Substantial increases in educational spending over the last thirty years have not lead to
improvements in the educational achievement of pupils in many OECD countries, according to
long-term international research by Erich Gundlach, Ludger Woessmann and Jens Gmelin,
published in the latest issue of the Economic Journal.
The researchers find that the productivity of the educational systems in OECD countries has fallen
steadily. Their results indicate that over the period 1970-94, the average annual fall in schooling
productivity was 2.8% in the UK, 3.2% in Germany, 3.4% in Japan, 3.8% in Italy and 4.4% in
France. This suggests that increasing educational budgets does not necessarily lead to improved
educational performance. As a consequence, the authors conclude, education reform should focus
on how educational systems function rather than on additional expenditure.
In the United States, a fierce political debate focuses on alternative measures to improve the
presumed low productivity of US public schooling, which is seen as a major threat to the future
standard of living in a knowledge-based society. But declining schooling productivity has not
attracted much attention outside the United States. Yet these results show that schooling
productivity has declined much faster in France, Germany, Italy, Japan and the UK than in
America, where inflation-adjusted public education expenditure per pupil rose by 2.5% per year
between 1970-94. This means that the US schooling sector as a whole received about twice as
much resource input per pupil in 1994 than it did in 1970. Since the number of pupils is the output
of the schooling sector and since productivity is defined as output per input, these raw numbers
suggest a decline in schooling productivity.
Of course, increased educational spending is not a problem if the quality of schooling has
improved as a result of the additional expenditure. In other words, pupils today may have better
skills than pupils of thirty years ago. But this possibility is not borne out by the facts. The
performance of US pupils in standardised cognitive achievement tests in mathematics and natural
sciences has not changed significantly between 1970-94.
A second objection against interpreting these findings as a productivity decline is that the raw
numbers have to be corrected for the economy-wide rate of productivity growth. This argument is
based on considerations that suggest sectors with relatively low or zero productivity growth have
to face increasing cost pressures, which translate into higher measured resource input.
Hence, part of the observed increase in schooling resources may simply reflect the so-called 'cost
disease' of services. But the researchers find that this factor is relatively small. Depending on the
specific measure used, they estimate that the decline in the productivity of US schooling is in the
range of 1.2-2.2% per year. This means that the US schooling sector as a whole received about
half as much more resource input per pupil in 1994 than it did in 1970, while their average skills
did not increase.
The results from the other OECD countries put the American experience into perspective. As with
the United States, the researchers find no evidence for an improved quality of education in the
other OECD countries - as measured by improved student performance. Interestingly, the only
countries where there has been a slight improvement in student performance are those reporting
the lowest average increase in schooling expenditure per pupil, namely Sweden and the
Netherlands.
Hence, in many OECD countries, even substantial increases in schooling resources did not boost
schooling quality. These findings confirm a large literature of individual case studies, which fail to
identify a positive relation between additional schooling inputs and student performance.
Indeed, Woessmann concludes 'The findings indicate that the schooling sector can be regarded as
inefficient in many countries. A more productive use of given resources seems to be a much more
pressing issue for schooling reform than increasing the educational budget.'
Notes for Editors: 'The Decline of Schooling Productivity in OECD Countries' by Erich
Gundlach, Ludger Woessmann, and Jens Gmelin is published in the May 2001 issue of the
Economic Journal. Gundlach and Woessmann are at the Kiel Institute of World Economics,
Germany; Gmelin is at Kreditanstalt fur Weideraufau, Germany.
For Further Information: contact Ludger Woessmann on 00-49-431-8814-497 (fax: 00-49-431-
8814-500; email: woessmann@ifw.uni-kiel.de); RES Media Consultant Romesh Vaitilingam on
0117-983-9770 or 07768-661095 (email: romesh@compuserve.com); or RES Media Assistant
Niall Flynn on 020-7878-2919 (email: nflynn@cepr.org).