Observations
(11.44)**
(14.21)**
-0.046
0.026
(14.26)**
(6.71)**
-0.046
0.018
(17.74)**
(5.84)**
-0.020
0.019
(6.82)**
(5.56)**
158194
158194
Robust z statistics in parentheses
* significant at 5%; ** significant at 1%
36
Graph 11. Probability of Child Labour vs. GDP percapita
0.07
0.05
0.03
0.01
-0.01
-0.03
Estimated probabily of being an employed child
GDPpercapita growth
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
-0.05
Graph 12. Probability of School Attendance vs. GDP percapita
0.07
0.05
0.03
0.01
-0.01
Probability of attending the school
-0.03
GDPpercapita growth
37
T00
T99
T98
T97
T96
T95
T94
T93
T92
T91
T90
T89
T88
T87
T86
-0.05
Table 11. Results Model 2
(1)
(2)
empch
School attendance
lnGDPpc
-0.060
0.095
(6.19)**
(8.63)**
boy
0.022
0.032
(14.64)**
(19.32)**
age
0.047
-0.062
(97.52)**
(111.03)**
HHsize
0.005
-0.009
(11.34)**
(20.19)**
HHHmale
-0.024
0.014
(1.69)
(0.83)
HHHage
-0.001
0.002
(9.25)**
(12.05)**
HHHeduc
-0.010
0.015
(14.35)**
(20.22)**
HHHeduc2
0.000
-0.001
(12.08)**
(16.61)**
HHSfemale
0.003
0.012
(0.21)
(0.72)
HHSage
-0.003
0.003
(20.30)**
(19.00)**
HHSeduc
-0.014
0.023
(20.30)**
(29.84)**
HHSeduc2
0.001
-0.001
(17.72)**
(24.62)**
lnHHHY
-0.005
0.019
(4.68)**
(16.32)**
lnHHSY
0.000
0.006
(0.23)
(5.77)**
S1
-0.030
-0.021
(4.98)**
(2.59)**
S2
-0.037
-0.001
(6.55)**
(0.07)
S3
-0.049
0.031
(8.30)**
(4.30)**
S4
-0.032
0.027
(5.83)**
(3.96)**
S5
-0.012
0.002
(1.94)
(0.30)
C1
-0.055
0.057
(21.13)**
(19.06)**
C2
0.028
-0.007
(9.03)**
(2.09)*
C3
-0.027
0.042
(10.17)**
(14.12)**
C4
-0.047
0.026
(14.49)**
(6.74)**
C5
-0.045
0.018
(16.85)**
(5.61)**
C6
-0.018
0.018
(6.15)**
(5.45)**
Observations
158194
158194
Robust z statistics in parentheses
* significant at 5%; ** significant at 1%
38
Table 12
Large differences in earnings due to education
(Earnings by level of education 2000)
Education level
No schooling
Primary
Secondary
Tertiary
Total workforce
Wage (Monthly)
('000 Col$)
137
208
278
886
323
Wage (Monthly)
(USD$)
62.4
94.7
126.2
402.7
146.7
Source: Blom and Hansen (2002), p. 4.
39
Wage in percentage
of average
42.6%
64.6%
86.1%
274.5%
100.0%
Table 13. Results Model 3
(1)
(2)
empch
School attendance
lnGDPpc
-0.057
0.079
(5.88)**
(7.17)**
boy
0.022
0.032
(14.62)**
(19.23)**
age
0.047
-0.061
(97.43)**
(110.59)**
HHsize
0.005
-0.009
(11.19)**
(19.62)**
HHHmale
-0.026
0.021
(1.85)
(1.26)
HHHage
-0.001
0.002
(9.37)**
(12.18)**
HHHeduc
-0.010
0.016
(14.43)**
(20.65)**
HHHeduc2
0.000
-0.001
(11.77)**
(15.52)**
HHSfemale
0.002
0.012
(0.16)
(0.76)
HHSage
-0.003
0.003
(20.20)**
(18.81)**
HHSeduc
-0.015
0.024
(20.64)**
(31.02)**
HHSeduc2
0.001
-0.001
(17.91)**
(24.50)**
S1
-0.026
-0.047
(4.29)**
(5.58)**
S2
-0.033
-0.022
(5.90)**
(2.96)**
S3
-0.045
0.013
(7.75)**
(1.90)
S4
-0.029
0.017
(5.43)**
(2.39)*
S5
-0.011
-0.002
(1.76)
(0.31)
C1
-0.055
0.058
(21.17)**
(19.35)**
C2
0.027
-0.003
(8.75)**
(0.91)
C3
-0.029
0.052
(11.26)**
(18.19)**
C4
-0.046
0.025
(14.40)**
(6.36)**
C5
-0.045
0.021
(17.09)**
(6.67)**
C6
-0.020
0.025
(6.68)**
(7.51)**
Observations
158194
158194
Robust z statistics in parentheses
* significant at 5%; ** significant at 1%
40
Table 14. Results Model 4
(1)
(2)
(mean) empch (mean) School Attendance
(mean) lnGDPpc
8.448
18.884
(1.18)
(3.38)**
(mean) lnGDPpc2
-0.330
-0.723
(1.19)
(3.35)**
Constant
-53.978
-122.532
(1.16)
(3.39)**
Observations
64
64
R-squared
0.22
0.68
Absolute value of t statistics in parentheses
* significant at 5%; ** significant at 1%
Graph 13. Collapsed fitted values vs. GDP percapita
Employed children
Fitted values
.127444
.104141
12.7907
(mean) lnGDPpc
13.0753
School attendance
Fitted values
.828043
.773703
12.7907
(mean) lnGDPpc
41
13.0753
Graph 14. Collapsed fitted values vs. Quarters
Employed children
Fitted values
.127444
.104141
100 104 108 112 116 120 124 128 132 136 140 144 148 152 156 160163
qdate
School attendance
Fitted values
.828043
.773703
100 104 108 112 116 120 124 128 132 136 140 144 148 152 156 160163
qdate
42
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
1
La coyuntura económica en Colombia y Venezuela
Andrés LANGEBAEK
Patricia DELGADO
Fernando Mesa Parra
Octubre 1992
2
La tasa de cambio y el comercio colombo-venezolano
Fernando Mesa Parra
Andrés LANGEBAEK
Noviembre 1992
3
¿Las mayores exportaciones colombianas de café redujeron
el precio externo?
Carlos Esteban POSADA
Andrés LANGEBAEK
Noviembre 1992
4
El déficit público: una perspectiva macroeconómica.
Jorge Enrique RESTREPO
Juan Pablo ZÁRATE
Carlos Esteban POSADA
Noviembre 1992
5
El costo de uso del capital en Colombia.
Mauricio OLIVERA
Diciembre 1992
6
Colombia y los flujos de capital privado a América Latina
Andrés LANGEBAEK
Febrero 1993
7
Infraestructura física. “Clubs de convergencia” y crecimiento
económico.
José Dario URIBE
Febrero 1993
8
El costo de uso del capital: una nueva estimación (Revisión)
Mauricio OLIVERA
Marzo 1993
9
Dos modelos de transporte de carga por carretera.
Carlos Esteban POSADA
Edgar TRUJILLO CIRO
Alvaro CONCHA
Juan Carlos ELORZA
Marzo 1993
10
La determinación del precio interno del café en un modelo
de optimización intertemporal.
Carlos Felipe JARAMILLO
Carlos Esteban POSADA
Edgar TRUJILLO CIRO
Abril 1993
11
El encaje óptimo
Edgar TRUJILLO CIRO
Carlos Esteban POSADA
Mayo 1993
12
Crecimiento económico, “Capital
humano” y educación: la teoría y el
caso colombiano posterior a 1945
Carlos Esteban POSADA
Junio 1993
13
Estimación del PIB trimestral según los componentes del gasto.
Rafael CUBILLOS
Junio 1993
Fanny Mercedes VALDERRAMA
14
Diferencial de tasas de interés y flujos de capital en Colombia
(1980-1993)
Andrés LANGEBAEK
15
Empleo y capital en Colombia: nuevas
estimaciones (1950-1992)
Adriana BARRIOS
Septiembre 1993
Marta Luz HENAO
Carlos Esteban POSADA
Fanny Mercedes VALDERRAMA
Diego Mauricio VÁSQUEZ
16
Productividad, crecimiento y ciclos en la economía
colombiana (1967-1992)
Carlos Esteban POSADA
Septiembre 1993
17
Crecimiento económico y apertura en Chile y México y
perspectivas para Colombia.
Fernando MESA PARRA
Septiembre 1993
18
El papel del capital público en la producción, inversión y
el crecimiento económico en Colombia.
Fabio SÁNCHEZ TORRES
Octubre 1993
19
Tasa de cambio real y tasa de cambio de equilibrio.
Andrés LANGEBAEK
Octubre 1993
20
La evolución económica reciente: dos interpretaciones
alternativas.
Carlos Esteban POSADA
Noviembre 1993
21
El papel de gasto público y su financiación en la coyuntura
actual: algunas implicaciones complementarias.
Alvaro ZARTA AVILA
Diciembre 1993
22
Inversión extranjera y crecimiento económico.
Alejandro GAVIRIA
Diciembre 1993
Agosto 1993
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
Javier Alberto GUTIÉRREZ
23
Inflación y crecimiento en Colombia
Alejandro GAVIRIA
Carlos Esteban POSADA
Febrero 1994
24
Exportaciones y crecimiento en Colombia
Fernando MESA PARRA
Febrero 1994
25
Experimento con la vieja y la nueva teoría del crecimiento
económico (¿porqué crece tan rápido China?)
Carlos Esteban POSADA
Febrero 1994
26
Modelos económicos de criminalidad y la
una dinámica prolongada.
Carlos Esteban POSADA
Abril 1994
27
Regímenes cambiarios, política macroeconómica y flujos
de capital en Colombia.
Carlos Esteban POSADA
Abril 1994
28
Comercio intraindustrial: el caso colombiano
Carlos POMBO
Abril 1994
29
Efectos de una bonanza petrolera a la luz de un modelo
de optimización intertemporal.
Hernando ZULETA
Juan Pablo ARANGO
Mayo 1994
30
.
Crecimiento económico y productividad en Colombia:
una perspectiva de largo plazo (1957-1994)
Sergio CLAVIJO
Junio 1994
31
Inflación o desempleo: ¿Acaso hay escogencia en Colombia?
Sergio CLAVIJO
Agosto 1994
32
La distribución del ingreso y el sistema financiero
Edgar TRUJILLO CIRO
Agosto 1994
33
La trinidad económica imposible en Colombia: estabilidad
cambiaria, independencia monetaria y flujos de capital libres
Sergio CLAVIJO
Agosto 1994
34
¿’Déjà vu?: tasa de cambio, deuda externa y esfuerzo
exportador en Colombia.
Sergio CLAVIJO
Mayo 1995
35
La crítica de Lucas y la inversión en Colombia:
nueva evidencia
Mauricio CÁRDENAS
Mauricio OLIVERA
Septiembre 1995
36
Tasa de Cambio y ajuste del sector externo en Colombia.
Fernando MESA PARRA
Dairo ESTRADA
Septiembre 1995
37
Análisis de la evolución y composición del Sector Público.
Mauricio Olivera G.
Manuel Fernando Castro Q.
Fabio Sánchez T.
Septiembre 1995
38
Incidencia distributiva del IVA en un modelo del ciclo de vida.
Juan Carlos PARRA OSORIO
Fabio José SÁNCHEZ T.
Octubre 1995
39
Por qué los niños pobres no van a la escuela?
(Determinantes de la asistencia escolar en Colombia)
Fabio SÁNCHEZ TORRES
Jairo Augusto NÚÑEZ M.
Noviembre 1995
40
Matriz de Contabilidad Social 1992.
Fanny M. VALDERRAMA
Javier Alberto GUTIÉRREZ
Diciembre 1995
41
Multiplicadores de Contabilidad derivados de la Matriz
de Contabilidad Social
Javier Alberto GUTIÉRREZ
Enero 1996
Fanny M. VALDERRAMA G.
42
El ciclo de referencia de la economía colombiana.
Martin MAURER
María Camila URIBE S.
Febrero 1996
43
Impacto de las transferencias intergubernamentales en la
distribución interpersonal del ingreso en Colombia.
Juan Carlos PARRA OSORIO
Marzo 1996
44
Auge y colapso del ahorro empresarial en Colombia:
1983-1994
Fabio SÁNCHEZ TORRES
Abril 1996
Guillermo MURCIA GUZMÁN
Carlos OLIVA NEIRA
45
Evolución y comportamiento del gasto público en Colombia:
1950-1994
Cielo María NUMPAQUE
Mayo 1996
Ligia RODRÍGUEZ CUESTAS
posibilidad de
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
46
Los efectos no considerados de la apertura económica en el
mercado laboral industrial.
Fernando MESA PARRA
Javier Alberto GUTIÉRREZ
Mayo 1996
47
Un modelo de Financiamiento óptimo de un aumento
permanente en el gasto público:
Una ilustración con el caso colombiano.
Alvaro ZARTA AVILA
Junio 1996
48
Estadísticas descriptivas del mercado laboral masculino y
femenino en Colombia: 1976 -1995
Rocío RIBERO M.
Carmen Juliana GARCÍA B.
Agosto 1996
49
Un sistema de indicadores líderes para Colombia
Martín MAURER
María Camila URIBE
Javier BIRCHENALL
Agosto 1996
50
Evolución y determinantes de la productividad en Colombia:
Un análisis global y sectorial
Fabio SÁNCHEZ TORRES
Jorge Iván RODRÍGUEZ
Jairo NÚÑEZ MÉNDEZ
Agosto 1996
51
Gobernabilidad y Finanzas Públicas en Colombia.
César A. CABALLERO R
Noviembre 1996
52
Tasas Marginales Efectivas de Tributación en Colombia.
Mauricio OLIVERA G.
Noviembre 1996
53
Un modelo keynesiano para la economía colombiana
Fabio José SÁNCHEZ T.
Clara Elena PARRA
Febrero 1997
54
Trimestralización del Producto Interno Bruto por el lado
de la oferta.
Fanny M. VALDERRAMA
Febrero 1997
55
Poder de mercado, economías de escala, complementariedades
intersectoriales y crecimiento de la productividad en la
industria colombiana.
Juán Mauricio RAMÍREZ
Marzo 1997
56
Estimación y calibración de sistemas flexibles de gasto.
Jesús Orlando GRACIA
Gustavo HERNÁNDEZ
Abril 1997
57
Mecanismos de ahorro e Inversión en las Empresas Públicas
Colombianas: 1985-1994
Fabio SÁNCHEZ TORRES
Guilllermo MURCIA G.
Mayo 1997
58
Capital Flows, Savings and investment in Colombia: 1990-1996
José Antonio OCAMPO G.
Camilo Ernesto TOVAR M.
Mayo 1997
59
Un Modelo de Equilibrio General Computable con
Competencia imperfecta para Colombia.
Juan Pablo ARANGO
Jesús Orlando GRACIA
Gustavo HERNÁNDEZ
Juan Mauricio RAMÍREZ
Junio 1997
Javier A. BIRCHENALL J.
Julio 1997
Alberto CASTAÑEDA C.
Gabriel PIRAQUIVE G.
Julio 1997
60
El cálculo del PIB Potencial en Colombia.
61
Determinantes del Ahorro de los hogares.
de su caída en los noventa.
62
Los ingresos laborales de hombres y
mujeres en Colombia: 1976-1995
Rocío RIBERO
Claudia MEZA
Agosto 1997
63
Determinantes de la participación laboral de hombres y
mujeres en Colombia: 1976-1995
Rocío RIBERO
Claudia MEZA
Agosto 1997
64
Inversión bajo incertidumbre en la Industria Colombiana:
1985-1995
Javier A. BIRCHENALL
Agosto 1997
65
Modelo IS-LM para Colombia. Relaciones de largo plazo y
fluctuaciones económicas.
Jorge Enrique RESTREPO
Agosto 1997
66
Correcciones a los Ingresos de las Encuestas de hogares y
distribución del Ingreso Urbano en Colombia.
Jairo A. NÚÑEZ MÉNDEZ
Jaime A. JIMÉNEZ CASTRO
Septiembre 1997
67
Ahorro, Inversión y Transferencias en las Entidades
Territoriales Colombianas
Fabio SÁNCHEZ TORRES
Mauricio OLIVERA G.
Giovanni CORTÉS S.
Octubre 1997
Explicación
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
68
Efectos de la Tasa de cambio real sobre la Inversión
industrial en un Modelo de transferencia de precios.
Fernando MESA PARRA
Leyla Marcela SALGUERO
Fabio SÁNCHEZ TORRES
Octubre 1997
69
Convergencia Regional: Una revisión del caso
Colombiano.
Javier A. BIRCHENALL
Guillermo E. MURCIA G.
Octubre 1997
70
Income distribution, human capital and economic
growth in Colombia.
Javier A. BIRCHENALL
Octubre 1997
71
Evolución y determinantes del Ahorro del Gobierno Central.
Fabio SÁNCHEZ TORRES
Ma. Victoria ANGULO
Noviembre 1997
72
Macroeconomic Perforrmance and Inequality in Colombia:
1976-1996
Raquel BERNAL
Mauricio CÁRDENAS
Jairo NÚÑEZ MÉNDEZ
Fabio SÁNCHEZ TORRES
Diciembre 1997
73
Liberación comercial y salarios en Colombia: 1976-1994
Donald ROBBINS
Enero 1998
74
Educación y salarios relativos en Colombia: 1976-1995
Determinantes, evolución e implicaciones para
la distribución del Ingreso
Jairo NÚÑEZ MÉNDEZ
Fabio SÁNCHEZ TORRES
Enero 1998
75
La tasa de interés “óptima”
Carlos Esteban POSADA
Edgar TRUJILLO CIRO
Febrero 1998
76
Los costos económicos de la criminalidad y la violencia en
Colombia: 1991-1996
Edgar TRUJILLO CIRO
Martha Elena BADEL
Marzo 1998
77
Elasticidades Precio y Sustitución para la Industria
Colombiana.
Juán Pablo ARANGO
Jesús Orlando GRACIA
Gustavo HERNÁNDEZ
Marzo 1998
78
Flujos Internacionales de Capital en Colombia:
Un enfoque de Portafolio
Ricardo ROCHA GARCÍA
Fernando MESA PARRA
Marzo 1998
79
Macroeconomía, ajuste estructural y equidad en Colombia:
1978-1996
José Antonio OCAMPO
María José PÉREZ
Camilo Ernesto TOVAR
Francisco Javier LASSO
Marzo 1998
80
La Curva de Salarios para Colombia. Una Estimación de las
Relaciones entre el Desempleo, la Inflación y los Ingresos
Laborales: 1984- 1996.
Fabio SÁNCHEZ TORRES
Jairo NÚÑEZ MÉNDEZ
Marzo 1998
81
Participación, Desempleo y Mercados Laborales en Colombia.
Jaime TENJO G.
Rocio RIBERO M.
Abril 1998
82
Reformas comerciales, márgenes de beneficio y
productividad en la industria colombiana
Juán Pablo ARANGO
Jesús Orlando GRACIA
Gustavo HERNÁNDEZ
Juán Mauricio RAMÍREZ
Abril 1998
83
Capital y Crecimiento Económico en un Modelo
Dinámico: Una presentación de la dinámica
Transicional para los casos de EEUU y Colombia
Alvaro ZARTA AVILA
Mayo 1998.
84
Determinantes de la Inversión en Colombia: E videncia sobre
el capital humano y la violencia.
Clara Helena PARRA
Junio 1998.
85
Mujeres en sus casas: Un recuento de la población
Femenina económicamente activa
Piedad URDINOLA C.
Junio 1998.
86
Descomposición de la desigualdad del Ingreso laboral
Urbano en Colombia: 1976-1997
Fabio SÁNCHEZ TORRES
Jairo NÚÑEZ MÉNDEZ
Junio 1998.
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
87
El tamaño del Estado Colombiano Indicadores y tendencias:
1976-1997
Angela CORDI GALAT
Junio 1998.
88
Elasticidades de sustitución de las importaciones para la
economía colombiana.
Gustavo HERNÁNDEZ
Junio 1998.
89
La tasa natural de desempleo en Colombia
Martha Luz HENAO
Norberto ROJAS
Junio 1998.
90
The role of shocks in the colombian economy
Ana María MENÉNDEZ
Julio 1998.
91
The determinants of Human Capital Accumulation in
Colombia, with implications for Trade and Growth Theory
Donald J. ROBBINS
Julio 1998.
92
Estimaciones de funciones de demanda de trabajo dinámicas
para la economía colombiana, 1980-1996
Alejandro VIVAS BENÍTEZ
Stefano FARNÉ
Dagoberto URBANO
Julio 1998.
93
Análisis de las relaciones entre violencia y equidad.
Alfredo SARMIENTO
Lida Marina BECERRA
Agosto 1998.
94
Evaluación teórica y empírica de las exportaciones
no tradicionales en Colombia
Fernando MESA PARRA
María Isabel COCK
Angela Patricia JIMÉNEZ
Agosto 1998.
95
Valoración económica del empleo doméstico femenino
no remunerado, en Colombia, 1978-1993
Piedad URDINOLA C.
Agosto 1998.
96
Eficiencia en el Gasto Público de Educación.
María Camila URIBE
Agosto 1998.
97
El desempleo en Colombia: tasa natural, desempleo cíclico
y estructural y la duración del desempleo: 1976-1998.
Jairo NÚÑEZ M.
Raquel BERNAL S.
Septiembre 1998.
98
Productividad y retornos sociales del Capital humano:
Microfundamentos y evidencia para Colombia.
Francisco A. GONZÁLEZ R.
Carolina GUZMÁN RUIZ
Angela L. PACHÓN G.
Noviembre 1998.
99
Reglas monetarias en Colombia y Chile
Jorge E. RESTREPO L.
Enero 1999.
100
Inflation Target Zone: The Case of Colombia: 1973-1994
Jorge E. RESTREPO L.
Febrero 1999.
101
¿ Es creíble la Política Cambiaria en Colombia?
Carolina HOYOS V.
Marzo 1999.
102
La Curva de Phillips, la Crítica de Lucas y la persistencia
de la inflación en Colombia.
Javier A.BIRCHENALL
Abril 1999.
103
Un modelo macroeconométrico para la economía
Colombiana
Javier A.BIRCHENALL
Juan Daniel OVIEDO
Abril 1999.
104
Una revisión de la literatura teórica y la experiencia
Internacional en regulación
Marcela ESLAVA MEJÍA
Abril 1999.
105
El transporte terrestre de carga en Colombia
Documento para el Taller de Regulación.
Marcela ESLAVA MEJÍA
Abril 1999.
Eleonora LOZANO RODRÍGUEZ
106
Notas de Economía Monetaria. (Primera Parte)
Juan Carlos ECHEVERRY G.
Abril 1999.
107
Ejercicios de Causalidad y Exogeneidad para Ingresos
salariales nominales públicos y privados Colombianos
(1976-1997).
Mauricio BUSSOLO
Jesús Orlando GRACIA
Camilo ZEA
Mayo 1999.
108
Real Exchange Rate Swings and Export Behavior:
Explaining the Robustness of Chilean Exports.
Felipe ILLANES
Mayo 1999.
109
Segregación laboral en las 7 principales ciudades del país.
Piedad URDINOLA
Mayo 1999.
110
Estimaciones trimestrales de la línea de pobreza y sus relaciones Jairo NÚÑEZ MÉNDEZ
con el desempeño macroeconómico Colombiano: (1977-1997) Fabio José SÁNCHEZ T.
Mayo 1999
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
111
Costos de la corrupción en Colombia.
Marta Elena BADEL
Mayo 1999
112
Relevancia de la dinámica transicional para el
crecimiento de largo plazo: Efectos sobre las tasas de
interés real, la productividad marginal y la estructura
de la producción para los casos de EEUU y Colombia..
Alvaro ZARTA
Junio 1999
113
La recesión actual en Colombia: Flujos, Balances y
Política anticíclica
Juan Carlos ECHEVERRY
Junio 1999
114
Monetary Rules in a Small Open Economy
Jorge E. RESTREPO L.
Junio 1999
115
El Balance del Sector Público y la Sostenibilidad
Fiscal en Colombia
Juan Carlos ECHEVERRY
Gabriel PIRAQUIVE
Natalia SALAZAR FERRO
Ma. Victoria ANGULO
Gustavo HERNÁNDEZ
Cielo Ma. NUMPAQUE
Israel FAINBOIM
Carlos Jorge RODRIGUEZ
Junio 1999
116
Crisis y recuperación de las Finanzas Públicas lecciones
de América Latina para el caso colombiano.
Marcela ESLAVA MEJÍA
Julio 1999
117
Complementariedades Factoriales y Cambio Técnico
en la Industria Colombiana.
Gustavo HERNÁNDEZ
Juan Mauricio RAMÍREZ
Julio 1999
118
¿Hay un estancamiento en la oferta de crédito?
Juan Carlos ECHEVERRY
Natalia SALAZAR FERRO
Julio 1999
119
Income distribution and macroeconomics in Colombia.
Javier A. Birchenall J.
Julio 1999.
120
Transporte carretero de carga. Taller de regulación.
DNP-UMACRO. Informe final.
Juan Carlos ECHEVERRY G. Agosto 1999.
Marcela ESLAVA MEJÍA
Eleonora LOZANO RODRIGUEZ
121
¿ Se cumplen las verdades nacionales a nivel regional?
Primera aproximación a la construcción de matrices de
contabilidad social regionales en Colombia.
Nelly.Angela CORDI GALAT Agosto 1999.
122
El capital social en Colombia.
La medición nacional con el BARCAS
Separata N° 1 de 5
John SUDARSKY
Octubre 1999.
123
El capital social en Colombia.
La medición nacional con el BARCAS
Separata N° 2 de 5
John SUDARSKY
Octubre 1999.
124
El capital social en Colombia.
La medición nacional con el BARCAS
Separata N° 3 de 5
John SUDARSKY
Octubre 1999.
125
El capital social en Colombia.
La medición nacional con el BARCAS
Separata N° 4 de 5
John SUDARSKY
Octubre 1999.
126
El capital social en Colombia.
La medición nacional con el BARCAS
Separata N° 5 de 5
John SUDARSKY
Octubre 1999.
127
The Liquidity Effect in Colombia
Jorge Enrique RESTREPO
Noviembre 1999.
128
Upac: Evolución y crisis de un modelo de desarrollo.
Juan Carlos ECHEVERRI G.
Orlando Gracia
B. Piedad Urdinola
Diciembre 1999.
129
Confronting fiscal imbalances via intertemporal
Economics, politics and justice: the case of Colombia
Juan Carlos ECHEVERRY G. Diciembre 1999.
Verónica Navas-Ospina
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
130
La tasa de interés en la coyuntura reciente en Colombia.
Jorge Enrique RESTREPO
Edgar TRUJILLO CIRO
Diciembre 1999.
131
Los ciclos económicos en Colombia. Evidencia empírica:
(1977-1998)
Jorge Enrique RESTREPO
José Daniel REYES PEÑa
Enero 2000.
132
Colombia'natural trade partners and its bilateral trade
performance: Evidence from 1960 to 1996
Hernán Eduardo VALLEJO
Enero 2000.
133
Los derechos constitucionales de prestación y sus
implicaciones económico- políticas. Los casos del
derecho a la salud y de los derechos de los reclusos
Luis Carlos SOTELO
Febrero 2000.
134
La reactivación productiva del sector privado colombiano
(Documento elaborado para el BID)
Luis Alberto ZULETA
Marzo 2000.
135
Geography and Economic Development:
A Municipal Approach for Colombia.
Fabio JOSÉ SÁNCHEZ T.
Jairo NÚÑEZ MÉNDEZ
Marzo 2000.
136
La evaluación de resultados en la modernización
del Estado en América Latina. Restricciones y
Estrategia para su desarrollo.
Eduardo WIESNER DURÁN
Abril 2000.
137
La regulación de precios del transporte de carga por
carretera en Colombia.
Marcela ESLAVA MEJÍA
Abril 2000.
138
El conflicto armado en Colombia.
Una aproximación a la teoría de juegos.
Yuri GORBANEFF
Flavio JÁCOME
Julio 2000.
139
Determinación del consumo básico de agua potable
subsidiable en Colombia.
Juan Carlos JUNCA SALAS
Noviembre 2000.
Incidencia fiscal de los incentivos tributarios
Juan Ricardo ORTEGA
Noviembre 2000.
Gabriel Armando PIRAQUIVE
Gustavo Adolfo HERNÁNDEZ
Carolina SOTO LOSADA
Sergio Iván PRADA
Juan Mauricio RAMIREZ
141
Exenciones tributarias:
Costo fiscal y análisis de incidencia
Gustavo A. HERNÁNDEZ
Carolina SOTO LOSADA
Sergio Iván PRADA
Juan Mauricio RAMIREZ
Diciembre 2000
142
La contabilidad del crecimiento, las dinámicas transicionales y
el largo plazo: Una comparación internacional de 46 países y
una presentación de casos de economías tipo:
EEUU, Corea del Sur y Colombia.
Alvaro ZARTA AVILA
Febrero 2001
143
¿Nos parecemos al resto del mundo?
El Conflicto colombiano en el contexto internacional.
Juan Carlos ECHEVERRY G.
Natalia SALAZAR FERRO
Verónica NAVAS OSPINA
Febrero 2001
144
Inconstitucionalidad del Plan Nacional de Desarrollo:
causas, efectos y alternativas.
Luis Edmundo SUÁREZ S.
Diego Mauricio AVILA A.
Marzo 2001
145
La afiliación a la salud y los efectos redistributivos
de los subsidios a la demanda.
Hernando MORENO G.
Abril 2001
146
La participación laboral: ¿qué ha pasado y qué
podemos esperar?
Mauricio SANTA MARÍA S. Abril 2001
Norberto ROJAS DELGADILLO
147
Análisis de las importaciones agropecuarias en la
década de los Noventa.
Gustavo HERNÁNDEZ
Juan Ricardo PERILLA
Mayo 2001
148
Impacto económico del programa de Desarrollo
alternativo del Plan Colombia
Gustavo A. HERNÁNDEZ
Sergio Iván PRADA
Juan Mauricio RAMÍREZ
Mayo 2001
.
140
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
149
Análisis de la presupuestación de la inversión de la Nación.
Ulpiano AYALA ORAMAS
Mayo 2001
150
DNPENSION: Un modelo de simulación para estimar
el costo fiscal del sistema pensional colombiano.
Juan Carlos PARRA OSORIO
Mayo 2001
151
La oferta de combustible de Venezuela en la frontera
con Colombia: una aproximación a su cuantificación
Hernando MORENO G.
Junio 2001
152
Shocks fiscales y términos de intercambio en el caso
colombiano.
Ómer ÖZAK MUñOZ.
Julio 2001
153
Demanda por importaciones en Colombia:
Una estimación.
Igor Esteban ZUCCARDI
Julio 2001
154
Elementos para mejorar la adaptabilidad del
mercado laboral colombiano.
Mauricio SANTA MARÍA S. Agosto 2001
Norberto ROJAS DELGADILLO
155
¿Qué tan poderosas son las aerolíneas colombianas?
Estimación de poder de mercado de las rutas colombianas.
Ximena PEÑA PARGA
156
Elementos para el debate sobre una nueva reforma
pensional en Colombia.
Juan Carlos ECHEVERRY
Septiembre 2001
Andrés ESCOBAR ARANGO
César MERCHÁN HERNÁNDEZ
Gabriel PIRAQUIVE GALEANO
Mauricio SANTA MARÍA S.
157
Agregando votos en un sistema altamente
desistitucionalizado.
Francisco GUTIÉRREZ S.
Octubre
158
Eficiencia -X en el Sector Bancario Colombiano
Carlos Alberto CASTRO I
Noviembre 2001
159
Determinantes de la calidad de la educación en Colombia.
Alejandro GAVIRIA
Jorge Hugo BARRIENTOS
Noviembre 2001
160
Evaluación de la descentralización municipal.
Descentralización y macroeconomía
Fabio SÁNCHEZ TORRES
Noviembre 2001
161
Impuestos a las transacciones: Implicaciones sobre el bienestar
y el crecimiento.
Rodrigo SUESCÚN
Noviembre 2001
162
Strategic Trade Policy and Exchange Rate Uncertainty
Fernando MESA PARRA
Noviembre 2001
163
Evaluación de la descentralización municipal en Colombia.
Avances y resultados de la descentralización
Política en Colombia
Alberto MALDONADO C.
Noviembre 2001
164
Choques financieros, precios de activos y recesión
en Colombia.
Alejandro BADEL FLÓREZ
Noviembre 2001
165
Evaluación de la descentralización municipal en
Colombia. ¿Se consolidó la sostenibilidad fiscal de los
municipios colombianos durante los años noventa.
Juan Gonzalo ZAPATA
Olga Lucía ACOSTA
Adriana GONZÁLEZ
Noviembre 2001
166
Evaluación de la descentralización municipal en
Colombia. La descentralización en el Sector de
Agua potable y Saneamiento básico.
Ma. Mercedes MALDONADO Noviembre 2001
Gonzalo VARGAS FORERO
167
Evaluación de la descentralización municipal en
Colombia. La relación entre corrupción y proceso
de descentralización en Colombia.
Edgar GONZÁLEZ SALAS
Diciembre 2001
168
Evaluación de la descentralización municipal en
Colombia. Estudio general sobre antecedentes,
diseño, avances y resultados generales del proceso de
descentralización territorial en el Sector Educativo.
Carmen Helena VERGARA
Mary SIMPSON
Diciembre 2001
169
Evaluación de la descentralización municipal en
Colombia. Componente de capacidad institucional.
Edgar GONZÁLEZ SALAS
Diciembre 2001
Agosto 2001
2001
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
170
Evaluación de la descentralización municipal en
Colombia. Evaluación de la descentralización en
Salud en Colombia.
Iván JARAMILLO PÉREZ
Diciembre 2001
171
External Trade, Skill, Technology and the recent
increase of income inequality in Colombia
Mauricio SANTA MARÍA S.
Diciembre 2001
172
Seguimiento y evaluación de la participación de los
resguardos indígenas en los ingresos corrientes de la
Nación para el período 1998 y 1999.
Dirección de Desarrollo
Territorial
Diciembre 2001
173
Exposición de Motivos de la Reforma de la Ley 60 de
1993. Sector Educación y Sector Salud
Dirección de Desarrollo
Social
Diciembre 2001
174
Transferencias, incentivos y la endogenidad del gasto
Territorial. Seminario internacional sobre Federalismo
fiscal - Secretaría de Hacienda de México, CEPAL,
ILPES, CAF - Cancún, México. 18-20 de Mayo de 2000
Eduardo WIESNER DURÁN
Enero 2002.
175
Cualificación laboral y grado de sindicalización
Flavio JÁCOME LIÉVANO
Enero 2002.
176
OFFSETS: Aproximación teórica y experiencia
Internacional.
Nohora Eugenia POSADA
Febrero 2002.
Yaneth Cristina GIHA TOVAR
Paola BUENDÍA GARCÍA
Alvaro José CHÁVEZ G.
177
Pensiones: conceptos y esquemas de financiación
César Augusto MERCHÁN H. Febrero 2002.
178
La erradicación de las minas antipersonal sembradas
en Colombia - Implicaciones y costos-
Yilberto LAHUERTA P.
Ivette María ALTAMAR
Marzo 2002.
179
Economic growth in Colombia: A reversal of "Fortune"?
Mauricio CÁRDENAS S.
Marzo 2002.
180
El siglo del modelo de desarrollo.
Juan Carlos ECHEVERRY G
Abril 2002.
181
Metodología de un Modelo ARIMA condicionado
para el pronóstico del PIB.
Juan Pablo HERRERA S.
Gustavo A. HERNÁNDEZ D.
Abril 2002.
182
¿Cuáles son los colombianos con pensiones
privilegiadas?
César Augusto MERCHÁN H. Abril 2002.
183
Garantías en carreteras de primera generación.
Impacto económico.
José Daniel REYES PEÑA.
Abril 2002
184
Impacto económico de las garantías de la Nación
en proyectos de infraestructura.
José Daniel REYES PEÑA.
Abril 2002
185
Aproximación metodológica y cuantitativa
de los costos económicos generados por el
problema de las drogas ilícitas en Colombia
(1995 - 2000)
Ricardo PÉREZ SANDOVAL
Andrés VERGARA BALLÉN
Yilberto LAHUERTA P
Abril 2002
186
Tendencia, ciclos y distribución del ingreso
en Colombia: una crítica al concepto de
"modelo de desarrollo"
Juan Carlos ECHEVERRY G.
Andrés ESCOBAR ARANGO
Mauricio SANTA MARÍA S.
Abril 2002.
187
Crecimiento y ciclos económicos. Efectos de los choques
de oferta y demanda en el crecimiento colombiano
Igor Esteban ZUCCARDI H.
Mayo 2002.
188
A general equilibrium model for tax policy
analysis in Colombia. The MEGATAX model.
Thomas Fox RUTHERFORD.
Miles Kenneth. LIGHT
Mayo 2002.
189
A dynamic general equilibrium model for tax
policy analysis in Colombia.
Thomas Fox RUTHERFORD.
Miles Kenneth. LIGHT
Gustavo HERNÁNDEZ
Mayo 2002.
190
Sistema Bancario Colombiano:
¿Somos eficientes a nivel internacional?
Alejandro BADEL FLÓREZ.
Junio 2002.
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
191
Política para mejorar el servicio de transporte público
urbano de pasajeros.
DNP: DIE- GEINF
Junio 2002.
192
Two decades of economic and social development
in urban Colombia: a mixed outcome
Carlos Eduardo VÉLEZ
Mauricio SANTA MARÍA,
Natalia MILLAN
Bénédicte DE LA BRIERE
World Bank (LAC/PREM)
Junio 2002.
Jairo NÚÑEZ MÉNDEZ
Roberto STEINER
Ximena CADENA
Renata PARDO
CEDE, U. de los Andes
Junio 2002.
193
¿Cuáles colegios ofrecen mejor educación en Colombia?
194
Nuevos enfoques de política regional en América Latina:
El caso de Colombia en perspectiva histórica.
Las nuevas teorías y enfoques conceptuales sobre el
desarrollo regional. ¿Hacia un nuevo paradigma?
Separata 1 de 7
Edgard MONCAYO J.
Julio 2002.
195
Nuevos enfoques de política regional en América Latina:
El caso de Colombia en perspectiva histórica.
Las políticas regionales: Un enfoque por generaciones
Separata 2 de 7
Edgard MONCAYO J.
Julio 2002.
196
Nuevos enfoques de política regional en América Latina:
El caso de Colombia en perspectiva histórica.
Un mundo de geometría variable:
Los territorios que ganan y los que pierden.
Separata 3 de 7
Edgard MONCAYO J.
Julio 2002.
197
Nuevos enfoques de política regional en América Latina:
El caso de Colombia en perspectiva histórica.
Enfoques teóricos y evidencias empíricas sobre
el desarrollo regional en Colombia.
Separata 4 de 7
Edgard MONCAYO J.
Julio 2002.
198
Nuevos enfoques de política regional en América Latina:
El caso de Colombia en perspectiva histórica.
Las políticas regionales en Colombia.
Separata 5 de 7
Edgard MONCAYO J.
Julio 2002.
199
Nuevos enfoques de política regional en América Latina:
El caso de Colombia en perspectiva histórica.
Tendencias del desarrollo regional en Colombia.
-Polarización, apertura y conflictoSeparata 6 de 7
Edgard MONCAYO J.
Julio 2002.
200
Nuevos enfoques de política regional en América Latina:
El caso de Colombia en perspectiva histórica.
Marco conceptual y metodológico para el diseño
de una nueva generación de políticas de desarrollo
regional en Colombia.
Separata 7 de 7
Edgard MONCAYO J.
Julio 2002.
201
Viabilidad de los servicios públicos domiciliarios
en la ciudad de Santiago de Cali.
Mauricio SANTA MARÍA
Francisco BERNAL
Carlos David BELTRÁN
David VILLALBA
Agosto 2002
202
Optimal enforcement: Finding the right balance
Jaime Andrés ESTRADA
Agosto 2002
203
Does corporate governance matter for developing
countries?
An overview of the Mexican case.
Paula ACOSTA MÁRQUEZ
Agosto 2002
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
204
Reflexiones sobre el proceso de paz del gobierno de Andrés
PASTRANA y las FARC-Ep: (1998-2002)
Camilo LEGUÍZAMO
Agosto 2002
205
Contratación pública en Colombia y teoría Económica.
Yuri GORBANEFF
Septiembre 2002.
206
Does planning pay to perform in infrastructure?
Deconstructing the babylon tower on the planning/
performance relationships in energy, telecommunications
and transport sectors – colombian case.
Daniel TORRES GRACIA
Septiembre 2002.
207
A dynamic analysis of household decision making in urban
Colombia, 1976-1998
Changes in household structure, human capital
and its returns, and female labor force participation .
Fabio SÁNCHEZ TORRES
Jairo Núñez Méndez
Octubre 2002.
208
Inversión pública sectorial
y crecimiento
Económico: Una aproximación desde la
Metodología VAR.
Alvaro A. PERDOMO S.
Octubre 2002.
209
Impacto macroeconómico y distributivo del Impuesto de
seguridad democrática.
Ömer ÖZAK MUÑOZ.
Oscar Mauricio VALENCIA
Octubre 2002.
210
Empleo informal y evasión fiscal en Colombia.
Jairo A. NÚÑEZ MÉNDEZ
Octubre 2002.
211
Diagnóstico del programa de reinserción en Colombia:
mecanismos para incentivar la desmovilización voluntaria
individual.
Maria Eugenia PINTO B.
Andrés VERGARA BALLÉN
Yilberto LAHUERTA P.
Noviembre 2002.
212
Economías de escala en los hogares y pobreza.
Tesis para optar el título de Magíster en Teoría y
Política Económica de la Universidad Nacional
de Colombia.
Francisco Javier LASSO V.
Noviembre 2002.
213
Nueva metodología de Encuesta de hogares.
¿Más o menos desempleados?
Francisco Javier LASSO V.
Noviembre 2002.
214
Una aproximación de la Política Comercial Estratégica
para el ingreso de Colombia al ALCA.
Ricardo E. ROCHA G..
Juan Ricardo PERILLA
Ramiro LÓPEZ SOLER
Diciembre 2002.
215
The political business cycle in Colombia
on the National and Regional level.
Allan DRAZEN
Marcela ESLAVA
University of Maryland
Enero 2003.
216
Balance macroeconómico de 2002 y
Perspectivas para 2003.
Dirección de Estudios
Económicos
Enero 2003.
217
Women workers in Bogotà ‘s Informal sector:
Gendered impact of structural adjustment
Policies in the 1990s.
Tesis para optar el título de Magíster en
Estudios de Desarrollo del Instituto de Estudios
Sociales de The Hague- Holanda.
Jairo G. ISAZA CASTRO
Febrero 2003.
218
Determinantes de la duración del desempleo en
el área metropolitana de Cali 1988-1998.
(Documento elaborado por profesores del Departamento de Economía de la Universidad del Valle)
Carlos E. CASTELLAR P.
José Ignacio URIBE G.
Marzo 2003.
219
Conflicto, violencia y actividad criminal en Colombia:
Un análisis espacial.
Fabio SÁNCHEZ TORRES
Ana María DÍAZ
Michel FORMISANO
Marzo 2003.
220
Evaluating the impact of SENA on earnings and
Employment.
Alejandro GAVIRIA URIBE
Jairo A. NÚÑEZ MÉNDEZ
Abril 2003.
221
Un análisis de la relación entre inversión extranjera y
Comercio exterior en la economía colombiana.
Erika Bibiana PEDRAZA
Abril 2003.
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
Abril 2003.
222
Free Trade Area of the Americas. An impact
Assessment for Colombia.
Miles Kenneth. LIGHT
Thomas Fox RUTHERFORD
223
Construcción de una Matriz de Contabilidad Social
Financiera para Colombia.
Gustavo Adolfo HERNÁNDEZ Mayo 2003.
224
Elementos para el análisis de Incidencia tributaria.
Andrés ESCOBAR
Gustavo HERNÁNDEZ
Gabriel PIRAQUIVE
Juan Mauricio RAMIREZ
Mayo 2003.
225
Desempeño económico por tipo de4 firma:
Empresas nacionales vs. Grandes y pequeñas receptoras
De inversión extranjera.
Erika Bibiana PEDRAZA
Mayo 2003.
226
El balance estructural del Gobierno Central en Colombia.
Natalia SALAZAR
Diego PRADA
Junio 2003.
227
Descentralización y Equidad en América Latina:
Enlaces Institucionales y de Política
Eduardo WIESNER
Junio 2003.
228
Ciclos económicos y mercado laboral en Colombia:
¿quién gana más, quién pierde más? 1984-2000.
Fabio SÁNCHEZ TORRES
Luz Magdalena SALAS
Oskar NUPIA
Julio 2003.
229
Efectos de un acuerdo bilateral de libre comercio con
Estados Unidos
Direcciones de Estudios
Económicos y de
Desarrollo Empresarial
del DNP
Julio 2003.
230
Pobreza, crimen y crecimiento regional en Colombia.
(Versión para comentarios)
Ricardo Ernesto ROCHA G.
Agosto 2003.
Hermes Fernando MARTÍNEZ
231
Contracciones leves y profundas:
Efectos asimétricos sobre la pobreza
El caso colombiano 1984-2000.
Jorge E. SÁENZ CASTRO
Juan Pablo HERRERA S.
Oscar E. GUZMÁN SILVA
Agosto 2003.
232
Sistema de modelos multivariados para la proyección
del Producto Interno Bruto
Carlos Alberto CASTRO I.
Septiembre 2003.
233
Yet another lagging, coincident and leading index for
The Colombian economy.
Carlos Alberto CASTRO I.
Septiembre 2003.
234
Posibles implicaciones de la legalización del consumo,
Producción y comercialización de las drogas en Colombia.
Andrés VERGARA BALLÉN
Yilberto LAHUERTA P.
Sandra Patricia CORREA
Septiembre 2003.
235
Impactos económicos generados por el uso de minas
antipersonal en Colombia.
Yilberto LAHUERTA P.
Septiembre 2003.
236
¿Cuánto duran los colombianos en el desempleo y en el
Empleo?: Un análisis de supervivencia.
Hermes Fernando MARTÍNEZ Septiembre 2003.
237
Barreras a la entrada en el mercado de compras del
Sector público.
Un análisis de estructura de mercado en la perspectiva
De la negociación del Area de Libre Comercio de las
Américas.
Fernando J. ESTUPIÑAN
238
Relative labor supply and the gender wage Gap:
Evidence for Colombia and the United States.
Diego F. ANGEL-URDINOLA Octubre 2003.
Quentin WODON
239
The gender wage Gap and poverty in Colombia.
.
Diego F. ANGEL-URDINOLA Octubre 2003.
Quentin WODON
240
The impact on inequality of raising the minimum wage:
Gap- narrowing and reranking effects.
Diego F. ANGEL-URDINOLA Octubre 2003.
Quentin WODON
Octubre 2003.
ARCHIVOS DE ECONOMIA
No
Título
Autores
Fecha
241
Inversión y restricciones crediticias en la década de
los 90 en Colombia.
Catalina DELGADO G.
Octubre 2003.
242
Metodologías de estimación del balance estructural:
Una aplicación al caso colombiano.
Luis Edgar BASTO M..
Noviembre 2003.
243
The cost of disinflation in Colombia
-A sacrifice Ratio Approach-
José Daniel REYES P..
Noviembre 2003.
244
Evaluación de la Eficiencia en Instituciones
Hospitalarias públicas y privadas con Data
Envelopment Analysis (DEA)
Ma. Cristina PEÑALOZA R.
Diciembre 2003.
245
Medición de eficiencia técnica relativa en hospitales
Públicos de baja complejidad, mediante la metodología
Data envelopment analysis – DEA
Tesis para optar al título de Magíster en Economía, de la
Pontificia Universidad Javeriana.
Maureen Jennifer PINZON M. Diciembre 2003.
246
Child labour and the Economic recession of 1999
In Colombia.
Claudia Marcela UMAÑA A.
:
Enero 2004.