III. Southern euro area five Countries: Trends in Value-Added1
European companies compete more on differentiation, less on cost. Whereas 86 percent of them consider differentiation to be important, low cost is important for only 58 percent. Europe is focusing more and more on high value-added goods and services. Best of European Business competition survey, 2006
This chapter analyzes whether the southern euro area five (SEA-5) countries changed their production structure in a “competitive” way and compares the pace of restructuring to that of other industrial countries over the past 15 years. This is defined as a reallocation of resources towards more dynamic (fastest growing) and higher-tech sectors, which are associated with stronger competitiveness and higher growth potential. Such restructuring that could boost overall competitiveness toward more dynamic and higher-tech sectors over the past 15 years. The dynamic sectors are defined as those with the highest growth of real value-added in the world, and the technological ranking of each sector is defined following the OECD classification. In order to compare relative restructuring performance across countries, indices of dynamism and technological intensity were also constructed.
The results indicate that since the second half of the 90s the SEA-5 countries had been relocating resources toward more dynamic sectors but starting from a lower level and at a slower pace than the EU peers and the United States. In addition, the share of high-tech sectors in their production had increased. However, the SEA-5 (except France) still remains specialized in “traditional,” slow-growing and (generally) low-tech sectors. The slower pace of restructuring in the SEA-5 countries is found to be associated with the lower ability to respond to global growth opportunities.
B. Data and Methodology
The dynamics of the value-added is best seen in the firm-level data. However, since the aggregate trends are ultimately reflected in the industry-level despite firms’ heterogeneity (Hawanini et al, 2003) and due to cross-country data limitations, industry-level data are used. The data are from the EU KLEMS data base, spanning EU countries, Japan, and the United States, 61 industries (defined in accordance with the NACE two-digit classification), and the period 1981–2004.2 This paper will henceforth use terms industry and sector interchangeably. While the dataset contains both manufacturing and services sectors, this study concentrates on the tradable goods industries (manufacturing). In order to eliminate annual volatility, the analysis is performed using nonoverlapping five-year averages.
Changes in production structure are analyzed along two dimensions: dynamic and technological content, with respective indices constructed to evaluate relative performance across countries, as follows:
Dynamic content. The dynamic industries are defined as those with the highest rates of real value-added (RVA) growth worldwide. Specifically, the average growth rate of RVA was calculated for every industry across all countries for each year. Then, industries were ranked according to their five-year average growth rates, in a descending order (see Table III.1 for examples of best and worst performers). The dynamic content of the country’s domestic production was defined as a share of top 10 fastest growing sectors in country’s total domestic output (in nominal terms) for each five-year period. The dynamic content is a useful concept to analyze changes in each country over time. However, for the cross-country comparison of the relative pace of restructuring the effects of changes in relative prices between products produced by the industries with high and low RVA growth need to be mitigated. For that, the index of dynamism is defined, for each country, as the aggregate share of country’s top (bottom) 10 industries that grow faster (slower) than the world’s average for that industry. Negative (positive) values of the index indicate that the country is relatively under (over) concentrated in fast (slow) growing industries.
Technological content. The technological content of the economy is measured as a share of industries in four tech categories (high, medium-high, medium-low, and low) in domestic output. The technological classification of industries follows the OECD methodology (see Bauman and di Mauro, 2007, and references therein and Table III.2). As in the case of dynamic content, to facilitate the cross-country comparison, the index of technological intensity is constructed mitigating the effects of changes in relative prices between tech and nontech goods on the production structure. The index is defined, for each country, as a difference between aggregate share of industries in each tech category and the share of each category in global production. The value of index above (below) zero indicates that the country is relatively more (less) concentrated in any particular tech category than the rest of the world.
|1||Radio and TV receivers||1||Office, accounting, and computing machinery|
|2||Electronic valves and tubes||2||Radio and TV communication equipment|
|3||Radio and TV communication equipment||3||Telecommunication equipment|
|4||Telecommunication equipment||4||Water transport|
|5||Electrical engineering||5||Radio and TV receivers|
|6||Recycling||6||Motor vehicles, trailers, and semi-trailers|
|7||Office, accounting, and computing machinery||7||Electronic valves and tubes|
|8||Electrical and optical equipment||8||Other electrical machinery and apparatus, nec|
|9||Other electrical machinery and apparatus, nec||9||Electrical machinery and apparatus, nec|
|10||Electrical machinery and apparatus, nec||10||Electrical engineering|
|57||Chemicals, excluding pharmaceuticals||57||Food and beverages|
|58||Food and beverages||58||Textiles|
|59||Textiles||59||Wearing apparel/dressing/fur dying|
|61||Wearing apparel/dressing/fur dying||61||Leather/footwear|
|High-tech||manufacturing of electrical machinery|
professional and scientific equipment
office, accounting, and computing equipment
drugs and medicines, pharmaceuticals
radio, TV, and communication equipment
medical, precision, and optical instruments
|Medium-high tech||scientific instruments|
electrical machines excl. communication equipment
transport equipment, motor vehicles, and railroad
chemical excluding drugs
manufacturing of transport equipment
manufacturing of agricultural and industrial machinery
|Medium-low tech||rubber and plastic products|
shipbuilding and repairing
nonmetallic minarl products
|Low-tech||nonmetallic mineral products|
textile, apparel, and leather
paper products and printing
food, beverage, and tobacco
wood porducts and furniture
C. Result Number 1: The SEA-5 is Moving in the Right Direction but Slower Than Others
Becoming more dynamic…
The concept of dynamism as defined in this study follows closely the approach adopted in the export growth literature (see ECB, 2005 and references therein) associating the higher past growth performance with the higher future growth. It also presumes that higher growth of value-added, especially if sustained over a period of time, should be associated with a “brownian motion” of industry’s firms toward higher value-added products. Indeed, higher dynamic content is associated with better export performance (and ultimately, higher competitiveness), and the most dynamic industries include the most dynamic export sectors (defined in ECB, 2005), such as professional and scientific equipment, and manufacturing of electrical machinery.
The dynamic content of the SEA-5’s production increased steadily over the sample period, although at a slower pace than that of the peer countries. In particular:
The share of the most dynamic industries in the region’s production increased in the 80s and the 90s, albeit edging down lately (Figure III.1). Specifically, the SEA-5 countries increased their specialization in industries such as electrical machinery and apparatus (Italy); medical, precision, and optical instruments (Italy, Portugal); computer and related activities (Spain); office, accounting, and computing machinery, High-tech manufacturing of electrical machinery professional and scientific equipment aerospace office, accounting, and computing equipment drugs and medicines, pharmaceuticals radio, TV, and communication equipment medical, precision, and optical instruments electronic valves and tubes, radio and TV receivers, and other instruments (France). Although the share of dynamic industries in the SEA-5 remained below the EU and U.S. averages (both above 8 percent), with high regional variance—from 3 percent in Portugal to 7½ percent in Spain and France—but the progress is obvious. Moreover, the weight of the least dynamic industries, the SEA-5’s traditional specialization, declined steadily (Table III.3).
Moreover, the SEA-5 countries not only restructured across sectors but also boosted the dynamism within sectors (Figure III.2). In particular, the index of dynamism indicates that the region’s dynamism gap vis-à-vis EU and U.S. peers declined, as growth rates increased within country’s individual industries. Specifically, Italian clothing and footwear industry is one example of increased intra-industry dynamism, employing strategies for higher-growth, high-value-added products.
Figure III.1.The Dynamic Content
Sources: EU KLEMS; and IMF staff calculations.
Note: The dynamic content of the country’s domestic production w as defined as a share of top 10 fastest grow ing sectors in country’s total domestic output (in nominal terms) for each five-year period.
Figure III.2.Index of Dynamism
Sources: EU KLEMS; and IMF staff calculations.
Note: The index of dynamism is defined, for each country, as the aggregate share of country’s top (bottom) 10 industries that grow faster (slower) than the world’s average for that industry. Negative (positive) values of the index indicate that the country is relatively under (over) concentrated in fast (slow) growing industries.
…and technologically advanced
The link between technological innovation, growth, and productivity is well-established (OECD, 2006), and hence higher technological content of a country’s production would indicate an upgrade toward higher-value added products and processes. In fact, higher tech industries tend to experience higher growth of value-added (Tables III.1 and III.2). SEA-5 countries, except France, still score low in a number of indicators in this area (Figure III.3). Moreover, countries with the predominantly low-tech specialization of production—Greece and Portugal—are also found to have similar export structure (Bauman, di Mauro, 2007).
Figure III.3.Technological Content 1/
Sources: EU KLEMS; OECD; and IMF staff calculations.
1/ The technological content of the economy is measured as a share of industries in four tech categories (high, medium-high, medium-low, and low) in domestic output. The technological classification of industries follows the OECD methodology.
The results indicate that the technological content of the SEA-5 countries increased, but less so than in the peer group. In particular:
The relocation of resources toward high-tech industries was much slower than in the peer group.3 Indeed, while the EU average share in high-tech industries doubled in the last 20 years, SEA-5’s average improved much less so (Figure III.3). As a result, as of 2004, 12 percent of region’s output was produced by the high-tech industries, a far cry from the EU-15’s average of 18 percent and the U.S.’s 27 percent. Within the region, France performed on par with the EU-15, while Greece and Portugal were below the SEA-5’s average.
The concentration in the medium high-tech industries increased in France, Spain (both from already relatively high base), and Greece (from a very low base), but declined in Italy and Portugal. However, the decline in Italy was from already relatively high base to the EU average. These developments are encouraging, given the positive contribution of the medium-tech sectors to export market shares of the euro area (ECB, 2005).
The share of low-tech industries declined notably during1991–2004 in all SEA-5 countries, with France dipping slightly below the U.S. level, and Spain and Italy almost reaching the EU average. While Greece and Portugal also made significant progress, low-tech industries remain important in these countries.
The increase in the technological intensity of the SEA-5’s production had been slow, while the EU-15 countries achieved notable increases in all directions—reducing underconcentration in high-tech, further boosting advantages in medium high-tech, and reducing overconcentration in low-tech industries (Figure III.4). It is only during the latest period that the SEA-5 countries slightly reduced their gap in high-tech industries and boosted medium low-tech content. On the bright side, their overconcentration in low-tech products declined during 1995–2004.
The negative high-tech gap is common for all SEA-5 countries, but the degree varies, with France being least and Greece the most underconcentrated. As regards medium high-tech concentration, Greece and Portugal are the only two countries that remained in the negative territory, while Spain and Italy increased their already positive gap. Last but not the least, all countries lowered their overconcentration in low-tech industries over the last 15 years.
Figure III.4.The Index of Technological Intensity 1/
Sources: EU KLEMS; OECD; and IMF staff calculations.
1/ The index of technological intensity is defined, for each country, as a difference between aggregate share of industries in each tech category and the share of each category in global production. The value of index above (below) zero indicates that the country is relatively more (less) concentrated in any particular tech category than the rest of the world.
D. Result Number 2: Because of Slower Restructuring the SEA-5 May Be Missing Growth Opportunities
The ability of the country to restructure its production structure quickly is also associated with its ability to benefit from the industry-wide global growth opportunities. This hypothesis is formally evaluated using the results from the dynamic content section. In particular, the degree of restructuring is defined for each country and each period as a sum of a percent increase in the dynamic content and a decrease in the “nondynamic” content (calculated analogously to the dynamic content as a share of bottom 10 industries in terms of RVA growth in country’s output). Then, adapting the approach from the “finance and growth” literature to measure responsiveness to global opportunities (see Fisherman and Love, 2003), the growth rate of real value-added is regressed on the lagged global growth component and the degree of restructuring, along with other control variables. The regression results—in particular, positive sign of the interaction term between restructuring and lagged global growth—indicate that countries with higher degree of restructuring tend to experience higher growth in response to global growth opportunities (Table III.3). The results are similar for nominal value-added growth and are generally robust across industries, time periods, and country groups.
Bauman, U., and F.di Mauro, 2007, “Globalization and Euro Area Trade Interactions and Challenges,” ECB Occasional Paper Series No. 55 (March).
European Central Bank, 2005, “Competitiveness and the Export Performance if the Euro Area,” Occasional Paper No. 30 (June).
Hawanini, G., V.Subramanian, and P.Verdin, 2003, “Is performance driven by industry- or firm-specific factors? A new look at the evidence,” Strategic Management Journal, No. 24, pp. 1–16.
McGahan, A., and M.Porter, 2002, “What do we know about variance in accounting profitability?” Management Science, No. 48, pp. 834–51.
Prepared by Iryna Ivaschenko.
See Timmer, et al (2007) for a detailed description of the dataset.
This fact, however, may be also driven by the declining prices of some high-tech goods, since the mid-90s. To some extent, this concern is mitigated by the fact that the United States, which high-tech industry is likely more concentrated in computers than that of the SEA-5, did not experience a significant drop in its high-tech share.