The Dynamic Effects of Educational Accountability

Abstract

This paper provides the first evidence that value-added education accountability schemes induce dynamic distortions. Extending earlier dynamic moral hazard models, I propose a new test for ratchet effects, showing that classroom inputs are distorted less when schools face a shorter horizon over which they can influence student performance. I then exploit grade span variation using rich educational data to credibly identify the extent of dynamic gaming, finding compelling evidence of ratchet effects based on a triple-differences approach. Further analysis indicates that these effects are driven primarily by effort distortions, with teacher reallocations playing a secondary role.

Transfers to Households with Children and Child Development

“…If the programme for children of age t was announced when the child was age t À 1, for example, the household would have an incentive to decrease the ability level of the child at age t in order to make the value-added criterion easier to satisfy. This is the essence of the ratchet effect explored by Macartney (2016) in his school quality application.…”

“…Second, we also limit attention to the case of a programme that exists only for one period. The multi-period case brings the possibility of strategic behaviour into play, as the analyses of Weitzman (1980) and Macartney (2016) demonstrate. While allowing for strategic long-term play by the household could alter the effect of the programme on child development, allowing for this behaviour makes the modelling exercise much more complex and less transparent.…”

Measuring quality for use in incentive schemes: The case of “shrinkage” estimators

“…Although deriving the (fully) optimal estimator for the environments considered in this paper is a technically difficult problem, future research making progress in this area could also quantify the effects of using such an estimator. Macartney (2016) documented that teachers are forward-looking, choosing effort levels in response to current and future incentives. Therefore, another important extension would extend this paper's static framework, where the most natural starting point would likely be the hidden action model, to examine how measurement and optimal remuneration policy would interact to affect the administrator's value in dynamic settings.…”

Performance Standards and Employee Effort: Evidence from Teacher Absences

“…North Carolina first implemented the ABC accountability policy at the start of the 1996 to 1997 academic year (Macartney, 2014). 3 The ABC policy is notable in that it focuses on achievement growth in addition to achievement levels, as schools have relatively more influence on the former.…”

“…Similarly, Reback, Rockoff, and Schwartz (2014) find evidence that NCLB accountability pressure caused schools to shift time away from non-tested subjects such as science and social studies. Even thoughtfully designed growth-based policies that account for initial performance levels are subject to ratchet effects, or dynamic distortions caused by actors making current decisions with the intent of manipulating future targets (Macartney, 2014). There is also evidence of nefarious unintended consequences of high-stakes accountability policies.…”