Since the launch of the euro, Spanish exporters have been successful in containing the loss of their export share in world markets. This is in contrast to several advanced economies that have experienced significant losses as a result of globalisation and the gain of exports shares by many emerging countries. From 1999 until 2011, Spain lost 8.9% of her export share, a relatively modest figure in comparison to the record of other large producers (for instance France lost -40.5%; the UK -39.2%; Italy -32.1%; US -31.9%; and, more modestly, Germany -12.2%). Spain’s achievement is typically overshadowed by its dismal evolution of competitiveness indicators based on international relative prices.
Figure 1 shows that Spain´s price competitiveness deteriorated substantially in the run-up to the 2008 financial crisis, reflecting an unprecedented shift of resources towards non-tradeable sectors. The combined evidence of a modest market share loss, on one hand, and a sizeable real exchange rate appreciation, on the other, has been referred to as the ‘Spanish paradox’ (cf. Antràs et al. 2010, Crespo-Rodríguez et al. 2012). The literature has tried to explain the paradox by looking into firm level data. In fact, large Spanish firms experienced both lower unit labor cost growth and higher export growth than in other countries (Antràs et al. 2010), yet this differential performance is not reflected in aggregate price indicators due to aggregation and dispersion bias (Altomonte et al. 2012). To the extent that the size of exporting firms is much larger than that of non-exporting firms, this can explain why the appreciation of exports prices has been much less intense than that of relative unit labour costs.
Export market shares and price competitiveness
Figure 2 summarises two results relevant to the export performance of Western European economies, Canada, the US and Japan between 1999 and 20111. On the horizontal axis, we have represented the variation of exports prices from 1999 to 2011, relative to 36 industrialised economies, and on the vertical axis the variation of export market shares over the same period.
The first result is that there is no clear cross-country relationship between variations in relative export prices and variations in export market shares. In fact, price competitiveness gains are positively correlated with market share losses (although the correlation coefficient is low (0.22) and not statistically significant). This evidence does not imply that relative prices are not relevant for export markets shares, but that non-price determinants have been more important during this period and have more than offset the effects of export prices.
Figure 1. Spanish real effective exchange rates, alternative deflators compared to the rest of the 36 industrialised countries (1Q2000=100)
Source: European Commission
Figure 2. Change in the world share of exports and in relative exports prices for goods and services, 1999-2011
Source: Calculations based on European Commission and WTO.
Table 1. Decomposition of the variation of export market shares for goods and services, 1999-2011
Source: Calculations based on European Commission and WTO
Variations in export market share
From an analytical viewpoint, the variation of the export market share of a country, Δ(ex – exw), can be decomposed into the variation of the international relative price, Δ(ptx – ptw), and the variation of non-price determinants, Δsx, as given by the following expression:
Δ(ext – extw) = Δstx – σΔ(ptx – ptw),
where σ denotes the price elasticity of exports2. The red line with negative slope depicted in Figure 2 captures the values of the market share variation that can be entirely attributed to the corresponding movements in international relative prices (as a result, Δsx = 0) under the assumption of a price elasticity of exports equal to -1.253.
That is, the export share loss experienced by Greece since joining the European Monetary Union can be explained by the behaviour of its price competitiveness. However, for a country located on a parallel line to the red one, the evolution of other non-price factors helps us explain export share variation. That is, Δsx is either positive or negative. The evidence presented in the figure thus suggests that the evolution of non-price determinants has been more important than movements in international relative prices to explain market share variation among advanced economies. Germany, France, the UK and the US experienced similar depreciation rates but very different market share performances, from the 12% export market share loss of Germany to the 40% loss of France, as summarised in Table 1.
The second result in Figure 2 is that, given the appreciation and depreciation rates recorded in the sample, Spain displayed the most favourable evolution of the non-price determinants of export market shares – amounting to 6pp4. If Spain had experienced the real effective exchange rate depreciation of, say, Germany, her export market share would have increased 20 percentage points (equivalent to 6% of Spanish GDP).
Firm-level factors are behind non-price determinants of export market shares
A number of factors, largely ascribed to the realm of the firm's strategic decision making, have shaped Spain´s internationalisation process over the last two decades (Correa-López and Doménech 2012). On the one hand, we find factors related to company size, investment in capital stock, skilled labour intensity, research and development spending, technology adoption, and improvements in process innovations (Hanley and Monreal-Pérez 2012). On the other, we find decisions pertaining to market strategy and finance, such as product differentiation and diversification; product innovation; and the reliance on alternative sources to long-term financing, including foreign ownership. On balance, the benign combination of all these factors has produced important feedback effects, underpinning the relatively good performance of the Spanish export market share, the strong recovery of exports during the financial crisis, and the inroads made into emerging markets with new and differentiated products. Ultimately, they provide an explanation to the so-called Spanish paradox.
Adolfson, M, S Laséen, J Lindé, and M Villani (2008), “Bayesian Estimation of an Open Economy DSGE Model with Incomplete Pass-through”, Journal of International Economics, 72, 481–511.
Altomonte C, T Aquilante, and G I P Ottaviano (2012), “The Trigger of Competitiveness – The EFIGE Cross Country Report”, Bruegel Blueprint Series, Vol XVII.
Antràs, P, R Segura-Cayuela, and D Rodríguez-Rodríguez- (2010), “Firms in International Trade (with an Application to Spain)”, SERIEs Invited Lecture at the XXXV Simposium of the Spanish Economic Society.
Boscá J E, R Doménech, J Ferri and J Varela (2011), The Spanish Economy: A General Equilibrium Perspective, Palgrave MacMillan.
Correa-López, M, and R Doménech (2012), “The internationalisation of Spanish firms”, BBVA Research Working Paper, forthcoming.
Crespo Rodríguez A, G Pérez-Quirós and R Segura Cayuela (2012), “Competitiveness Indicators: The Importance of an Efficient Allocation of Resources”, Economic Bulletin, Bank of Spain, January.
Hanley, A and J Monreal-Pérez (2012), “Can Spain learn from its ‘export starters’?”. VoxEU.org, 5 November.
Ratto M, R Werner and J Veld (2009), “QUEST III: An Estimated Open-Economy DSGE Model of the Euro Area with Fiscal and Monetary Policy", Economic Modelling 26, 222-233.
1 For relative export prices the data source is the European Commission and refers to 36 industrialised countries. Data for export market shares come from WTO.
2 In open economy models, the demand for a country´s exports is typically expressed as:
EXt = Stx (Ptx / Ptw)-σ EXtw
where Px is the price level of exported goods, Pw is the price level of competitors' goods in world markets, σ is the price elasticity of exports, EXw is the world's demand for exports, and Sx is a time-varying variable that gathers all the factors that are relevant to the export market share different from relative prices. By taking logs (lower case letters) and first differences in the above expression, we obtain equation (1).
3 For EMU countries, Ratto, Werner and Veld (2009) estimate a price elasticity of exports equal to -1.25. This is a very similar estimate to the one reported in Boscá et al. (2011) for Spain (-1.30) but slightly below the range estimated in Adolfson et al. (2008). Notice that, for the ease of comparison, we are assuming the same elasticity for all countries in the sample, as represented by the slope of the red line depicted in Figure 2. Although elasticities may differ across countries (due to, e.g., compositional differences in exporting baskets), we do not expect that the differences in elasticities can explain a large part of the observed cross-country variability of export share variation, especially for the countries that combine real exchange rate depreciation with export share loss.
4 For the case of Spain, notice that using the real effective exchange appreciation based on unit labour costs (Figure 1) would increase the importance of residual factors in explaining export market share variation.