Inequality and wage structure in Mexico: going beyond the mean

Keyword: 
Labor
Topic: 
Labor

The Mexican case has emerged as an interesting outlier in terms of the relationship between changes in wage inequality and schooling premia in the international context.

From 1987 to 2008 the Mexican economy underwent numerous reforms —domestic financial market reforms, capital account liberalization, tax reforms, privatization of state-owned enterprises and labour reforms (Lustig, 2001). Specifically, in this period earnings inequality presented varying trends. The period from 1987 to 1994 was marked by structural reforms and trade and financial liberalization in the economy, which increased relative demand for skilled labour and also increased inequality. Subsequently, the period between 1994 and 2001 was one of growth and relative stability, with an increased supply of skilled workers and a fall in inequality. Moreover, in this period the levels of education showed a marked increase. Between 2001 and 2008, other reforms were introduced.1  Thus, between 1987 and 1994, there was an increase in wage inequality, but then in the period after 1994, and unlike many developed countries, Mexico presented a declining wage inequality.

However, the analysis and evolution of inequality is much more complex than just looking at statistics on wage averages. In particular, it is important to look beyond the average values so as to obtain a complete picture for three reasons: First, because recent studies in other countries using quantile regression techniques have shown that different characteristics have different effects on individuals’ wages at the top of the wage distribution from those recorded at the bottom of the wage distribution; second, because Mexico is a heterogeneous society and, for this reason, the effects of reforms may well be heterogeneous also; and, third, because there is growing evidence from other countries (e.g., the US) that suggests that, far from being ubiquitous, the growth in wage inequality is increasingly concentrated at the top end of the wage distribution (Lemieux, 2008).

In the case of Mexico, changes in wage structure display interesting patterns in terms of wage levels at different parts of the wage distribution between 1987 and 2008. In the table I show the results of applying in the 3 subperiods a decomposition method similar to Melly (2005) that permits to identify the relevance of three factors: characteristics, coefficients and residuals. In particular, I report the estimated variation over time of some selected quantiles (10, 25, 50, 75, 90), and the related decomposition into the three components. The effects for different quantiles show that the differences in characteristics are much more important at the bottom (10th percentile) than at the top (90th percentile) of the wage distribution. Indeed, some significant wage structure effects emerge at the 90th percentile.

Table
Quantile and inequality decomposition in the contributions related to covariates, coefficients and residuals in Mexico

Quantile and inequality decomposition in the contributions related to covariates

The attached figure illustrates not only that inequality due to characteristics of labour force is still significant and that characteristics change contribute in opposite direction to increase or decrease of the inequality, but also it will have important implications regarding the changes in the schooling premia. A glance at the figure immediately convinces us that high heterogeneity in the within component can explain the changes in inequality. In order to provide an interpretation of the within component, which basically underlines two main effects. On the one hand, the positive (negative) changes in the coefficients component exert a positive (negative) impact on the residual component, along the wage distribution, providing a measure for “unmeasured price skills” On the other hand, the residual component is also related positively at the bottom of the distribution between 1987 and 1994 and at the top of the distribution in 2001-2008 and there are offsetting by the covariate or coefficient components in the labor force.2

Figure
Decompositions of differences in distribution, 1987-2008

1987-1994
Decompositions of differences in distribution, 1987-2008

1994-2001
Decompositions of differences in distribution, 1987-2008

2001-2008
Decompositions of differences in distribution, 1987-2008

Decompositions of differences in distribution, 1987-2008

Source: Own elaboration from ENEU-ENOE 1987-2008

Therefore, the interplay between these forces determines the changes in the wage structure at the selected percentiles. The decomposition of the changes in the wage distribution for the period 1987–2008 displays different results of trends in earnings inequality, the increase of wage inequality between 1987 and 1994. And as opposed to many developed countries, wage inequality in Mexico has been falling for the period after 1994.

My estimates suggest that changes both in individuals’ attributes and in the returns to these attributes contributed in different direction to the observed increase or decrease in wage inequality over time. Besides, the contributions of both changes are variable in magnitude as per the different portions of the wage distribution are considered. The arguments put forward concerning the importance of that rising education leads to lesser wage inequality. The analysis indicates that, contrary to this, in Mexico increases in educational levels do not necessarily translate into a more equal wage distribution.

Even though the levels of education enlarged very rapidly, educational inequality is the variable that accounts for, by far, the largest share of wage inequality in Mexico. There can be substantial heterogeneity among workers of each type of level education. The marginal contribution of education to the explanation of inequality in Mexico is almost equal to the joint contribution of other relevant variables such as occupation, economic sector, firm size and urban areas. It is worth pointing out that the difference between the marginal contributions has been increasing over time, indicating that, as the economy progresses, education becomes even more important in determining the choices of sectors, occupations and firm size.

Taking all this into account, our estimates also suggest that changes both in individuals’ attributes and in the returns to these attributes contributed in different directions to the increases and decreases observed in wage inequality over time. Moreover, the contributions of the two changes varied in magnitude depending on the section of the wage distribution under analysis. It is claimed that raising education serves to reduce wage inequality; however, our analysis, by contrast, suggests that in Mexico increases in educational levels do not necessarily result in a more equitable wage distribution. Thus, even though educational levels rose rapidly and educational inequality is the variable that accounts for the largest share, by far, of wage inequality in Mexico, there remains considerable heterogeneity among the workers in terms of their educational level.

It is overwhelming documented that the rising wage inequality in Mexico has been a constant phenomenon in Mexico over the last two decades and, to varying degrees. Moreover the rise in the pay-off to skill is an important part of the reason for the rise in inequality.

In my view, the rate of return of education tends to either fall or remain constant and for those at the top of the wage distributions are reduce. The evidence suggests that there are declining returns to invest in human capital. Consequently, we can observe is whether changes in bottom (or in top) wages represent differences in transitory or permanent wage.

Finally, the current debated of inequality seems to be to and fro. The trends move in opposite directions even in less wealthy countries (such as Mexico, Korea among others). This remains one of the most important challenges for understanding how wage or income inequality may affect other aspects of life in the poor and rich countries.


1. See Campos-Vázquez (2010), Robertson (2007) and López-Acevedo (2006).

2. These results are also consistent with other analysis concerning Europe, such as Barth and Lucifora (2006), which document an increasing trend in within-group wage inequality, especially for tertiary education.

This article is based on Tello, C. Ramos, R., and Artís, M. (2012) “Changes in wage structure in Mexico going beyond the mean: An analysis of differences in distribution, 1987-2008”. IZA DP No. 6576.


References

Autor, D.A, Katz, L.F. and Kearney, M.S. (2005) “Rising wage inequality: The role of composition and prices”, NBER No. 11628.

Barth, E. and Lucifora, C. (2006) “Wage dispersion, markets and institutions: The effects of the boom in education on the wage structure”, IZA DP No. 2181.

Campos-Vázquez, R. M. (2010) “Why did wage inquality decrease in Mexico sfter NAFTA?”, Serie documentos de trabajo No. 15, CEE. Colmex.

Lemieux, T. (2008) “The changing nature of wage inequality”, Journal of Population Economics, 21: 21-48.

López-Acevedo, G. (2006) “Mexico: two decades of the evolution of education and inequality”, World Bank Policy Research, WP No. 3919.

Lustig, N (2001) “Life is not easy: Mexico quest for stability and growth”, Journal of Economic Perspectives, 15: 85-106.

Melly, B. (2005) “Decomposition of differences in distribution using quantile regression”, Labour Economics, 12(4): 577-590.

Robertson, R. (2007), ‘Trade and wages: two puzzles from Mexico’, The World Economy, 30: 1378-1398.
 

Share this