Emiliana Vegas, Lant Pritchett and William Experton*
The World Bank
1818 H Street, NW
Washington, DC 20433
In this paper, we compare teachers’ salaries to those of workers in other occupations in Argentina. We ask, do teachers in Argentina fare well or poorly in the labor market as compared to observationally-equivalent workers in other fields?
Our findings suggest that a national pay raise, or even a national policy on teacher compensation, may not be the best policy for attracting and retaining qualified teachers in Argentina. Our results call for adjustments by province in the real relative salaries of teachers, so that the profession may be attractive to highly qualified individuals in all areas. Our findings also highlight the need for changing the uniformity in the structure of rewarding seniority in teaching.
In March of 1997 Argentina’s teachers unions put up a tent in front of Congress to demand across-the-board increases in teacher salaries. Two years later, the tent is still up. Teachers have been protesting against a proposal by the National Ministry of Culture and Education (NMCE) to reform the teaching career, including changing the structure of teacher compensation. In March of 1998, the NMCE publicly announced a proposal to levy a new tax on transportation vehicles to finance an increase in teacher salaries. Congress approved the proposal in September of that year, though it is still unclear how the salary increase will be implemented.
In this period of intense national debate, more information is needed about whether the rewards to the teaching profession are sufficient to attract, retain, and motivate the highly qualified individuals that are needed to implement the ambitious education reform agenda that the country launched in 1994. In particular, more information is required to answer two questions. First, is the level of the total package of monetary compensation of the profession (e.g. wages, benefits, and pension) and non-monetary rewards (e.g. social status, work flexibility, intrinsic rewards) adequate to attract people with the necessary skills to become high-quality teachers? Second, does the structure of the total compensation package lead those attracted to the teaching profession to perform to the best of their abilities and continue to develop their professional capacity?
Because teachers are key in the education process, and because their salaries often account for (at least) three-quarters of the education budgets of Argentine provinces, these are very important questions. But, they are also very complex questions, to which we seek to make a modest empirical contribution. In this paper, we compare teachers’ earnings to those of workers in other occupations. We ask, do teachers in Argentina fare well or poorly in the labor market as compared to observationally-equivalent workers (e.g. same gender, age, educational attainment) in other fields?
The paper is organized as follows: in Section 2, we discuss our data and methodology. Section 3 contains a brief overview of the Argentine education system and the structure of teacher compensation. In Section 4, we present our analysis of the teacher profile for the entire nation. Section 5 contains our findings of the relative teacher earnings analysis by metropolitan area. In Section 6, we discuss our findings and their policy implications, especially regarding the structure of teacher compensation. In Section 7, we present some conclusions and suggestions for future research.
A limitation of our analysis is the use of teachers wages as the only indicator of the adequacy of compensation. While we recognize that wages are but one indicator of teacher compensation, unfortunately, data on other aspects of teacher compensation (monetary and non-monetary, such as pension and benefit plans as well as flexible schedules and greater job security) are unavailable in Argentina at this time. It is usually argued that the non-salary components of the teacher compensation package in Argentina constitute an important factor in making the profession attractive. Better data on the full compensation package of teachers and other workers in Argentina is needed to thoroughly understand the dynamics of teacher incentives and the teacher labor market.
Our paper is only a first step toward understanding the influence of teacher compensation on who chooses to enter and stay in teaching in Argentina. To thoroughly understand the dynamics of the teacher labor market in Argentina, other indicators are necessary. For example, to better assess whether the profession is attracting sufficient highly qualified individuals, we would need to analyze the enrollment and flows into and out of teacher training institutes. Similarly, to learn whether the profession is able to retain sufficient numbers of highly qualified teachers, we would need to study the turnover rates (taking into account the effects of benefits and pension schemes on teachers’ decisions their positions versus simply requesting leaves of absence). Also, a key question left unanswered here is how pension and other non-salary benefits affect who becomes and stays in the teaching profession. The data available were inadequate to address these very important issues.
Nevertheless, our paper contributes to inform the debate on teacher earnings and working conditions in Argentina. By using complementary sources of information, the paper constitutes a useful step in characterizing the earnings structure and the relative earnings of teachers in Argentina, as well as the (often ignored) variation in relative earnings across provinces.
2.Data And Methodology
We use two datasets: the 1994 National Census of Teachers and the October 1997 Encuesta Permanente de Hogares (EPH), a household survey conducted in Argentina’s metropolitan areas. The census contains comprehensive information on all of the nation’s teachers, their characteristics and those of their jobs. However, it does not contain any information on earnings. The EPH contains information on general worker characteristics for individuals of all ages and occupations, including gender, hours worked per week, educational attainment, age, experience, and earnings. However, the EPH’s occupational categories are broad; one cannot directly identify teachers. Instead, one can only identify individuals who report that their occupation is in education. In addition, the EPH does not contain specific information on teachers that is of interest to answer our research question, such as whether they work in the private or public sector. Because both data sets have advantages and shortcomings, we draw on both sources in addressing the different components of our research question.
Our empirical strategy includes first analyzing the national-level data from the Census and presenting tables that describe the characteristics of teachers and their profession in Argentina. Given that the education system is decentralized and that there may be important regional variations in teacher labor markets, we use the EPH to analyze how teachers in the nation’s main metropolitan areas fare in terms of earnings relative to comparable workers in other occupations (see the Technical Appendix for a presentation of our linear regression model and a description of the variables used in our analyses).
One must recall that in Argentina teachers are also paid during the three months of annual vacation. One of the limitations of our data is that they do not allow us to account for this difference in compensation between teachers and non-teachers. Hence, when in our analyses we compare teachers’ reported hourly earnings to those of observationally-equivalent workers in other occupations, we may underestimate teachers’ earnings.