Evaluating the Relationship between Pay and Research Productivity: Panel Data Evidence from Ontario Universities

Abstract

À l'aide des données rendues disponibles par la Loi de 1996 sur la divulgation des traitements dans le secteur public de l'Ontario, nous élaborons un modèle de données de panel qui contiennent les salaires individuels de professeurs d'économie titulaires d'un poste permanent ou menant à la permanence de seize universités de l'Ontario entre 1996 et 2006, afin d'évaluer la relation entre le rendement scolaire et le salaire. Nos évaluations suggèrent que les universités récompensent la productivité, car la publication de la recherche dans une revue importante est souvent associée à une majoration d'environ 1 à 3 pour cent du salaire annuel. Nous interprétons nos résultats comme une preuve empirique de l'importance possible de la productivité de la recherche dans la détermination des salaires des universitaires à partir du financement public.Mots clés : loi sur la transparence, salaire, rendement, productivité de la recherche, publications Exploiting data made available by the 1996 Ontario Public Sector Salary Disclosure Act, we construct a unique panel data set containing individual salaries of tenured and tenure-track economics professors from 16 universities in Ontario between 1996 and 2006 to evaluate the relationship between academic performance and salary. Our estimates suggest that universities reward research productivity, as a top journal publication is significantly associated with a roughly 1 percent-3 percent increase in annual salary. We interpret our results as empirical evidence of the potential importance of research productivity in determining publicly funded salaries of academics.

Impact of Research on Economics Professors’ Compensation

“…Moore et al (2007) found the number of years since a PhD was obtained to be approximately 16 years, and length of career at current university was approximately 12 years, indicating that there is limited shifting between institutions for professors. Sen et al (2014) found being an associate professor provided a 5% increase in salary, while being a full professor increased salary by 10%. Warman, Wooley, and Worswick, (2006) found significant pay discrepancies between male-female cohorts in academia.…”

“…The two studies conducted on Ontario economics professors, by Sen, Voia, and Woolley (2010) and Sen, Ariizumi, and Desousa (2014), obtained salary information through the Ontario Government's mandated salary disclosure. This allowed for the study of the entire population of economics professors earning over $100,000.…”

“…Naturally, this is a difficult variable to measure; it is challenging to tell if a professor shirks or exceeds in these duties. All models which include a measure for administrative duties, such as Sen et al (2014) and Hamermesh and Pfann (2012), used a similar proxy measurement. Additional responsibilities such as chair or above are flagged, with the logic being that these are an indicator of additional duties which are likely to be rewarded.…”

“…Additional responsibilities such as chair or above are flagged, with the logic being that these are an indicator of additional duties which are likely to be rewarded. Sen et al (2014) only flagged administrative chair positions, excluding any more senior position, while other studies such as Hamermesh and Pfann (2012) flagged all positions at the chair level and above. Across models, the variable had significance at the 1% level.…”

“…Studies often have difficulty finding a proxy for teaching ability and do not include the parameter (Moore et al, 2007). Others use proxies such as receiving a teaching award (Moore et al, 1998) or a survey site such as ratemyprofessor.com (Sen et al, 2014), which have potentially large uncertainties relating to whether they are an appropriate proxy. Of those proxies which are used, they are frequently insignificant (Sen et al, 2014) and if not, the coefficient value is often small.…”