Content uploaded by Aleksander Aristovnik
Author content
All content in this area was uploaded by Aleksander Aristovnik
Content may be subject to copyright.
AE Relative Efficiency of Higher Education in Croatia and Slovenia:
An International Comparison
Amfiteatru Economic
362
RELATIVE EFFICIENCY OF HIGHER EDUCATION IN CROATIA
AND SLOVENIA: AN INTERNATIONAL COMPARISON
Alka Obadić1 and Aleksander Aristovnik2∗
1) University of Zagreb, Zagreb, Croatia
2) University of Ljubljana, Ljubljana, Slovenia
Abstract
The article measures the relative efficiency of government spending on higher education in
selected new EU member states (with special focus on Croatia and Slovenia) in comparison
to selected OECD countries. The article applies a non-parametric approach, i.e. data
envelopment analysis (DEA), to assess the relative technical efficiency of higher education
across selected countries. When estimating the efficiency frontier we focus on measures of
quantities outputs/outcomes. The results show that the relatively high public expenditure
per student in Croatia should have resulted in a better performance regarding the
outputs/outcomes, i.e. a higher rate of higher education school enrolment, a greater rate of
labor force with a higher education and a lower rate of the unemployed who have tertiary
education. On the other hand, regardless of the input-output/outcome mix, the higher
education system in Slovenia is shown to have a much higher level of efficiency compared
to both Croatia and many other comparable new EU member states and OECD countries.
Keywords: public expenditure, efficiency, higher education, data envelopment analysis,
Croatia, Slovenia, new EU member states, OECD
JEL Classification: H52, I21, I23
Introduction
The review of different empirical evidence (Norman, 1998, p. 129) on the relationship
between education and economic growth rates or income levels show that education
influences economic growth. Most evidence comes from cross-section regression analysis
on samples of developing and/or OECD countries, though increasingly times-series testing
on individual (of groups of) countries are being pursued. Several studies have found that
countries with more educated labour forces tend to grow faster, other thing equal; other
studies have failed to find a significant education-growth relationship. Results also vary
widely in terms of magnitude. The reasons for this are unclear, but may be related to
differences and inaccuracies in the educational datasets (Keller, 2006; Norman, 1998).
∗ Corresponding author, Aleksander Aristovnik - aleksander.aristovnik@fu.uni-lj.si
The Knowledge Based Economy:
Implications for Higher Education in Economics and Business AE
Vol. XIII • No. 30 • June 2011 363
Denison (1985) estimates that typically a quarter of the growth of output per person arises
from increases in educational attainment. Higher education plays vital role in driving
economic growth and social cohesion. Greater investment in universities increases the
quality and quantity of highly educated graduates. The commonly held perception of
universities as merely institutions of higher learning is gradually giving way to the view
that universities are important engines of economic growth and development. Universities
not only generate new knowledge through primary research, they also provide technical
support and specialised expertise and facilities for on-going firm-based research and
development (R&D) activities. Academic research and development is now seen as one of
the key drivers of economic growth. Countries that have academic institutions performing
large amounts of R&D are more able to attract and grow technology oriented companies.
The most comprehensive evidence from cross-section regression comes from Barro and
Sala-i-Martin, who finds that public educational expenditures significantly improve growth
performance and confirm a positive role (Barro, Sala-i-Martin, 1995).
The attainment of an upper secondary education has become the norm in most countries
today. In addition, the majority of students are graduating from upper secondary programs
designed to provide access to tertiary education, in turn leading to increased enrolments at
this higher level. Countries with high graduation rates at the tertiary level are also those
most likely to develop or maintain a highly skilled labor force (OECD, 2009a, p. 64). The
emerging knowledge-based information society requires a large supply of highly skilled
people. There is strong demand for tertiary graduates (especially in the fields of science and
engineering, along with other fields like languages and economics) in the economy. The
characteristics of the higher education (HE) sector make it difficult to measure efficiency: it
does not make a profit; there is an absence of output and input prices; and higher education
institutions (HEIs) produce multiple outputs from multiple inputs (Johnes, 2006, p. 273).
The HE sector, however, has characteristics which make it difficult to measure efficiency:
it is non-profit making; there is an absence of output and input prices; and HEIs produce
multiple outputs from multiple inputs (Johnes, 2006, p. 273). This article tries to assess the
relative efficiency of government spending on higher education in selected new EU
member states and OECD countries, with special focus on Croatia and Slovenia. In this
respect, the efficiency of higher education systems is computed using the non-parametric
approach of data envelopment analysis (DEA) to capture the different dimensions of those
systems and to measure their relative efficiency. The performance of higher education is
measured by how well it transforms inputs into outputs. This is the first time DEA
estimations have been used to measure the performance of HE systems in these two
countries on the macroeconomic level by using a wide range of inputs and
outputs/outcomes.
The article is divided into four main parts. After introductory part, second part analysis
higher education systems, their expenditures and outcomes in selected new EU member
states and OECD countries. The third part explicates methodology of data envelopment
analysis for measuring higher education achievements. Research results of efficiency
effects of higher education attainment in selected countries are presented in the fourth part
od the research. Conclusions regarding the efficiency of the Croatian and Slovenian higher
education in comparison to new EU member states and OECD countries are drawn in final
section.
AE Relative Efficiency of Higher Education in Croatia and Slovenia:
An International Comparison
Amfiteatru Economic
364
1. Descriptive Analysis
Croatian GDP per capita has been relatively low compared to Slovenian or other European
countries. One of the many explanations of this difference could be the effectiveness and
efficiency of the country’s education system. From this perspective, universities generate
spill-over effects from their academic research and teaching, thereby stimulating economic
growth (Audretsch and Lehmann and Warning, 2003). Indeed, the close nexus between the
university system and economic growth has seen significant attention being paid to the
efficiency and quality of Croatian universities. The majority of Croatian and Slovenian
universities are government-owned and largely funded by the Ministry of Education and
Science1. Universities are autonomous bodies established by legislation allowing
considerable freedom in their activities. The next section describes the Croatian and
Slovenian tertiary systems in more detail.
1.1 The Higher Education Systems of Croatia and Slovenia and Its Expenditure
Higher education (HE) institutions in Croatia encompass universities, polytechnics and
schools of professional higher education. Universities may include faculties and academies
of arts as legal entities, and may establish a number of other constituent units (departments,
institutes etc.). In contrast, polytechnics and professional higher education schools may not
establish other TE institutions (MoSES, 2007, p. 33). There are seven public universities
and two private universities and 16 private two-, three- or four-year colleges, polytechnics,
or academic programs. The central government funds public higher education, although
management is fully decentralized to the level of individual institutions (WB, 2008a, pp.
107-109). On the other hand, the higher education system in Slovenia is currently based on
four universities with 49 faculties, three art academies or professional colleges, and 30
individual higher education institutions generally established as private institutions. The
funds for financing academic activity are allocated from the national budget as aggregate
funds for a university or an independent higher education institution (integral financing)
and take into consideration the field of study and the numbers of enrolled students and
graduates from regular first- or second-degree studies (MHEST, 2010a).
Education expenditure in both countries is financed by two distinct types of funding: public
funding (public expenditure) and private funding. In all EU countries, public financing
accounts for at least 75% of education expenditure when taking all education levels
together (Eurostat, 2009, p. 129). However, since the early 1980s changes in the direction
of diversified sources have been observed, with an emphasis on student contributions (Bevc
and Uršič, 2008, p. 233). Namely, higher education has expanded and today is in need of
better quality. The OECD believes that graduates should contribute to the cost of their
tuition – balanced by measures to support students from poor backgrounds (OECD, 2006).
Notwithstanding, a higher student contribution in higher education expenditures the total
number of higher graduates has grown in the EU-27 since 2000 by 35% or 4.3% per year
and hence twice as fast as the general student population. Of course, one reason for this is
the Bologna Process, with a higher share of students taking second degrees (European
1 Observing the higher education institutions as a whole, the ratio of public funds used exceeds 70%,
in extreme cases – mainly in Scandinavia – it can reach even 97-98%. It is fair to ask why the state
finances universities and colleges to such a high extent, that is, why the state should have a role in
higher education (Tóth, 2008, p. 79).
The Knowledge Based Economy:
Implications for Higher Education in Economics and Business AE
Vol. XIII • No. 30 • June 2011 365
Commission, 2009, p. 59). The overall growth in graduates was particularly strong (over
10% per year) in some selected new EU member states (Romania, the Czech Republic and
Slovakia) in the 2000-2007 period (Figure no. 1).
The major inputs for education and higher education in EU and OECD countries come from
public expenditures. Public expenditures for higher education in Croatia are less than those
in the EU-27, selected new EU member states, OECD countries and Slovenia, which are
nearly the same. Private spending on education in Croatia accounts for around 0.75% of
GDP compared with ratios of around 0.4% in the EU-15 and EU-25. Despite the relatively
high private spending on education there are very few private schools, although there is a
growing number of private pre-school providers. School enrolment at the higher education
level in Croatia is almost half that seen in Slovenia, but relatively close to the selected new
EU member states average. Although the completion rates are low, in 2008 the number of
higher education graduates in Croatia was higher than in Slovenia (Table no. 1). While the
number of graduates is rising, there is still a mismatch between skills demanded by the
market and the skills produced by the education system (World Bank, 2008, p. 104).
Figure no. 1: Tertiary graduates in selected EU-27 and Croatia (by ISCED levels 5
and 6 per 1,000 population aged 20-29/25-34), 2000-2007
Note: * estimates
Source: Eurostat according to the European Commission (2009).
Overall public expenditure on education as a share of GDP in both countries is comparable
with the EU average. Slovenian government expenditure on higher education has shown a
positive trend in recent years, with nominal expenditures tending to increase faster than the
inflation rate. The total amount of government expenditure rose by 5.9% in 2005 and 7.2%
in 2006, while the amount of funds for educational purposes went up by 6.4% and 8.4%,
respectively, in the same years (Tajnikar and Debevec, 2008, p. 290). By contrast,
AE Relative Efficiency of Higher Education in Croatia and Slovenia:
An International Comparison
Amfiteatru Economic
366
expenditure at the higher education level in Croatia is far behind that in the selected new
EU member states and OECD countries. The Croatian higher education system currently
has too little by way of assured financial funds compared to European standards. The
amount of outlays on tertiary education as a percentage of GDP in 2007 was 0.81%,
namely, much lower than the EU average (1.3%).
Table no. 1: Higher Education Indicators – Expenditure, Output and Outcomes
in Croatia, Slovenia, New EU Member States and OECD in 2007
Total Public
spending on
education (%
of GDP)
Public
Expenditure
on Higher
Education
(% of GDP)
School
Enrolment
Tertiary (%
gross)
Graduates of
Tertiary
Education (25
to 29)
Population
with tertiary
education
(ISCED 5-6)
aged 25-39
Croatia 4.1 0.8 44.1 20.7 18.2
Slovenia 5.2 1.0 88.0 20.1 21.0
EU-27 5.0 1.1 67.0 38.2 23.2*
OECD
average 5.2 1.0 72.0 38.0 26.1*
Selected
new EU
member
states
average**
5.1 1.1*** 51.1 43.3*** 24.5
Note: * Figure for 1999-2007 average. ** Selected new EU member states - Czech Republic,
Bulgaria, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia.*** Selected new EU member
states – Czech Republic, Estonia, Hungary, Poland, Slovakia.
Source: Eurostat (2010b); OECD (2009a), OECD (2010), UNESCO (2010); World Bank
(2010).
The main characteristics of Croatian education financing are: chronic under-funding, a lack
of equity and transparency in budgetary allocation, an unbalanced structure of the education
budget in terms of categories of expenditure and source of funds, and a lack of synergy
(legislative, professional and institutional) for system change. The 4.1% of GDP share of
total education expenditure in 2007 is well below the European average (5.0%), and the
current level of funding is insufficient to support the reform process. Physical conditions
vary widely from institution to institution, but facilities are often inadequate (OECD, 2001).
Conversely, the main objective of financing higher education institutions in Slovenia is to
implement the goals of the national higher education program, along with respecting these
institutions’ autonomy in terms of the independent formulation of their institutional strategy
and how they define the ways to achieve the set goals. The mechanisms of financing using
public funds should enable higher education institutions to independently adopt decisions
on expenditure and sustainable asset management. An important mechanism for ensuring
the financial autonomy of higher education institutions is the integral (lump sum) financing
of their academic activity. In the future, an internationally comparable share of GDP will
have to be appropriated for the higher education and scientific and research activity of
higher education institutions in Slovenia, meaning that the total funds allocated to higher
education activity will have to rise (MHEST, 2010b). In this context, at least 1.3% of GDP
from the budget and 0.3% of GDP from other sources is planned to be provided for higher
The Knowledge Based Economy:
Implications for Higher Education in Economics and Business AE
Vol. XIII • No. 30 • June 2011 367
education in Slovenia by 2015, and a total of 2.5% of GDP by 2020, of which 2.0% of GDP
would come from the budget. At the same time, a new system of financing higher education
would be introduced, consisting of a basic and a development pillar as of 2011.
Another input of the higher education system in Croatia and Slovenia could be seen from
public research and development (R&D) expenditures as a percentage of GDP. The lowest
expenditures for R&D in the 1999-2008 were in the selected new EU member states (below
0.8% of GDP). Similar situation was in Croatia, especially from 2006 when expenditures
for R&D diminished. The state sector is largely dominant sector in the Croatian R&D
system, especially comparing to research potentials in the business sector (MoSES, 2007, p.
55). With expenditures around 1.5% of GDP for R&D, Slovenia stands much better and is
catching EU-27 average. The United States, with public investments in R&D higher than
2.6% of GDP, are with good reason labelled „knowledge-economy“ (Figure no. 2).
Figure no. 2: Research and development expenditures (% of GDP)
Source: Eurostat Database.
1.2 Outcomes of higher education system
The basic assumption is that higher education systems are multi-product organizations
which “produce” at least two different outputs – research and teaching – using multiple
inputs. Generally accepted outputs of the higher education production process are the
number of graduates of tertiary teaching as a proxy for teaching and the number of
publications as a proxy for research (Warning, 2004, p. 396). Number of higher education
graduates by universities and single higher education institutions by ISCED 5 and 6 levels
in some selected EU countries, Croatia and Slovenia is visible in Table no. 1. In last ten
years there was a significant increase in the number of enrolled and graduates students in
Croatia compared to Slovenia (Figure no. 3), as also the number of student programmes. In
AE Relative Efficiency of Higher Education in Croatia and Slovenia:
An International Comparison
Amfiteatru Economic
368
Croatia, almost three-quarters of young people who successfully complete secondary
school enter tertiary education. These changes correlate to the enlargement and re-
organization of HE institutions, primarily establishment of polytechnics and schools of
professional higher education (MoSES, 2007, p. 72). In such way, gross enrolment rates for
tertiary level has been improving steadily over the past few years in Croatia, but they are
still significantly lagging behind the Slovenian’s. Gross enrolment at the tertiary level stood
at 47.01 in 2007 compared to 85.47% in Slovenia.
Figure no. 3: Tertiary school enrolment in Croatia and Slovenia, (% gross)
Source: World Bank, 2010.
Regarding outcomes in tertiary education, for example, although gross enrolment was about
46 percent in 2006 compared to around 53 percent in the EU-10 (Jafarov and Gunnarsson,
2008, p. 11), the proportion of graduates in Croatia is not high enough. Further, only one-
third of students at the tertiary level reportedly complete their programs, with an average
completion rate of 6.7 years for four-year programs (World Bank, 2008). Non-completion
rates in tertiary education were also very high, with the Ministry estimating that only one-
third of all those enrolled were completing their courses of study. The serious internal
inefficiencies at the tertiary level do not seem to have diminished in recent years (World
Bank, 2008, p. 114). The number of graduates in TE over the last 10 years has been rising
constantly. A vast majority of students has finished their undergraduate programs (on
average 92.3%), whereas 7.7% of students finished postgraduate studies (5.3% a Master of
Science degree and 2.4% a doctoral degree). The average share of graduates in the natural
sciences was only 4% and has been falling constantly since 1997 (from 4.9% to 2.9% in
2003) (MoSES, 2007, p. 73).
A similar situation can be found in Slovenia where the majority of graduates come from the
social sciences, business sciences and law, accounting for nearly one-half of graduates at
the tertiary education level; as many as 70% of them were women. The smallest number of
graduates was recorded in the fields of science, mathematics and computer science as well
The Knowledge Based Economy:
Implications for Higher Education in Economics and Business AE
Vol. XIII • No. 30 • June 2011 369
as agriculture and veterinary medicine – just 1,255 (slightly less than 7% of all) graduates.
An observation over time of the trend in graduate numbers at the tertiary education level in
Slovenia reveals that this number oscillated around 6,000 in the 1980s and at the start of the
1990s, and then started soaring after 1994. Twelve years ago it exceeded the 10,000 limit.
By 2009 the number of all graduates had doubled compared to 1996 and even tripled
compared to the period before 1990 (SORS, 2010).
An additional outcome of the higher education systems in Croatia and Slovenia could be
seen from total number of researchers. The total number of researchers has remained almost
the same during 1999-2008 years period in Croatia, at around 6,700 researchers making
Croatia lag behind to developed European countries regarding research work force. During
the same period this number has increased for almost 2,600 in Slovenia (Table no. 2).
Table no. 2: Total number of researchers (FTE) for teaching and research
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Croatia - 6,772 6,656 8,572 5,861 7,140 5,727 5,778 6,129 6,697
Slovenia 4,427 4,336 4,498 4,642 3,775 4,030 5,253 5,857 6,250 7,032
Source: Eurostat Database.
2. Methodology
This research measures the relative (technical) efficiency of higher education in Croatia and
Slovenia, as well as in comparison with other selected new EU member states and OECD
countries. Yet the characteristics of the higher education sector make it difficult to measure
efficiency: it does not make a profit; there is an absence of output and input prices; and
higher education institutions (HEIs) produce multiple outputs from multiple inputs (Johnes,
2006, p. 273). Therefore, a performance evaluation of higher education is based on multiple
inputs and outputs and thus regressions based on only one output are unsuitable. To be
precise, a non-parametric frontier analysis, namely data envelopment analysis (DEA), is the
most recent methodology that is commonly used to examine the problems of measuring the
performance of HE institutions (Athanassopoulos and Shale, p. 1997). Therefore, this
research uses data envelopment analysis2 as a methodological tool.
In a multi-output, multi-input production context, DEA provides estimates of the distance
function, which is a generalisation of the single output production function (Johnes, 2006,
p. 274). The DEA literature has meanwhile become one of the success stories of the
operational research area. The estimation of frontier or best practice models found its way
to a large variety of domains application (Kerstens and Woestyne, 2009, p. 1). In terms of
empirical surveys, examples of DEA applications were analyzed in many sectors, as also in
education. Some of the key empirical studies of the performance of higher education
include those about measurement techniques (Worthington, 2001); performance differences
in German higher education (Warning, 2004); comparative efficiency of higher education
institutions in the UK (Athanassopoulos and Shale, 1997); efficiency of economic
2 DEA was developed by Charnes, Cooper, and Rhodes (1978) following work by Dantzig (1951) and
Farrell (1957), and estimates a piece-wise linear production function relative to which the efficiency
of each firm or decision-making unit (DMU) can be measured (Johnes, 2006, p. 275).
AE Relative Efficiency of Higher Education in Croatia and Slovenia:
An International Comparison
Amfiteatru Economic
370
departments at Australian universities (Madden et al., 1997); efficiency investigation of
Canadian universities (McMillian and Debasish, 1997) and Japanese universities
(Hashimoto and Cohn, 1997). Chapple et al. in their paper from 2005, presented evidence
of the relative performance of U.K. university technology transfer offices (TTOs) using
data envelopment analysis (DEA) and stochastic frontier estimation (SFE). U.K. TTO’s are
found to exhibit low-levels of absolute efficiency. There also appear to be decreasing
returns to scale, implying that TTO’s may need to be reconfigured into smaller units.
DEA is a non-statistical and non-parametric approach which makes no assumptions
regarding the distribution of inefficiencies or the functional form of the production function
(although it does impose some technical restrictions such as monotonicity and convexity).
Instead, it uses the input and output data themselves to compute, employing linear
programming methods, the production possibility frontier. The efficiency3 of each unit is
measured as the ratio of weighted output to weighted input, where the weights used are not
assigned a priori but are calculated by the technique itself so as to reflect the unit at its
most efficient relative to all others in the dataset. In a multi-output, multi-input production
context, DEA provides estimates of the distance function which is a generalization of the
single output production function (Johnes, 2006, p. 274).
In the first step, the frontier is drawn up by the efficient units. In the second step,
hypothetical units are generated on the frontier to serve as reference units for inefficient
higher education systems. These reference units are constructed as linear combinations of
the most efficient units on the frontier. All inefficient units are enveloped by the frontier.
On the basis of the empirical production function, in terms of best practice, DEA reveals
those HE systems that are on the efficient frontier. It indicates the level of inefficiency of
each system compared to the efficient systems4.
The DEA method is essentially a linear program which can be expressed as follows:
(1)
subject to
All urk > 0, vik > 0 (2)
where Y is a vector of outputs; X a vector of inputs; i inputs (m inputs); r outputs (s
outputs); n is the number of decision-making units (DMUs), or the unit of observation in a
DEA study.
DEA fits a piecewise linear surface to rest on top of the observations. This is referred to as
the “efficient frontier”. The efficiency of each DMU is measured relative to all other
DMUs, with the constraint that all DMUs lie on or below the efficient frontier. The linear
programming technique identifies best-practice DMUs, or those that are on the frontier. All
3 Efficiency is defined as the relationship between inputs and outputs (outcomes), wherein monetary
inputs are considered. Inputs (educational expenditure, students etc.) are “transformed” into
outputs/outcomes (number of graduates, their knowledge etc.) through the “production” (pedagogic)
process (Bevc and Uršič, 2008, p. 234).
4 Modified according to Warning (2004, p. 396).
The Knowledge Based Economy:
Implications for Higher Education in Economics and Business AE
Vol. XIII • No. 30 • June 2011 371
other DMUs are viewed as being inefficient relative to the frontier DMUs (Chapple, et al.,
2005, p. 371). As already mentioned, the article analyzes the relative efficiency of
government spending on education in Croatia and Slovenia. It does so by comparing
spending on these sectors and key higher education (outcome) indicators in the two
countries. Relative efficiency is defined as the distance of a country’s observed input-
output combination from an efficiency frontier. This frontier is estimated using the DEA
approach that was explained earlier and represents the maximum attainable outcome for a
given input.
The data set in this research includes input data, i.e. expenditure per student, tertiary (% of
GDP per capita) and output/outcome data, i.e. school enrolment, tertiary (% gross), labor
force with a tertiary education (% of total) and the unemployed with a tertiary education (%
of total unemployment) in thirty-seven countries are included in the analysis (selected new
EU member states, and OECD countries). In order to assess different inputs and
outputs/outcome relative to technical efficiency, two models have been tested. Model 1 is
based on expenditure per student, tertiary (% of GDP per capita, 1999-2007 averages) (as
input) and school enrolment, tertiary (% gross), labor force with a tertiary education (% of
total, 1999-2007 averages) and the unemployed with a tertiary education (% of total
unemployment, 1999-2007 averages) (as output/outcome). Relative efficiency scores for
Model 2 are based on expenditure per student, tertiary (% of GDP per capita, 1999-2007
averages) (as input) and school enrolment, tertiary (% gross) and labor force with a tertiary
education (% of total, 1999-2007 averages) (as output/outcome). The program used for
calculating the technical efficiencies is the DEAFrontier software. The data are provided
by Eurostat, the IMF, the OECD, UNESCO, and the World Bank’s World Development
Indicators database.
3. Research Results
When looking at the results5, by using Model 1 and applying the DEA efficiency frontier
technique within a selected group of the new EU member states and OECD countries to
measure efficiency of higher education, Canada, Czech Republic, Finland, the Republic of
Korea, Latvia, Lithuania, Poland, Russia, Slovakia and even Slovenia are seen as efficient.
Here, the average expenditure per student, tertiary (% of GDP per capita) in the 1999-2007
period measures the input and as the output/outcome we use school enrolment, tertiary (%
gross), labor force with a tertiary education (% of total, 1999-2007 averages) and the
unemployed with a tertiary education (% of total unemployment, 1999-2007 averages). One
can see that some countries come very close to the frontier (e.g. Hungary and Romania),
while the other countries are further away and therefore less efficient (e.g. Cyprus and
France) (Table no. 3).
The results of the DEA analysis (Model 1) also suggest a relatively high level of
inefficiency in higher education in Croatia and, correspondingly, significant room to
rationalize public spending without sacrificing, while also potentially improving, higher
education outputs and outcomes (Table no. 3). Indeed, Croatia is ranked in the third quartile
5 All of the results relate to DEA with an output orientation, allowing for variable returns to scale
(VRS). An output orientation focuses on the amount by which output quantities can be proportionally
increased without changing the input quantities used. Using an input orientation approach leads to
similar efficiency results as those presented in the text.
AE Relative Efficiency of Higher Education in Croatia and Slovenia:
An International Comparison
Amfiteatru Economic
372
and in terms of the efficiency scores for public spending, Croatia ranks in the 69th
percentile among the 37 countries. With respect to individual output/outcome indicators,
Croatia’s ranking is in the last quartile for higher education school enrolment, the third
quartile for labor force with a tertiary education and the second quartile for the unemployed
with a tertiary education. In order to become an efficient country, Croatia should
significantly reduce its average expenditures on higher education per student by around 10
percentage points (to around 29% of GDP per capita), to bring it near to the OECD average
level.
Table no. 3: The Relative Efficiency of Selected New EU Member States and OECD
Countries in Tertiary Education – Model 1 (Distribution by quartiles
of the ranking of efficiency scores)
I. quartile II. quartile III. quartile IV. quartile
Canada Italy Norway Cyprus
Czech R. Ireland Croatia Mexico
Finland Austria New Zealand Denmark
Korea Australia Japan France
Latvia Bulgaria Sweden Netherlands
Lithuania Romania United Kingdom Spain
Poland Hungary Estonia Switzerland
Russia Portugal Iceland
Slovakia Greece Turkey
Slovenia Belgium
United States
Source: World Bank, 2010; UNESCO, 2010; Eurostat, 2010a; OECD, 2010; own
calculations
Further empirical analysis, now focusing on Model 2, suggests even worse relative
efficiency results for Croatia. When using only two outputs/outcomes, Croatia’s ranking is
only 32 (out of 37). Similar to the results for Model 1, in order to become efficient Croatia
should cut its average expenditures on higher education per student by 6.3 percentage
points. In terms of the efficiency scores, the efficiency benchmark is represented by
Canada, Finland, the Republic of Korea and the USA. In contrast, some new EU member
states lag well behind (e.g. Slovakia, Romania and the Czech Republic), especially due to
relatively poor output/outcome results (relatively low school enrolment and labor force
with a tertiary education). Slovenia is ranked in 13th position and would improve its
efficiency score by significantly expanding its labor force with a tertiary education (by
around 8.5 percentage points) (Table no. 4).
According to our descriptive and empirical analysis, it is obvious that the higher education
systems in Croatia and Slovenia suffer from relatively high technical inefficiencies (in
particular in Croatia). To improve each system’s efficiency, performance-based funding
models for higher education should be developed and further emphasis should be placed on
quality assurance in higher education and the integration of the facilities. Moreover,
curricula in universities should also be reformed to better reflect the needs of the economy,
whereas dialogue and cooperation between the private sector and universities should be
greatly increased. Indeed, trade unions and employers should be actively involved in
The Knowledge Based Economy:
Implications for Higher Education in Economics and Business AE
Vol. XIII • No. 30 • June 2011 373
education reform. That is especially important in the area of vocational higher education
programmes in order to reduce labour market mismatches. Improvement of the education
system should be a top priority of tripartitive dialogue.
Table no. 4: The Relative Efficiency of Selected New EU Member States and OECD
Countries in Tertiary Education (Model 1 and Model 2)
Source: World Bank, 2010; UNESCO, 2010; Eurostat, 2010a; OECD, 2010; own
calculations.
Conclusion
Expenditures on higher education systems made an important role in improving economic
growth and development. It is familiar that public educational expenditures significantly
improve growth performance and confirm a positive role between educational attainment
and economic growth. Countries with more educated labour have tended to grow faster
over the post-1960 period (Norman, 1998, p. 133). At the same time, expenditures on
higher education signify an important tax burden on taxpayers. The efficiency with which
inputs produce the desired outputs is thus an important public policy issue. In this research,
an attempt was made to measure the relative efficiency of higher education across selected
OECD and new EU member states, in particular in Croatia and Slovenia, using data
envelopment analysis (DEA) in a VRS framework. The findings of the article are generally
supported by other similar studies by OECD, the World Bank, and Jafarov and Gunnarsson
(2008). Indeed, the research results suggest the significant inefficiency of higher education
spending in Croatia and therefore the considerable potential to reduce government
expenditure and/or to increase the higher education output/outcome. Conversely, regardless
of the input-output/outcome mix, the higher education system in Slovenia is shown to have
a much higher level of efficiency compared to Croatia as well as many other comparable
Model 1
Model 2
Country Output-Oriented VRS
Efficiency Rank
Output-Oriented VRS
Efficiency
Rank
Cyprus 1.18366 37 1.67953 27
Czech R. 1.00000 1 2.22684 33
Estonia 1.04146 21 1.30988 16
Finland 1.00000 1 1.00000 1
Hungary 1.00243 12 1.68296 28
Lithuania 1.00000 1 1.24196 12
Poland 1.00000 1 1.48874 22
Republic of
Korea
1.00000
1 1.00000 1
Romania 1.00460 13 2.31993 35
Slovakia 1.00000 1 2.32826 36
USA 1.00000 1 1.00000 1
Croatia 1.06280 26 2.21438 32
Slovenia 1.00000 1 1.25579 13
AE Relative Efficiency of Higher Education in Croatia and Slovenia:
An International Comparison
Amfiteatru Economic
374
new EU member states and selected OECD countries. The results also indicate that some
developed nations (e.g. Korea, the USA and Finland) can serve as benchmarks for their
efficient use of higher education resources. Nevertheless, the improvement of data quality
and testing the influences of the environmental factors (such as climate, socio-economic
background etc.) remain important issues for further research.
References
1. Athanassopoulos, A.D. and Shale, E., 1997. Assessing the Comparative Efficiency of
Higher Education Institutions in the UK by Means of Data Envelopment Analysis.
Education Economics, 5(2), pp. 117-134.
2. Audretsch, D.B., Lehmann, E. and Warning, S., 2003. University Spillovers: Strategic
Location and New Firm Performance. CEPR Discussion Paper, no. 3837, London:
CEPR.
3. Aver, B. and Čadež, S., 2009. Management accountants' participation in strategic
management processes: a cross-industry comparison. J. East Eur. manag. stud., 14(3),
pp. 310-322.
4. Bevc, M. and Uršič, S., 2008. Relations between funding, equity, and efficiency of
higher education. Education Economics, 16(3), pp. 229-244.
5. Barro, R. and Sala-i-Martin, X., 1995. Economic Growth. Boston: McGraw-Hill.
6. Burcea, M. and Marinescu, P., 2011. Students’ Perceptions on Corporate Social
Responsibility at the Academic Level. Case Study: The Faculty of Administration And
Business, University of Bucharest. Amfiteatru Economic, XIII(29), pp.207-220.
7. Chapple, W., Lockett, A., Siegel, D. and Wright, M., 2005. Assessing the relative
performance of U.K. university technology transfers offices: parametric and non-
parametric evidence. Research Policy, Issue 34, pp. 369-384.
8. Charnes, A., Cooper, W.W. and Rhodes, E., 1978. Measuring the Efficiency of
Decision Making Units. European Journal of Operational Research, Issue 2,
pp. 429-444.
9. Dantzig, G.B., 1951. Maximization of a linear function of variables subject to linear
inequalities. In: T.C. Koopmans, ed. 1951. Activity analysis of production and
allocation. New York: Wiley, pp. 230-321.
10. Denison, E.F., 1985. Trends in American economic growth, 1929-1982. Washington,
DC: Brookings Institute.
11. European Commission, 2009. Progress towards the Lisbon Objectives in Education
and Training – Indicators and benchmarks. [online] Available at:
<http://ec.europa.eu/education/lifelong-learning-policy/doc/report09/report_en.pdf>
[Accessed 17 September 2010].
12. Eurostat, 2010a. Education in Europe – Key statistics 2008. [online] Available at:
<http://epp.eurostat.ec.europa.eu/cache/ITY_OFFPUB/KS-QA-10-037/EN/KS-QA-
10-037-EN.PDF > [Accessed 29 September 2010].
13. Eurostat, 2010b. Indicators on education expenditure for 2007. [online] Available at:
<http://epp.eurostat.ec.europa.eu/cache/ITY_OFFPUB/KS-QA-10-038/EN/KS-QA-
10-038-EN.PDF > [Accessed 15 September 2010]
The Knowledge Based Economy:
Implications for Higher Education in Economics and Business AE
Vol. XIII • No. 30 • June 2011 375
14. Eurostat, 2009. Key data on education in Europe – 2009. [online] Available at:
<http://eacea.ec.europa.eu/education/eurydice/documents/key_data_series/105EN.pdf
> [Accessed 15 September 2010].
15. Farrell, M., 1957. The Measurement of Productive Efficiency. Journal of the Royal
Statistical Society, Series A (General), 120(3), pp. 253-290.
16. Hashimoto, K. and Cohn, E., 1997. Economies of scale and scope in Japanese private
universities. Education Economics, 5(2), pp. 107-115.
17. Jafarov, E. and Gunnarsson, V., 2008. Government Spending on Health Care and
Education in Croatia: Efficiency and Reform Options. [online] Available at:
<http://www.imf.org/external/pubs/ft/wp/2008/wp08136.pdf> [Accessed 28 September
2010].
18. Johnes, J., 2006. Data envelopment analysis and its application to the measurement of
efficiency in higher education. Economics of Education Review, Issue 25, pp. 273-288.
19. Keller, K. , 2006. Investment in Primary, Secondary, and Higher Education and the
Effects on Economic Growth. Contemporary Economic Policy, 24(1), pp. 18-34.
20. Kerstens, K. and Woestyne Van de, I., 2009. Negative Data in DEA: A Simple
Proportional Distance Function Approach. Lille Economie & Management; Document
de travail du LEM, 06, pp. 1-11.
21. Madden, G., Savage, S. and Kemp, S., 1997. The relative efficiencies of Canadian
universities: a DEA perspective. Research paper, No. 97-4, Department of Economics,
University of Alberta.
22. McMillian, M.L. and Debasish, D., 1997. The relative efficiencies of Canadian
universities: a DEA perspective. Research paper, No. 97-4, Department of Economics,
University of Alberta.
23. MHEST, 2010a. Higher education: Directorate for Higher Education. [online]
Available at: <http://www.mvzt.gov.si/si/delovna_podrocja/visoko_solstvo/>
[Accessed 20 November 2010].
24. MHEST, 2010b. A draft of the National Higher Education Master Plan. [online]
Available at: <http://www.mvzt.gov.si/fileadmin/mvzt.gov.si/pageuploads/pdf/
odnosi_z_javnostmi/14.10._NPVS.pdf [Accessed 10 November 2010].
25. MoSES, 2007. OECD Thematic Review of Tertiary Education – Country Background
Report for Croatia. [online] Available at: <http://www.oecd.org/document/5/
0,3746,en_2649_39263238_35580240_1_1_1_1,00.html> [Accessed 5 November
2010].
26. Norman, G., 1998, Reviewing the New Growth Literature. New Political Economy,
Issue 3, pp. 129-134.
27. OECD, 2010. Education at Glance 2010. [online] Paris: OECD Indicators. Available
at: <http://www.oecd.org/document/52/0,3343,en_2649_39263238_45897844_1_1_1_
1,00.html > [Accessed 9 November 2010].
28. OECD, 2009a. Education at Glance 2009. [online] Paris: OECD Indicators. Available
at: <http://www.oecd.org/dataoecd/41/25/43636332.pdf> [Accessed 25 October 2010].
AE Relative Efficiency of Higher Education in Croatia and Slovenia:
An International Comparison
Amfiteatru Economic
376
29. OECD, 2009b. Higher Education to 2030. [online] Paris: OECD. Available at:
<http://www.oecd.org/document/18/0,3343,en_2649_35845581_43908242_1_1_1_1,0
0.html > [Accessed 29 September 2010].
30. OECD, 2001. Thematic Review of National Policies for Education: Croatia. Paris:
OECD.
31. Rosca, I.G. et al., 2008. Quality Assurance Systems in Higher Education. Amfiteatru
Economic, X(Sp. Iss. 2), pp. 6-12.
32. SORS, 2010. Graduates from vocational colleges and higher education institutions,
Slovenia, 2009 - final data. [online] Ljubljana: Statistical Office of Republic of
Slovenia. Available at: <http://www.stat.si/eng/novica_prikazi.aspx?id=3199>
[Accessed 4 September 2010].
33. Stare, J. and Klun, M., 2008. Improving public administration performance demands
investment in human resources. Zbornik radova Ekonomskog fakulteta u Rijeci. 26(1),
pp. 151-173.
34. Tajnikar, M. and Debevec, J., 2008. Funding system of full-time higher education and
technical efficiency: case of the University of Ljubljana. Education Economics, 16(3),
pp. 289-303.
35. Tóth, R., 2008. Using DEA to evaluate efficency of higher education. [online]
Budapest: Agroinform Publishing House. Available at:
<http://ageconsearch.umn.edu/bitstream/53548/2/17_Using%20DEA_Apstract.pdf>
[Accessed 18 December 2010].
36. Unesco, 2010. Data Centre. [online] Montreal: UNESCO Institute for Statistics.
Available
at:<http://stats.uis.unesco.org/unesco/TableViewer/document.aspx?ReportId=143&IF_
Language=eng> [Accessed 28 December 2010].
37. Warning, S., 2004. Performance Differences in German Higher Education: Empirical
Analysis of Strategic Groups. Review of Industrial Organization, 24(4), pp. 393-408.
38. World Bank, 2010. World Development Indicators. [online] Available at:
<http://data.worldbank.org/indicator> [Accessed 15 December 2010].
39. World Bank, 2008. Croatia Restructuring Public Finance to Sustain Growth and
Improve Public Services: A Public Finance Review. [online] Washington, USA.
Available at:
<http://siteresources.worldbank.org/INTCROATIA/Resources/Croatia_pfr_hr.pdf>
[Accessed 20 January 2011].
40. Worthington, A., 2001. An Empirical Survey of Frontier Efficiency Measurement
Techniques in Education. Education Economics, 9(3), pp. 245-268.
