وسواس برای رقابت پذیری و تاثیر آن بر آمار: ساخت و ساز شاخص های فن آوری پیشرفته

High technology is a concept much in vogue in OECD countries, for it is a symbol of an “advanced” economy. This paper looks at why and how the concept and its indicator acquired such fame. It explains that the reason has to do with the fact that the indicator emerged in the context of debates on the competitiveness of countries and their efforts to maintain or improve their positions in world trade. The first part looks at the early statistics behind the indicator (R&D/sales), statistics developed in official analyses of industrial R&D surveys before the 1950s. The second part traces the evolution of the R&D/sales ratio in the 1960s through its use as an indicator of research or technological intensity. The third part discusses the internationalization of the indicator via the OECD.

Early official economic statistics dealt with national income, industrial production, and productivity. Following World War II, international trade was increasingly added to statistical series (Tomlinson, 1996). In fact, international relations between countries, growing trading exchanges and competitiveness came to dominate the political agendas of several governments. It was in this context that two of the main science and technology indicators that are currently in vogue came to be developed: the technological balance of payments (TBP) and high technology. This paper deals with the emergence of the latter. High technology (or technology intensity) is an indicator much in vogue in OECD countries, for it is a symbol of an “advanced” economy. The indicator is in fact the analog to industry of the GERD/GDP indicator for countries: a ratio of R&D divided by production. Industries are classified according to whether they are above or below the average ratio. An industry that invests above the average in R&D is considered to be a high-technology industry. The indicator remains a controversial one for conceptual and for methodological reasons. Nevertheless, governments use it continually as part of their economic and innovation policy. As the US National Science Foundation (NSF) summarized the situation, high-technology industries are important to nations for several reasons (NSF, 2002, pp. 6–5): 1. High-technology firms innovate, and firms that innovate tend to gain market share, create new products/markets, and/or use resources more productively. 2. Industrial R&D performed by high-technology industries benefits other commercial sectors by generating new products and processes that increase productivity, expand business and create high-wage jobs. 3. High-technology firms develop high-valued-added products and are successful in foreign markets, which results in increased competition. What is characteristic about the indicator on high technology is its linkage to competitiveness issues. This paper explains that the reason has to do with the fact that the indicator emerged in the context of debates on the competitiveness of countries and their efforts to maintain or improve their positions in world trade. High technology rapidly came to be viewed as the solution to the issue, and statistics were developed to document the case. Some authors have qualified the debates on competitiveness as obsessive. For P. Krugman, for example, countries do not compete economically with each other as corporations do (Krugman, 1994). But the metaphor “derives much of its attractiveness from its seeming comprehensibility” (p. 39). First, the competitive image is exciting, and thrills sell tickets. Second, the metaphor makes difficulties easier to solve (subsidize high technology or be tough on competitors). Third, it is a political device that assists in justifying choices. This article shows how statistics contributed to the rhetoric on competitiveness, looking at the efforts of official statisticians to rhetorically transform early statistics on R&D into indicators on high technology. Where does the indicator come from? Who was behind its construction? What discourses did governments conduct using the indicator? This paper intends to answer these questions. It is divided into three parts. The first looks at the basic statistics behind the indicator, statistics developed in early official analyses of industrial R&D surveys. The second part traces the evolution of the statistic through its use as an indicator of research or technological intensity. The third part discusses the internationalization of the indicator via the OECD.

How has the concept of high technology improved over the previous concepts? Certainly, one could argue with Kelly that: “research-intensity and technologyintensity are not necessarily the same concept. What one is really trying to measure [with technology intensity] is the degree of technical sophistication of products that gives them a competitive edge (. . .)” (Kelly, 1976, p. 8). However, it should have become clear from the above analysis that both concepts are actually the same, according to their measurement. Because, as Kelly himself admitted: “as in many areas of economics, proxies must be used as an indicator (. . .). [And we] chose the R&D intensity”. The OECD technology intensity indicator is also based on R&D expenditures to sales ratios, and is frequently discussed in terms of R&D intensity rather than technology intensity. When more than one criterion is used, the criteria usually center around Boretsky’s three, to the point that “limited progress appears to have been made on the measurement of technology intensiveness since the original Boretsky paper” (Chesnais and Michon- Savarit, 1980, p. 14), except rhetorical inventiveness. For example, a newly-coined concept appeared recently at theOECDin itswork on the knowledge-based economy: knowledge-based industries (Godin, 2004a). Knowledge-based industries are defined as those that have the following three characteristics: (1) a high level of investment in innovation, (2) intensive use of acquired technology, and (3) a highly-educated workforce (Webb, 2000, 2001). This is a perfect example of a variation on the high technology indicator. Briefly stated, if technology intensity is a replica of research-intensity, high technology is simply a rhetorical exercise renaming technology intensity.Why? The label was the way to link and align the statistical work to political and normative issues, where buzzwords are the rule. In fact, high technology is the perfect example of a fuzzy concept of much value for rhetorical purposes. Officials use it constantly simply for its prestigious appeal.17 What role have academics played in all this? It is clear that the indicator originally came from official organizations. Early on, and in general, academics (economists) satisfied themselves with models correlating R&D with exports to assess the role of science on trade performances (Lowinger, 1975; Wolter, 1977; Legler, 1987; Fagerberg, 1988; Dosi and Soete, 1988). But they also acted as consultants to public organizations, helping to define indicators on technology intensity,18 and increasingly used the high-technology label and its indicators in the 1990s or developed their own classifications (Papagni, 1992; Amendola and Perrucci, 1994; Grupp, 1995; Guerrieri and Milana, 1995; Roessner et al., 1996, 2002; Porter et al., 1996, 1999, 2001). Above all, they were at the heart of the rhetoric on competitiveness (Porter, 1990). Today, the indicator remains a highly contested measure (Baruch, 1997). One of the main criticisms has to do with the basic statistics behind the indicator: R&D expenditures. Some authors therefore have suggestedreplacing R&D with patents (Soete, 1987), and others have argued for using several dimensions and statistics to define the indicator (Sciberras, 1986; Hughes, 1988; Felsenstein and Bar-El, 1989). A second frequentlyvoiced criticism has to do with the fact that a firm may not be considered technology-intensive only because it conducts R&D, but also if it adopts and uses advanced technologies in its activities and employs highly-trainedworkers (Palda, 1986; Baldwin and Gellatly, 1998).Athird criticism refers to the fact that there is no standardization yet, and therefore organizations and authors produce different results. T.A. Abbott, for example, has documented how a surplus of $ 3.5 billion (1985–1988) is measured when products are used as units, and a deficit ($ 17 billion) appears when the measurement is based on industries (Abbott, 1991). In the light of these criticisms, some have decided to ignore the indicator, arguing that what distinguishes industries is not products but technologies—and market characteristics. Instead, they have suggested alternative sectoral taxonomies.19 Certainly, the indicator brought simplification to statistical analyses. Before high-technology groups appeared, analysis of R&D and trade was conducted according to industrial classes individually.20 The high-technology indicator reduced the classification to only three groups. But the indicator also highlighted statistical discrepancies between studies. As the OECD itself argued, each author, organization or country has its own idea of what constitutes high technology and uses its own vocabulary (OECD, 1993c). The OECD was only partly right, however, when it suggested that “the concept of high technology became part of our everyday vocabulary before economists and scientists had even managed to produce a precise and generally-accepted definition of the term” (OECD, 1988a, p. 3). Very early on, official economists (and statisticians) invented the concept and constructed a measurement—which more or less focused on data on R&D. The problem stems rather from the political obsession to which it was applied— competitiveness21 — and the urge to support the case quantitatively.