Korea's per capita income has risen from 15 percent of the U.S. level in 1970 to around half by 2005 (Figure 1). Rapid economic development has been based primarily on inputs of capital and labor, driven by the highest rate of business investment in the OECD area, a growing working-age population, long working hours and rising participation in the labor force. In addition, investment in education and R&D has facilitated a "catch-up" model of innovation. Looking ahead, though, the outlook for input-based growth is dimming. The size of the working-age population is projected to begin falling from 2016 in the context of rapid population aging. Indeed, the share of the population over age 65 is projected to double from 7 percent in 2000 to 14 percent in 2018, and increase further to 20 percent in 2026. Meanwhile, fixed investment has dropped from an average of 37 percent of GDP during the 1990s, prior to the crisis, to less than 30 percent during the past few years.

With inputs of labor and capital slowing, sustaining high growth depends increasingly on innovation to drive productivity gains. There is considerable scope to raise labor productivity, which is 60 percent below the U.S. level (Figure 2, right-hand column). Productivity gains are closely linked to innovation - the successful development and application of knowledge - which in turn depends to a large extent on investment in R&D and human capital. Such investment is particularly important to Korean exporters, many of whom have reached the technology frontier in their fields. Indeed, according to a survey by the Korea International Trade Association, around half of the big companies report that the quality of their products was at least as good as rivals in more advanced countries. Increasing productivity and continuing the convergence to the income levels in the most advanced countries depend to a large extent on improving Korea's R&D system and upgrading the education system in order to increase the capacity for innovation.

Improving the R&D system in Korea

Total R&D spending in Korea increased from 2.6 percent of GDP in 2003 to 3 percent in 2005, surpassing the 2.7 percent ratio in the United States (Figure 3). R&D spending in Korea is concentrated in the business sector, which accounts for three quarters of the total.

Weaknesses in Korea's R&D framework

Despite the high level of spending, Korea's R&D framework has a number of weaknesses:

(1) Korea needs to make the transition from a catch-up strategy of innovation, which has emphasized imports of technology and reverse engineering, to a more creative approach. In a survey of 6,000 manufacturing firms, the overall level of technology was estimated to be 80 percent of the world technology frontier, while 13 percent of firms replied that they have already reached the frontier. Consequently, foreign sources can no longer provide the needed expertise in a growing number of fields. Moreover, foreign firms appear to have become more reluctant to release their technology.

(2) R&D activities are focused on a relatively small number of industries and firms in the manufacturing sector. The Information and Communication Technology (ICT) sector and automobiles accounted for 60 percent of business R&D in 2003, with the top five companies alone in these two industries responsible for 37 percent of total business R&D. The heavy concentration in two industries contributes to the dualism in the Korean economy and may not provide a broad enough base to promote the convergence of per capita income in Korea to the levels in the most advanced OECD countries.

(3) R&D in services is small. Although the service sector produces 57 percent of value-added in the Korean economy, it accounts for less than 10 percent of business R&D, well below the OECD average of 25 percent and the 40 percent share in the United States. The low level of R&D in services in Korea contributes to the large productivity gap between its manufacturing and service sectors. Indeed, service sector productivity was only 65 percent of that in manufacturing in Korea in 2003, well below the OECD average of 97 percent.

(4) Universities are not well integrated in the R&D system. Although they employ about three-quarters of researchers with a Ph.D. degree, they account for only 10 percent of the total R&D performed in Korea, about half of the OECD average, indicating that the human resources in universities are underutilized in research. More generally, interaction between business, government and academic R&D activities is weak. For example, only 2 percent of R&D financed by the business sector is performed in universities and only 1 percent in government research institutes.

(5) Korea's R&D system is relatively isolated internationally. Foreign sources financed only 0.7 percent of R&D activities in Korea in 2005, the second lowest figure in the OECD area. Korea also ranks second to last in terms of the share of foreign ownership of domestic inventions, partly reflecting the low level of FDI in Korea and weak linkages with foreign research institutes. International isolation may limit the scope for technological progress, as foreign sources of knowledge are increasingly important for innovation.

Focusing on "growth engines" and creating "innovation cities"

In its 2003 overhaul of the national innovation system, the government unveiled a plan for the development of certain technologies to act as growth engines for the Korean economy over the next five to 10 years. The 10 strategic industries identified as growth engines are: bio-medical products, next-generation computer displays, next-generation semiconductors, next-generation batteries, future automobiles, intelligent robots, digital TV and broadcasting, next-generation mobile communications, intelligent home networks and digital content and software solutions. The government argues that since the absolute amount of R&D in Korea is small compared to some of the leading countries, it is necessary to pick strategically important industries and concentrate R&D spending in them. The criteria for selecting the growth engines include world market size and strategic importance.

To promote the growth engines, the government is encouraging the development of the necessary technology, the creation of infrastructure to support R&D investment and the training of skilled workers and measures to enhance collaboration between firms, universities and government research institutes. To accomplish these goals, the government has established 10 consortiums, which provide R&D funds and develop the necessary infrastructure, including the training of skilled workers. Moreover, they receive other advantages. For example, the shareholding ceiling imposed on chaebol affiliated companies is waived for the 10 strategic industries. Perhaps equally important as the financial support and regulatory advantages for the growth engine industries is the signal provided to the private sector.

However, picking winners and giving them undue emphasis could lead to government failure and distortions. The experience of the Heavy and Chemical Industry drive in Korea during the 1970s, which focused investment on certain strategic industries, illustrates the risks of government measures to pick winners. The risks may be even larger today, given globalization and the rapid pace of technological change, which make it difficult for the government to accurately anticipate what areas will be most important in the future. In addition, the growth engine program focuses on a number of key manufacturing industries, such as cars, semiconductors, computer displays and mobile telecommunication equipment. Such an approach tends to maintain the emphasis on manufacturing at the expense of the service sector. The growth engine approach may also widen the gap between large companies and small and medium-sized enterprises as priority areas are driven by large firms.

Another risk of focusing on key high technology products, such as semiconductors and mobile telecommunications, is a further deterioration in the terms of trade as other countries also increase production in these areas. Terms of trade losses have reduced Korea's national income by 14 percent since 1998, the largest such loss in the OECD area (Figure 4), reflecting the downward trend in the price of key ICT products. Consequently, the growth of national income has lagged behind the growth of output. The constraint on income growth has limited private consumption and slowed the improvement in living standards. Other countries, which are major producers of ICT products, such as Sweden, Finland and Japan, also report large terms of trade losses. Countries that target priority industries for R&D tend to pick some of the same sectors as Korea, suggesting that there may be continued terms of trade losses in the future. This perspective suggests that there are gains to a more diversified approach to R&D.

Meanwhile, Korea is establishing "innovation cities" as part of its objective of regional development to reduce concentration in the capital region. This initiative aims at strengthening the link between public organizations, industry and universities to create a favorable environment for innovation. The innovation cities will benefit from the transfer of 175 public institutions from the capital region, thus reducing the share of public organizations located in the capital region from 85 to 35 percent. However, this initiative raises several concerns, notably the high degree of government intervention in creating the cities. Moreover, mixing regional development programs with measures to upgrade the innovation system will tend to reduce the effectiveness of both initiatives. Instead, granting more autonomy to local governments and enhancing their revenue sources are essential for more balanced regional development.

(To be continued tomorrow)

By Randall Jones

2007.08.30