competition from community schools on public school performance in Poland.1 Commu-nity schools are more autonomous than public schools with respect to teachers’ hiring and salaries and collecting external funds, however, they have to follow a nationwide curriculum. As an identification strategy we use the amendment to the education act introduced in March 2009 which facilitated the creation of community schools but only for schools that have 71 and less students. Therefore, public schools located in areas where there is a higher percentage of students who attend schools with less than 71 students are more exposed to competition. We show that this is indeed strongly related to actual creations of community schools. Using year 2009 as a breakthrough date in the Difference-in-Difference estimations we find that the higher threat of competition caused by the mentioned reform has significant negative impact on the performance of urban public schools. Urban areas are more competitive educational markets than rural areas, with dense school network and better parental background. The effect is similar for public schools that are larger (more than 300 students) and becomes stronger for ur-ban schools that already have a community school in their neighbourhood and may thus be more aware of the consequences of the reform. The generalized model and placebo experiments show it does not result from a pre-existing or concurrent trend either. We argue that in our case changes in students composition between schools are unlikely, but also similarly toHsieh and Urquiola (2006) we analyse changes in test scores at the municipality level, which cancels out the direct effects of sorting between schools. In cities (above 20000 inhabitants) test scores drop at the aggregate community level.
Our paper is most similar to Hoxby (2003, p.32) in that there is an exogenous varia-tion in thethreat of competition, which we exploit. As statesHoxby (2003, p.32) "choice options do not arise randomly, but are frequently a response to school conduct”. Con-sequently, the actual change in competition might be endogenous with respect to the characteristics of educational markets. Therefore, it is the threat of competition that matters. Comparing toHoxby(2003, p.32) we do not discreticize our treatment variable.
We focus on the short run consequences of the reform. In the short run, decisions regard-ing the school might be different than predicted by reputational model of (MacLeod and Urquiola, 2009). If school’s reputation depends on both productivity and on the ability of its students, private schools have the incentive to boost their reputation by cream skimming the best students rather than by raising productivity. Then, increasing choice may result in lower average school productivity. This is consistent with what happened
1Poland has experienced substantial gains in PISA scores, rising from 470 points in 2000 to 518 in 2012, placing Poland fifth in Europe and eleventh in the world.
CEUeTDCollection
in Chile. In 1981 Chile introduced nationwide school choice by providing vouchers to students, resulting in 20 percentage points increase in private enrollment rate. Hsieh and Urquiola (2003) find that the main effect of this expansion is school stratification.
Comparing the change in educational outcomes in urban and wealthier communities to that in communities where private schooling increased by less the authors find no evidence that more competition improved test scores, but that repetition and grades actually worsened. On the contrary, if schools cannot select on ability (e.g. they must select students via lotteries), then their model implies that school choice will unam-biguously raise school performance and student outcomes. It seems that this is what happened in Sweden. In 1991 Sweden introduced a reform which led to the creation of independent municipality fund schools and to greater school choice as students were allowed to attend any school in their jurisdiction. Admissions into elementary level were based on proximity to the school, so cream-skimming was not allowed. Bohlmark and Lindhal(2012) analyse all ninth-grade cohorts from 1988 to 2009 and report an increase in a broad set of outcomes following the increase in competition e.g. test scores in lan-guage and math, grades in lanlan-guage and math, university attendance and average years of schooling at age 24. Sandstrom and Bergstrom (2005) analyse how higher share of independent schools impacts on the students results in the neighbouring public schools.
To account for the endogeneity in the share of independent schools, they use proxies for the general attitude that a given municipality has towards delegating its duties to the private sector. Municipalities more open to the private sector are less likely to block independent schools entry into the market. They find that greater school competition leads to better performance of public school students on standardised exams.
In the short run, school adjustment to increased competition may include activities that are attractive and visible to parents (e.g. school trips), but are not necessarily related to productivity. As Hoxby (2003) states, in the short run principals “(...) can induce their staff to work harder; they can get rid of unproductive staff and programs;
they can allocate resources away from non-achievement-oriented activities (building self-esteem) and toward achievement-oriented ones (math, reading, and so on)”. In Poland, teachers enjoy high level of employment security (ensured by the collective bargaining arrangement, so-called Karta Nauczyciela) and they cannot be laid off easily, in par-ticular, not in the short run. Therefore, what remains available to school principals, is either efficiency changes i.e. incentivising teachers to work harder, or boosting their school’s prestige. We find no evidence for increasing teachers’ salaries or investment in infrastructure in the expenditure dataset. Anecdotal evidence suggest that principals
CEUeTDCollection
may deter the entry of community schools by shifting their attention from tasks oriented at performance of students, to those which are visible to parents (e.g. school trips).
Polish community schools differ from such schools in other countries, but in general community schools have their idiosyncratic aspects (Heers, Van Klaveren, Groot and van den Brink, 2011). In Poland such schools are not targeted at low or high-income students, they operate like regular schools, however they are given a substantial auton-omy when it comes to the management. As such it makes them also similar to charter schools in the USA or autonomy schools in the UK. The main reason for their creation is, however, financial. The move toward greater cost rationalisation of the Polish local governments and the actions taken by the central Polish government to reduce fiscal debt force many gminas(i.e. municipality - NUTS 5 administration unit) to close small elementary schools and move students into bigger establishments. Parents object to the closure of a school because of the increased distance to a new school and the role the school plays in cultural and social life of the local community. Thus closures lead to significant tensions between local governments and citizens. One way to release these tensions is to allow non-public associations (i.e. typically parents’ associations) to take control over a school that is planned to be closed, and create non-public community school instead. Local government provides the community school with a subsidy fi-nanced by the central government (i.e. tied to the number of pupils), but it does not need to cover the remaining costs which can be as high as 50 percent. Community schools are free of charge and are more autonomous than public schools. In particular, they do not need to abide by collective bargaining agreements with teachers’ unions that determine teachers’ salaries. They are also more flexible in acquiring external funds, but they have to follow a standard, nation-wide curriculum.
The threat of handover of a small school to a non-public association creates com-petitive pressure for other local public schools for two reasons. Firstly, since central government subsidy is tied to the student, a public school loses money when a student goes to a community school. For the same reason, liquidation of a small school is ben-eficial for another public school, because students from a liquidated school attend the public school. Secondly, because of the cost-effectiveness of community schools, local politicians may consider the transformation of all public schools into community schools.
Principal-agent problems make it unlikely that principals form a coalition with parents and transform their school into a community school. Principals are more free in their decisions when they respond to a local politician than directly to parents whose kids attend the school. The principals are motivated to influence local politicians to close
CEUeTDCollection
endangered schools and thus block the entry of new schools. Therefore, the ensuing number of liquidations and community schools is endogenous and subject to the degree of competition.
This paper contributes to the literature on more autonomous schools too, although we do not study such schools themselves, but rather their spillover effects. Clark (2009) analyses British reform that allowed public high schools gain more autonomy. He finds no evidence of spillover effects of such schools on the schools in their neighborhood. Machin and Vernoit(2010) study the introduction of academy schools in UK that allowed schools to change their structure and become more autonomous. They report positive spillovers to neighboring schools. There is also related literature on the effects of decentralization in the US (Hoxby,2000;Rothstein,2007). This, however, provides choice between public school districts rather than between private and public schools, so this literature answers a different question and so far has produced mixed results. Similarly for the effects of private voucher-induced competition on public school performance in the USA (Hoxby, 2002;Figlio and Hart,2014;Abdulkadiroglu, Angrist, Hull and Pathak, 2014).
The paper is organised as follows. In Section 2 we describe the educational system in Poland with particular emphasis on community schools. In Section 3 we present empirical strategy and data. Section 4 contains the results and a series of robustness checks. In Section6 we discuss the results and conclude.