The Impact of Degree Duration on Higher Education Participation: Evidence from a Large‐scale Natural Experiment
| Author | Fabio Berton,Daniele Bondonio |
| Published date | 01 October 2018 |
| DOI | http://doi.org/10.1111/obes.12231 |
| Date | 01 October 2018 |
905
©2018 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd.
OXFORD BULLETIN OF ECONOMICSAND STATISTICS, 80, 5 (2018) 0305–9049
doi: 10.1111/obes.12231
The Impact of Degree Duration on Higher Education
Participation: Evidence from a Large-scale Natural
Experiment*
Daniele Bondonio† and Fabio Berton‡
†Department of Law and Political, Economic and Social Sciences, University of Piemonte
Orientale, via Cavour 84–15121, Alessandria, Italy (e-mail: daniele.bondonio@uniupo.it)
‡Department of Economics and Statistics, University of Torino, Lungo Dora Siena
100–10153, Torino, Italy (e-mail: fabio.berton@unito.it)
Abstract
This paper investigates the effects on enrolment, retention rate and on-time graduation of a
nationwide Bologna Process reform introduced in Italy to establish BAs with a duration of
three years, followed by optional second-tier degrees of two years, in place of single-tier
degrees of four or five years. The analysis exploits exogenous delay of treatment condi-
tions and the unique availability of microdata that cover the universe of the departments.
We estimate that the reform boosted first-year enrolments by 14.5–17.3 percentage points,
compared to a counterfactual status of no reform. This enrolment shift was due to partici-
pation gains rather than substitution effects, and it is likely to have persisted in the longer
term. Moreover, no trade-off between increased participation and deteriorated retention
and on-time graduation emerged.
I. Introduction
Higher education is recognized as a strategic growthand development factor in all advanced
economies. Nonetheless, in OECD countries the share of the population with tertiary
educations is less than 35% (OECD, 2016), and policies aimed at increasing participation
in higher education maintain a prominent role on domestic political agendas. In many
European countries, the adoption of such policies stemmed from the implementation of
the so-called Bologna Process, a joint initiative by29 European Ministers of Education that
met in Bologna in 1999 and designed a new standard for university education in Europe
JEL Classification numbers: C21, I21.
*The data used in the analysishave been provided by the former Italian National Committee for the Evaluation of the
University System (MIUR-Ministry of Education, Universityand Research). We thank the Statistical office of MIUR,
the National Committee for the Evaluation of the University System and Matilde Bini for assistance in acquiring
the data Wealso thank the editor, Prof. James Fenske and three anonymous referees for their helpful comments, and
Giliberto Capano and Giunio Luzzatto for the information on the institutional process that accompanied the Bologna
Process in Italy.Funding was provided by the University of Piemonte orientale. The usual disclaimers apply.
906 Bulletin
comprising a two-tier system based upon three-year long BA programmes, and two-year
masters (MAs).
In a number of countries (including Albania, Austria, Croatia, Bosnia-Herzegovina,
Estonia, Hungary, Italy, Germany, Norway, Portugal, Romania, Spain and The Nether-
lands), the adoption of the new Bologna Process standard entailed the transformation of
the university system from a prevailing one-tier degree system with a duration of four to
six years (D’Hombres, 2007). In these countries, and notably in Italy and Germany, the
purpose of breaking down the excessively long one-tier programs was also to increase
participation in tertiary education by reducing the risk of students dropping out without a
degree or graduating many years after the already long statutory duration of the degrees
(Perotti, 2002; Hahm and Kluve, 2016).
For these reasons, providingreliable empirical evidence on the impacts of such policies
is of crucial importance. Yet it is not an easy task. Previous attempts to assess the im-
pact of the Bologna Process have suffered from identification drawbacks due to non-ideal
institutional frameworks and data limitations. The bulk of the existing literature focuses
on Germany, Portugal and Italy. The empirical literature on Germany (Horstschr¨aer and
Sprietsma, 2015; Hahm and Kluve, 2016; Lerche, 2016) has had to face a problem of
self-selection by departments into the reform.1This circumstance has made adoption of
the Bologna Process standard endogenous to the specific underlying trends that mayaffect
the participation and educational outcomes of interest in any given department. Similar en-
dogeneity issues have emerged also in the empirical literature on Portugal (Cardoso et al.,
2008).2
In the case of Italy, much more favourable identification conditions have been in place
also due to the exogenous timing of the reform’s adoption and the occurrence of some
large-scale natural experiment conditions in terms of an unintended delay-of-treatment
mechanism. The existing empirical literature on Italy, however, has not fully exploited
these favourable conditions due to data limitations: Boero, Laureti and Naylor (2005) and
Bratti, Broccolini and Staffolani (2006) examined only a small sample of a few depart-
ments; Cappellari and Lucifora (2009), Di Pietro (2012), Di Pietro and Cutillo (2008)
and D’Hombres (2007) focused on the future labour market outcomes of repeated cross
sectional samples of high-school graduates; while Bosio and Leonardi (2011) analysed the
repeated cross-sectional samples of individuals included in the Italian labour force survey.
For the purposes of this paper, instead, we gathered unique administrative microdata,
at the department level, which cover longitudinally the universe of all existing Italian
universities, with degree-specific information on the number of enrolees and graduated
students sorted by the year of their first enrolment. Using such data, we estimate the impact
on enrolment, retention and on-time graduation of the adoption of the Bologna Process
standard in Italy by fully exploiting a number of favourable identification conditions that
can be summarized as follows.
1The Bologna Process was introduced by some German departments as early as in the year 2000. In Germany,
however, the adoption of the new standard was not ruled as mandatory. As a result, by 2008 14% of German
departments still offered curriculum degrees entirely within the pre-Bologna system (Horstschr¨aer and Sprietsma,
2015), and it took around 10 years for a German department to complete the transition to the new standard for all its
degree curricula (Hahm and Kluve, 2016).
2In Portugal, the adoption of the BolognaProcess was compulsory for all universities, but no deadline was enforced
to switch to the new two-tier system.
©2018 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd
To continue reading
Request your trial