Official price statistics, which mostly reflect price surveys of importers, rely on the so-called overlap method to adjust price observations for changing product baskets. The implicit assumption is that price differences between old and new products entirely reflect differences in quality. Many problems remain: officially reported import prices, even when adjusted for quality, measure changes in the price for one unit of a product, while welfare analysis looks into the price per unit of unobserved utility.
The academic literature on the role of variety, quality, and taste in international trade is large and growing. Krugman (1979) emphasised the role of the ‘extensive’ margin in view of consumers’ love for variety, i.e. the importance of taking a differentiated set of goods into consideration (such as German and Japanese cars). Feenstra (1994) showed how to incorporate new product varieties into an aggregate of import prices. Building on this work, Broda and Weinstein (2006) found that changes in variety, if ignored, account for a significant upward bias in estimates of US import prices – yet this is still subject to the restrictive assumption that consumer tastes and product quality remain constant over time. This assumption may be increasingly problematic given rapid technological change and possible shifts in consumer preferences. More recently, many attempts were made to evaluate quality explicitly. For example, Khandelwal (2010) exploited price and quantity information to estimate the quality of products, while Di Comite et al. (2014) used unique firm-product-country data to capture product quality and consumer taste.
Traditional import price statistics understate welfare gains for European consumers from trade integration
In a recent paper we integrate changes in quality, taste, and variety into the calculation of import price indices for the four largest EU economies (Benkovskis and Wörz 2014). We follow the approach of Khandelwal (2010) in spirit, and combine it with Feenstra’s (1994) aggregate of import prices. Our results show that import prices in these four economies increased less over the period 1995 to 2012 when developments in quality and taste are taken into account (see Figure 1). This implies higher consumer welfare gains than suggested by official import prices.
Figure 1. Traditional and adjusted import price indices (1995 = 100)
We base our results on Eurostat data on imports made by Germany, France, Italy and the UK between 1995 and 2012, thus including recent data for the crisis and post-crisis years.1 We proxy for unobservable changes in relative taste and quality with observable unit values (price per kilogram) and volumes of imports.2
Some welfare gains from product variety
First, we focus on the gains from variety. Figure 1 reports that ignoring the increase in variety causes import price growth to be overestimated slightly – it causes welfare gains to be underestimated. However, the bias appears to be minor: cumulatively, the welfare gains between 1995 and 2012 amounted to only 0.5% of GDP for Germany and 0.1% for France, with the effect being neutral or even slightly negative for Italy and the UK. This is in sharp contrast to the cumulated 2.6% of GDP reported by Broda and Weinstein (2006) for the US over the period between 1972 and 2001 (average annual effects more than three times larger than those of Germany).
Impressive welfare gains from changes in consumer taste and product quality
Our decomposition shows that ignoring changes in consumer preferences and enhanced physical quality of imported products introduces a substantially larger upward bias in the price index under some plausible assumptions. The welfare gains from changes in taste and quality of imports impress: 10.6% of GDP in Germany cumulatively over the period of 1995 to 2012, 5.9% in Italy, 5.7% in the UK, and 1.4% in France. While one can dispute the preciseness and robustness of these results, the relative magnitude of reported figures is conspicuous. Product quality and consumer tastes are an important (but often overlooked) issue in international trade.
Another interesting observation relates to changes in consumer preferences and in the quality of imports during the financial crisis. In the four largest European economies the valuation of quality and taste of imports fell in 2008-2009 and recovered in 2010 (with a narrower gap between conventional and quality-adjusted price indices in 2009). We interpret this as a temporary shift to second-best, lower-quality products when the recession caused consumers to retrench. In other words, the decrease of prices appeared to be lower, but losses in welfare were higher than estimated without taking the taste and quality aspect into account. This is another example of taste and quality assessments altering the evaluation of welfare dynamics.
Some evidence on the role of taste in international trade
Di Comite et al. (2014) emphasise that the role of taste in international trade remains unexplored. By focusing on food products, we can obtain a rough indication of the role of taste versus quality. Unlike consumer preferences, the quality of individual food products is not expected to fluctuate a lot over time. Therefore, any variation in our parameter capturing relative taste and quality within individual food products should primarily refer to changes in consumer taste rather than changes in product quality.
Figure 2 shows the results for three basic food product categories. An upward sloping line indicates growing taste relative to the average taste for imported food and tobacco products. Indeed, in accordance with rising awareness for a healthy lifestyle, we find a pronounced shift in preferences toward fish products and fruits, while meat products have become less popular. Our rough calculations indicate that such shifts in consumer preferences have a major direct effect to consumers′ welfare. The fact that statistical offices do not (and cannot) account for changes in taste when constructing price indices reduces the usefulness of these indicators for welfare analysis.
Figure 2. Taste for selected imported products (relative to average taste for imported food and tobacco products, 1995 = 100)
This column argues for a comprehensive view on the gains from trade. Traditional import price statistics understate welfare gains for European consumers from trade integration. Import prices of major EU countries have risen since 1995, but so has the quality and valuation of these imports, often outpacing the price rise in terms of utility gains. Mirroring the underestimation of welfare gains from importing in past decades, our results also show that welfare losses in the recent crisis were in fact higher than traditional import price statistics suggest.
Another interesting question relates to the factors that may drive quality improvements in import goods, such as technological progress in the exporting countries (or even further up the global value chains). In this sense, production outsourcing from developed countries to emerging economies may serve as another explanation for the growing quality of European imports. Many hi-tech products which used to be assembled in Europe 10 or 15 years ago are now assembled in China, using hi-tech intermediate inputs from European countries. Interestingly, the global integration of production showed a temporary decline during the financial crisis, which coincides with our results in Figure 1 (see Stehrer et al. 2012).
In conclusion, we want to stress the importance of focusing on non-price factors. The policy debate is often too narrowly focused on price developments, which are simple to calculate and uncontroversial in interpretation, when in fact a much more differentiated and holistic view is needed.
Benkovskis K and J Wörz (2014), "How Does Taste and Quality Impact on Imports Prices?", Review of World Economics, DOI: 10.1007/s10290-014-0196-3.
Broda C and DE Weinstein (2006), "Globalization and the Gains from Variety", Quarterly Journal of Economics, 121(2), 541–585.
Di Comite F, JF Thisse and H Vandenbussche (2014), "Verti-Zontal Differentiation in Export Markets", Journal of Economics, 93(1), 50–66.
Feenstra, RC (1994), "New Product Varieties and the Measurement of International Prices", The American Economic Review, 84(1), 157–177.
Khandelwal, A (2010), "The Long and Short (of) Quality Ladders", Review of Economic Studies, 77(4), 1450–1476.
Krugman, P R (1979), "Increasing Returns, Monopolistic Competition, and International Trade", Journal of International Economics, 9(4), 469–479.
Stehrer R, N Foster and G de Vries (2012), "Value Added and Factors in Trade: A Comprehensive Approach", Vienna Institute for International Economic Studies Working Paper No 80, June.
1 The data are derived from Eurostat′s Comext database and thus based on the common EU nomenclature with the eight-digit subheadings used in export declarations.
2 Our framework is based on maximising consumer utility, which is modeled by a nested CES utility function wherein consumers can choose between domestic and imported goods and different varieties of individual import goods. See Benkovskis and Wörz, (2014) for details of the methodology.