![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
Abstract
This paper examines how technology specialization, measured by citation-weighed patents, affects trade flows. The paper analyzes the relationship between (i) technology specialization and export specialization across regions and (ii) the technology specialization of origin and destination and the quality of export flows. We find that the export specialization of regions corresponds to their technology specialization. Regions with higher technology specialization export products of higher quality, as indicated by higher prices. Moreover, export flows to destination countries with a high technology specialization consist of products of higher quality in the specific technology. The results are consistent with knowledge and technology being important for export performance and with regions with higher specialization in a technology being better equipped to produce high-quality products. They are also consistent with destinations of higher technology specialization, having a more pronounced demand for products of higher quality in the same technology.
Acknowledgements
We thank Carolina Castaldi, Ari Kokko and two anonymous referees for useful comments. Financial support from the Swedish Governmental Agency for Innovation Systems (VINNOVA) is gratefully acknowledged.
Notes
1 An associated assumption in this framework is that technology knowledge do not diffuse instantly.
2 See also Kaldor Citation(1981).
3 Advocates of the use of USPTO data often put forward the argument that the US is the world's single largest economic market and any technological advantage sought here should therefore best reflect technological leadership. On the other hand, European firms are more familiar with the European market. There is a home bias effect in patenting (Criscuolo, Citation2006).
4 Against this background, analyses of regions’ specialization across sectors frequently rest upon a specification of a vector of factors related to the productivity in each sector and an assessment of the relative endowments of such factors across regions.
5 Learning-by-exporting has also been advanced as an alternative explanation for the empirical regularity that exporters are more productive than non-exporters, but does not preclude self-selection. The hypotheses are not mutually exclusive (Wagner, Citation2007).
6 See Fagerberg Citation(1996) and Dosi and Soete Citation(1988) for reviews of related literature.
7 Examples of empirical studies that point to these kinds of effects are Bernard and Jensen Citation(1999), Castellani Citation(2002), Castellani and Zanfei Citation(2003) and Criscuolo et al. Citation(2004).
8 Their findings do, however, not preclude ‘learning-by-exporting’. Moreover, the two hypotheses are not mutually exclusive.
9 In this sense, own R&D can potentially be seen as a prerequisite for materializing ‘learning-by-exporting’ effects.
10 By ranking all goods such that lower values of x corresponds to higher home-relative productivity, the home region will specialize and export those goods whose index is lower than a threshold value [xtilde]. Home exports all goods for which x<[xtilde]. At the threshold, [xtilde], where wa([xtilde])=w* a* ([xtilde]), trade is indeterminate.
11 The production of products or development of product innovations with high θ, i.e. high quality, is as aforementioned, assumed to require a high technology specialization.
12 A similar implication follows from the so-called lead market literature (e.g. Beise Citation2004), which asserts that a concentration on certain technologies (or products) in an area can be attributed to a high willingness-to-pay on behalf of the consumers in the same area. Firms that succeed in such an environment have then a strategic advantage that can be used to penetrate export markets.
13 We thank Bart Verspagen for providing EPO data divided by country. The material originates from Maurseth and Verspagen Citation(2002).
14 EPO was started in the late 1970s, so using the first few years may bias counts downward. Also, using data after 1999 is likely to lead to truncation biases, since most patents issued after that date have not yet received many citations.
15 This concordance can be found on Eurostat's Ramon project homepage: http://europa.eu.int/comm/eurostat/ ramon/index.cfm?TargetUrl=DSP_PUB_WELC
16 Some studies normalize the specialization indices used in this study through a monotonic transformation, such that the specialization indices are bounded between−1 and 1 (e.g. Malerba and Montobbio, Citation2003). The analyses presented in subsequent Sections do not apply this transformation. However, results with transformed specialization indices are identical to the ones reported in the paper. These results are available from the authors upon request.
17 The second variant is equivalent to the location quotient.
18 Max(P i ), for instance, describes how much patenting occurs in the most patent-intensive sector.
19 Since the CV-measures are rather similar across variables, we only present the coefficient of variation for weighed patenting in order to conserve space.
20 We ran a few simple linear regressions between amount of patenting and specialization, which confirmed a statistically significant negative association between the two.
21 A different paper by Ejermo Citation(2006) describes the Swedish regional distribution of unadjusted and quality-adjusted patenting.
22 This technological stability of Swedish regions is documented in Andersson and Ejermo Citation(2004) and Ejermo Citation(2005).
23 Mean, median and standard deviations of the categories are presented in Appendix B.
24 In view of this, we focus on the citations-weighed specialization measures in the subsequent empirical analysis.
25 It could be argued that the number of exporting firms as an explanatory variable for exports is endogenous. However, the results presented in this Section are not affected if we instead use the regional production value. Results with regional production values are available from the authors upon request.
26 GDP data was extracted from World Development Indicators 2005. We do not use Purchasing Power Parity (PPP)-adjusted Figures, as international trade transactions are conducted according to nominal exchange rates.
27 Appendix C shows the formula that was used for calculating the distance with latitudinal and longitudinal data.
28 The model is one-sided since it only includes exports from regions and not their respective imports.
29 Results based on estimating x ri, s on the RHS are omitted here but are available from the authors upon request.
30 The variance inflation factors associated with the variables indicate that multicollinearity is not a problem for any of the variables in the model.
31 The Alchian–Allen conjecture is that the presence of a per unit transport cost lowers the relative price of high-quality products (see Hummels and Skiba, Citation2004).
