tokyo

Despite a struggling economy, overall research spending in Japan grew by 3.4 per cent in the fiscal year 1996 (which ended in March 1997), according to a survey on research and development (R&D) expenditure by the Management and Coordination Agency, a government agency that reports to the prime minister's office.

Onward and upward: R&D expenditures have shaken off the recession of the early 1990s.

The survey, based on data provided by several thousand companies, universities and research institutes, estimates total R&D spending at ¥14.9 trillion (US$ 115 billion). If the software industry — included for the first time in this year's survey — is added, this figure rises to ¥15.1 trillion.

Although this year's increase is below last year's rise of 6 per cent, it is still considerably more than the growth in gross domestic product (GDP). With an estimated 3 per cent of total GDP spent on research during 1996, Japan remains well ahead of all the other large industrialized countries, whose growth rates have been below those of GDP for much of the past decade.

But although industrial R&D — excluding software — displayed a record 5.2 per cent increase, that for higher education is estimated at a modest 1 per cent. Indeed, although spending appears to have increased markedly in private universities, public funding for university research actually declined by several percentage points. Funding at research institutes — which includes both public and private institutions — also went down, by 1.1 per cent.

Furthermore, overall public research spending decreased by some 4 per cent, according to the survey, a figure that appears to conflict with the large nominal increases in long-term government research funding budgeted for the same period (see Nature 385, 104; 1997). This discrepancy is partly explained by unusually generous ‘supplementary budgets’ for the previous year, which added considerably to a 12 per cent increase in public research spending.

As part of a ¥14 trillion government effort to stimulate the economy, supplementary budgets in 1995 included an additional ¥685 billion for research. Most of this, however, went into new buildings and facilities for universities and national institutes; extra funds for R&D in last year's supplementary budgets amounted to only ¥155 billion.

The large increase in private expenditures revealed by this year's survey appears to confirm claims that the political objective of the ‘Basic Law’ was aimed less at raising public research spending than at reinvigorating private-sector R&D. After periods of sustained growth during the 1980s, companies had been cutting their research budgets in the early 1990s because of the recession.

But since last year, industrial R&D has been rising aggain. This trend appears to be confirmed by a recent survey of some 200 large corporations by the Nikkan Kogyo Shimbun, a daily newspaper for business and technology, which found growth rates for 1997 of between 2 and 10 per cent for all industries except petroleum refining. Growth in industrial R&D has been most markked in the electronics, telecommunications and transport sectors. Total R&D in software was esstimated at only ¥177 billion, a sum that is below R&D spending by some iindividual US companies.

According to Shinichi Kobayashi, a science policy analyst at the University for Electro-Communications near Tokyo, R&D expenditure in smaall and medium-sized enterprises is inadequately measured, and as a result the data provided by the survey may underestimate structural changes in the Japanese economy.

In particular, analysis of inter-company R&D funding, as measured by the annual survey, reveals a widening gap between the total amount received by industry and that paid out for research, leaving a growing proportion of funds spent on extramural R&D unaccounted for. According to Kobayashi, this suggests a trend towards out-sourcing of R&D to small companiess that does not show up in the survey data.