Archives

  • 2018-07
  • 2018-10
  • 2018-11
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • 2021-10
  • 2021-11
  • 2021-12
  • 2022-01
  • 2022-02
  • 2022-03
  • 2022-04
  • 2022-05
  • 2022-06
  • 2022-07
  • 2022-08
  • 2022-09
  • 2022-10
  • 2022-11
  • 2022-12
  • 2023-01
  • 2023-02
  • 2023-03
  • 2023-04
  • 2023-05
  • 2023-06
  • 2023-07
  • 2023-08
  • 2023-09
  • 2023-10
  • 2023-11
  • 2023-12
  • 2024-01
  • 2024-02
  • 2024-03
  • 2024-04
  • 2024-05

    • The results obtained with the second

    2018-11-13

    The results obtained with the second way of measuring education are show in Tables A.2 and A.3. The returns to education are positive, suggesting that increases in earnings are substantially higher from 10 years of schooling for both genders. It should be noted that the first year of schooling yielding the highest return is the 11th grade, the last grade of high school. We also observed that the threshold effect (from 10 years of schooling) of education increases with the age at which one started to work. This is new evidence suggesting the negative effects of early work. Therefore, we considered the existence of a threshold effect, besides the years of schooling variable, and included the variable S=Z(S−λ) in the specification, where λ is the threshold, i.e. the value of schooling from which the return on education increases, and Z is a dummy variable that assumes value 0 for S≤λ and value 1 for S>λ. Using the variable S=Z(S−10) in the specification, the results were observed on earnings for each additional year of education from 10 years of schooling. Table 3 shows the percentage marginal effects of years of schooling and the threshold effect (from 10 years of schooling) on earnings, by the age intervals in which one began to work. For both men and women, the percentage increase in earnings for each additional year of education after 10 years of schooling increases as the age at which they nicergoline entered the labor market increases. This is an indirect evidence that the negative effects of working at an early age on future earnings are caused, among other reasons, by the quality of the education received. This has both a direct and an indirect effect on the kind of work obtained. For men, for example, an additional year of education from 10 years of schooling yields a return of 26.7% for those who start working after the age of 23, while the return for those who have to work before reaching the age of 9 is 12.4%. It should be noted that, for both genders, the positive threshold effect on earnings increases steadily as people enter the labor market at a later age. We recognize that our findings are not convincing that endogeneity issues were solved by simply restricting the sample to those who ultimately complete high-school or post-secondary education. Nonetheless, we believe that child labor reduces earnings in adult life after applying estimations conditional on the years of schooling completed by the individuals. The results presented in Table 4 reinforce those obtained at using the samples restricted to maximum level of education achieved. Since schooling is the same for all observations in each of the 16 samples used in the estimation (0–15 years of schooling), the endogeneity between the age at which an individual began to work and his/her education is no longer a relevant problem, as bursae is much less likely to occur. According to Emerson and Souza (2011) the impact of entering the labor market is negative for young children (in the sample used) and that negative effects turn positive between the ages of 12 and 14. Specifically, a concave down parabola was observed that assumes a maximum value at the age of 13–14. In our opinion, even taking into account that child labor occurred several years before, this age is very low. It is important to consider that no control was applied to the number of hours worked in the main job. In this regard, it is interesting to observe, in Fig. 1, the averages of the logarithm of earnings and number of hours worked per week in main job according to the age at which one started to work (7–25 years old) for Brazilian males aged from 25 to 55 years old. This figure shows that the number of hours worked per week decreases as the age at which an individual enters the labor market increases. In the earnings equation, one can clearly see that controlling for the number of hours worked by an individual today (i.e. in his or her adult life) is key, especially if the objective is that of isolating the effect of working at an early age in the past. However, this figure may lead one to think that it depicts an unreal situation, due to differences in the number of individuals in each first job age group. Indeed, the proportion of children who start working before the age of 14 today is relatively small as compared to later age groups. The same can be said for people who start working after they are 25 years old in relation to those who do so at the age of 14–18, for example. One should consider, however, that the information about the first job age is related to the past of individuals who are at least 25 years old. Table 1 shows the proportion of Brazilian individuals by the age at which they started to work, gender, and survey (2001–2009 and 2011). The high rates of men in the 25–55 age intervals who started working before they were 14 years old is not surprising, as child and adolescent labor was common in the past, especially in agriculture.