There is an extensive consent that good teaching has a positive effect on students' cognitive, social and emotional growth. Studies have shown that effective teachers improve student achievement (Hanushek, 2011; Leigh, 2010). No other attribute of schools comes close to having this much influence on student achievement. It is not surprising therefore, that it is suggested that educational systems should make efforts to recruit high-quality candidates and ensure they remain in the profession for many years (Barber & Mourshed, 2007).

One important indicator proposed for measuring teacher quality is human capital. It is assumed that higher education provides a broad foundation of professional knowledge, academic skills and advance methods in both discipline content and pedagogy, which in turn make a positive contribution to student achievements (Darling-Hammond, 2004). Some studies support this assumption (Clotfelter, Ladd, & Vigdor, 2007). Others found no relationship between teachers' academic education and student achievements (Hanushek, 2011). Ingersoll (2005) claims that the effect of teachers' education on student achievement is significant, however the main problem is that many teachers actually teach fields of study which they were not trained to teach. It is also mentioned that critical characteristics of teacher education predicting student achievement are the academic institutions' selectivity and prestige rather than academic degrees teachers attain (Harris & Sass, 2011).

Although  findings concerning the effect of teachers' education on student achievement are inconclusive, the educational system aspires to absorb educated and talented teachers. The challenge to maintain these teachers within the system, however, becomes complicated in light of the high proportion of dropout among teachers each year, especially among good teachers (Ingersoll, Merrill & May, 2012).

The reward-resource model (Sorensen and Tuma, 1981) assumes that employees make rational economic decisions by systematically assessing the benefits and costs of remaining in an occupation. They usually examine the relationship between the job rewards (e.g. salary, promotion opportunities, and autonomy) and their personal resources (e.g. human capital, experience etc.). Therefore, educated and experienced employees who receive high wages or hold prestigious or leadership positions are expected to have low rate of job shift as a result of the benefits they gain through having a "good" job.

The relationship between teachers' education and teachers' attrition received limited attention in the professional literature. So far common outcomes on this regard have indicated that higher academic achievements of teachers increase the likelihood of leaving teaching (Guarino, Santibanez & Daley, 2006), mostly among those who studied mathematics and science (Borman & Dowling, 2008). Human capital is a resource with economic value in the labor market. It demonstrates knowledge, expertise and professional skills. People expect to have high economic rewards in return for this investment, especially in professional, attractive occupations. Many evidences revealed that teacher' salary in developed countries is significantly lower than other professional employees' salary (Ladd, 2007). This trend is one of the main causes of teacher attrition (Ingersoll & May, 2012). Moreover, other evidences specified that teachers hardly get professional support and autonomy, indicators that also increase teacher attrition (Skaalvik & Skaalvik, 2010). Thus, it seems reasonable to assume that the balance of rewards and resources is violated in the case of teacher as their educational credentials increases.

According to the reward-resource model, this study intends to explore the relationship between human capital and teacher attrition in the Israeli educational system. Human capital will be defined by three dominant features: 1. academic degree, 2. academic institution, 3. field of study. The effect of the relationship between human capital (resources) and income components (rewards) on the likelihood to leave the teaching profession will be also estimated.

Context: Teaching in Israel is highly female-dominated occupation. It is associated with high level of job security, option of part time work and long vacations. Most teachers are employed by the state, and usually receive tenure after 3 years; afterwhich they are protected and cannot be easily dismissed. On the other hand, Teachers' salary is rather low (Addi-Raccah, 2005) and promotion channels are limited (Avidav-Ungar & Arviv-Elyashiv, in press). Teaching credentials requires academic degree and a teaching certificate. It also comprises participation in internship program in the first year of teaching. Teacher attrition is high in the beginning of the career, around 16%, in the first year of teaching and after five years it increases to 30% in average (Arviv-Elyashiv & Zimmerman, 2015). Participants: The study is based on a sample of 10,340 Israeli teachers teaching in the Jewish and Arab sectors at all K-12 levels, who commenced their first year of work in 2000 (3,183 teachers), 2003 (3406 teachers), or 2005 (3,751 teachers). The database includes yearly follow-ups of their career until 2010. The sample is based on the Israeli Central Bureau of Statistic (CBS) records, and it includes demographic information, educational data as well as employment characteristics and schools features. This sample includes 50% of the teachers who began their first year of work during this period. It represents the teachers' population in the official state schools of the Hebrew secular sector. Analysis: The data was analyzed by descriptive statistics, independent t-tests and 2 tests. A logistic regression was estimated to examine teachers' likelihood to leave teaching (at the first five years) in comparison with staying in the profession – focusing on the effect of the human capital components (academic degree, academic institution, field of study.) on teacher attrition.

The results indicate that high level of human capital increases teaching attrition. Teachers who have acquired advanced degrees, teachers who have studied in selective academic institutions (universities) and teachers who have acquired higher education in a professional field of study (e.g. economy, law) or in mathematics and science are more likely to leave teaching compared with teachers who have acquired first degrees or teaching certificate, teachers have studied in educational colleges and those who have studied education and teaching, respectively. We also examined the effect of the interaction between teachers' academic degree and their salary on teacher attrition. The results show that high level this component decrease the likelihood to leave teaching. Among highly educated teachers, higher salary increase teacher attrition. This finding is surprising at first glance, but it is reasonable to assume that, as teachers' salaries are lower than other professional employees, educated teachers have attractive alternatives to earn higher salary outside the teaching profession; therefore, increasing their salary relative to their less educated counterparts is not enough. It seems that job shifting decisions among teachers, in accordance with the reward-resource model involve a rational calculation of evaluating the gap between the personal resources they bring with them and the rewards they gain for their efforts. Indeed, teachers with high level of human capital are more likely to drop out. In light of the vast consent on the effect of quality teachers, these findings raise the question: Does high educated teacher attrition mean a loss of good teachers? The abandonment of educated personnel may have negative consequences on other aspects of the quality of teaching, among other things a possible downgrading of teachers’ status and the entry of teachers of low quality and low cultural capital.

Adi-Raccah, A. (2005). Gender and teachers’ attrition: The occupational destination of former teachers. Sex Roles, 53(9/10), 739-752. Arviv-Elyashiv, R. & Zimmerman, V. (2015). Which teachers are liable to drop up? Demographic and institutional characteristics of teaching dropouts. Dapim, 59, 175-206 (Hebrew). Avidav-Ungar, O & Arviv-Elyashiv, R. (in press). Teacher perceptions of empowerment and promotion during reforms. Educational Management. Barber, M. & Mourshed, M. (2007). How the World's Best-Performing School Systems Come Out on Top, McKinsey & Company, Social Sector Office. Borman, G.D. & Dowling, M.N. (2008). Teacher attrition and retention: A meta-analytic and narrative review of the research. Review of Educational Research, 78(3), 367-409. Clotfelter, C.T, Ladd, F.L. & Vigdor, J.L.(2006). Teacher-Student matching and the assessment of teacher effectiveness. Economics of Education Review, 26, 673-682. Darling-Hammond, L. (2003). Keeping good teachers: Why it matters, what leaders can do? Educational Leadership, 60(8), 6-13. Guarino, C.M., Santibanez, L. & Daley, G. (2006). Teacher recruitment and retention: A review of the recent empirical literature. Review of Educational Research, 76(2), 173-208. Hanushek, E. A. (2011). The economic value of higher teacher quality. Economics of Education Review, 30(3), 466-479. Harris, D.N. & Sass, T.R. (2011). Teacher Training, Teacher Quality, and Student Achievement. Washington: National Center for Analysis of Longitudinal Data in Education Research Ingersoll, R. M.. (2005). The problem of unqualified teachers: A sociological perspective. Sociology of Education, 78(2), 175-178. Ingersoll, R.M. & May, H. (2012 ). The Magnitude, Destinations, and Determinants of Mathematics and Science Teacher Turnover. Educational Evaluation and Policy Analysis, 34(4), 435-464. Ingersoll, R.M Merrill, L. & May, H. (2012). Retaining Teachers: How Preparation Matter? Educational Leadership, 30-34. Ladd, H. (2007). Teacher labor market in developed countries. The Future of Children, 17(1), 203-217. Leigh, A. (2010). Estimating teacher effectiveness from tow-year changes in students' test scores. Economics of Education Review, 29(3), 480-488. Skaalvik, E. & Skaalvik, S. (2010). Teacher self-efficacy and teacher burnout: A study of relations. Teaching and Teacher Education, 26(4), 1059-1069. Sorensen, A.B. & Tuma, N.B. (1981). Labor market structure and job mobility. Research in Social Stratification and Mobility, 1(1), 67-94.