Guillaume Gaulier

BACI: International Trade Database at the Product-Level (the 1994-2007 Version)

This paper documents the construction of BACI, our international trade database, which covers more than 200 countries and 5,000 products, between 1994 and 2007. New approaches have been developed to reconcile data reported by almost 150 countries to the United Nations Statistics Division, collated via COMTRADE. When both exporting and importing countries report to Comtrade, we have two different figures for the same flow, so it is useful to reconcile these into a single figure. To do this, firstly, as import values are reported CIF (cost, insurance and freight) while exports are reported FOB (free on board), transport and insurance rates have to be estimated and removed from import values. We regress the observed CIF/FOB ratios for a given flow on gravity variables and a product-specific world median unit value. In a second step we evaluate the reliability of countries reporting. We decompose the absolute value of the ratios of mirror flows using a (weighted) variance analysis. These measures of the reliability of reported data are used as weights in the reconciliation of each bilateral trade flow which is reported twice. Taking advantage of this bilateral information on each flow, we end up with a large coverage of countries and more reliable data, especially in terms of unit-values. BACI is freely available online to users of COMTRADE database, in different product classifications.

Specialization Across Varieties and North-South Competition

Analyzing a new database that makes it possible to disaggregate trade flows across many countries according to unit values, we show that international specialization in terms of quality within industries and product categories plays an important role in the dynamics of North-South competition. The different specialization of countries at different levels of development within products and across varieties is mirrored in the recent shifts in world market shares, which are very different across quality segments: the South is not gaining market share in high-value portions of trade pattern. In this respect Europes specialization pattern appears to be different from that of the US and Japan, and may allow it to better resist the competitive pressure of the South.

Specialisation Across Varieties Within Products and North-South Competition

Recent works in trade theory and related empirical studies have drawn a revised picture of trade patterns: international specialisation has been proved to take place within products, across varieties, rather than across products or across industries. Systematising this repeated empirical evidence, we ask here what are the precise patterns regarding the specialisation of countries within products and across varieties and what are the determinants of such specialisation. Our value added is twofold. Firstly, we use BACI, the new CEPII data base of world trade covering the largest available set of countries over a decade at the most detailed level of the product classification. BACI reconciles the declarations of trading partners to the United Nations (COMTRADE), extracting trade costs from unit values of imports, and correcting for the quality of the declarations. We consider varieties of products inside each heading of the 6-digit level of the harmonised nomenclature, which comprises some 5,000 products. Secondly, we take advantage of this extensive coverage to systematically address the determinants of specialisation using a 10 year panel of 163 countries and 25 manufactured sectors.Our results point to four stylised facts. Firstly, the similarity of exports between North and South is much more limited when we consider differentiated varieties than when industries are considered. Secondly, the unit value of exported products to a certain market varies with the level of development of the exporter. Thirdly, the observed redistribution of market shares has been especially detrimental to advanced economies for low unit value varieties, while the EU has better resisted competition in high unit value varieties. Fourthly, we use a gravity equation to explain the bilateral trade in varieties among developing and developed economies.On the basis of such detailed and systematic empirical evidence regarding the specialisation of countries within - rather than between - products, we ask whether the fears raised by North-South competition are exaggerated.

International Comparisons of Living Standards by Equivalent Incomes

We propose a measure of living standards for international comparisons. Based on GDP per capita, the measure incorporates corrections for international flows of income, labor, risk of unemployment, healthy life expectancy, household demography and inequalities. The method for comparing populations that differ in some non-income dimension consists of computing the equivalent variation of income that would make each population indifferent between its current situation and a reference situation with respect to the non-income dimension. This is applied to 24 OECD countries. The obtained ranking of countries differs substantially from the GDP ranking.

World Trade Competitiveness: A Disaggregated View by Shift-Share Analysis

We develop an improved shift-share methodology and employ it to estimate the trade competitiveness of 88 world countries during the 1995-2002 period and to identify factors that drove each country’s increase or drop in exports market share. Along with the export competitiveness, we consider the geographical and sectoral dimension of countries’ initial position on different import markets and of their capacity to adapt to shifts in the world economy. Differently from the traditional method employed in the literature, our procedure yields identical results regardless the order in which trade is decomposed in geographical and sectoral factors. Moreover, it produces standard errors and permits to evaluate the statistical significance of each effect.During this period, the strong demand for elaborated products has benefited the developed countries and disadvantaged most of the South, especially the least-performing countries. Despite the unfavorable sectoral breakdown of the global import demand, exporters from the South recorded remarkable gains in market share over the 1995-2002 period. The emergence of the South, not just as a supplier of the North but as well as a rapidly expanding market, was mainly driven by Central and East European countries along with Turkey on one hand, and emerging Asian countries on the other hand. These gains are mainly explained by the competitiveness of their exports, which largely compensated for disadvantages linked to their sectoral specialisation. In Latin America, the improvement in competitiveness just offsets the major handicap in sectoral specialisation, without allowing most of the countries to maintain their market share. African and the Middle East countries accumulate both unfavourable sectoral and geographic specialisations, as well as a poor adaptation to the markets’ dynamics.The EU recorded the best performance in the North by maintaining its market share, especially on the domestic market, regardless the emergence of Southern exporters. On the contrary, competitiveness and market share losses by the US and Japan are significant and of similar magnitude. Japanese exports have suffered from the general weakening of the domestic economy, while the high level of the dollar in the beginning of the period has reduced the competitiveness of American products. For both countries, however, such sluggish competitiveness was slightly compensated by other factors: geographical advantages for the US and sectoral advantages for Japan.

Exchange‐rate pass‐through at the product level

We use a detailed database to investigate exchange-rate pass-through at the product level for a large number of countries. The empirical analysis suggests that pricing behaviours are dichotomous, with complete pass-through in around 25% of sectors and significant pricing-to-market in the remaining ones. Average long-run pass-through coefficient is close to 80%; this result hides a strong heterogeneity of pass-through behaviours across sectors. Even when composition effects are controlled for, average pass-through varies across importing countries. The econometric analysis shows that pass-through tends to be higher in volatile environments; in less developed countries; in weakly integrated markets.

Financial constraints and foreign market entries or exits: firm-level evidence from France

This paper studies the effect of credit constraints on the expansion and survival of firms in foreign markets. It develops a model in which, lower access to external finance, or reduced internal liquidity, hampers the firm ability to finance the recurrent costs to serve foreign markets and decreases firm survival in foreign markets. Additionally, financial constraints act as a barrier to firm export expansion by decreasing the firm ability to finance the entry costs into new export markets; thus, they push firm to avoid losing destinations. We use a unique longitudinal dataset on French firms that contains information on export destinations of individual firms and allows us to construct various firm-level measures of financial constraints to test these predictions. We obtain two main results. First, credit constraints have a negative effect on the number of newly served destinations. Second, higher probability of exit from the export market is also associated with credit constraints; that is consistent with constraints limiting the financing of recurrent export costs.

International Trade Price Indices

Export and import price indices are useful instruments in international economics. We document here TradePrices, our database of aggregated and sectoral trade price indices for all countries, computed using unit values given by BACI, the CEPII’s database of international trade at the product-level covering the period 1995-2004. Its rich country dimension allow an international comparison of prices evolutions. We compute “common” Laspeyres and Paasche indices but also “superlative” Fisher and Tornqvist indices, in both chained and fixed-base forms (chained Laspeyres and Paasche are also provided in their geometric form). In a first part, we discuss the characteristics of these different aggregation methods. In particular, we highlight the links between statistical methods and economical assumptions about implicit elasticity of substitution between goods. In a second step we describe the data. An exhaustive sensitivity analysis is done in order to determine the appropriate way to deal with technical difficulties, such as the treatment of measurement errors in bilateral unit values. Finally, this paper provides some stylized facts illustrating the necessity to provide a trade price database allowing international comparison and the usefulness of both aggregated and sectoral trade price indices to study international economics. TradeP rices database and SAS programs are freely available online, offering to users the possibility to choose the most accurate index for each particular purpose.

In the Wake of the Global Crisis: Evidence from a New Quarterly Database of Export Competitiveness

Over the past two decades, international trade has become a privileged engine of growth for much of the developing world. With the global economy evolving continuously and rapidly, countries must pay close attention to their positioning on the map of global trade and production. Within this framework, countries must also become aware of how they fare relative to competitors and to their past export performance. Of particular importance is the extent to which their performance is driven by exporter own supply-side capacity as opposed to external or compositional factors, including product and geographical specialization and how these trends compare across countries. This paper describes a new initiative that uses quarterly data for 2005q1-2013q1 to compute comparable indicators of export performance for 228 countries and territories. The database, the Export Competitiveness Database, reveals interesting patterns in trade performance. Export performance, stripped of compositional effects, was strongest for countries from the Asia and Pacific region, on average. Moreover, such performance was almost entirely driven by exporting country specific factors, with changes reflecting growth in volume rather than price developments. All emerging and developing regions have, on average, improved export performance. The indicators in the database trace the legacy of supply-side capacity and the overall export performance of the double-dip recession in the euro area. An illustrative set of results suggests that the papers measure of competitiveness correlates to a countrys nominal and real effective exchange rate, factors that are commonly perceived as important determinants of competitiveness.

Market Shares in the Wake of the Global Crisis: The Quarterly Export Competitiveness Database

Over the past two decades, international trade has become a privileged engine of growth for much of the developing world. In the wake of the global crisis, countries must pay close attention to their positioning on the global map of trade and production and become aware of how they fare relative to competitors and to their past export performance. To which extent changes in their market shares are driven by exporter own supply-side capacity as opposed to external or compositional factors, dues to their product and geographical specialization? This paper uses quarterly data, covering all exchanges flows at the product level since 2005, to compute indicators of export performance stripped of compositional effects. The resulting Export Competitiveness Database (ECD) reveals that emerging and developing regions, particularly the Asia and Pacific one, had strongest capacity to gain market shares in the most recent period, with changes reflecting growth in export volumes rather than price developments (once controlled for the composition effects). In contrast, ECD indicators also trace the legacy of the double-dip recession in the euro area, which have turned into negative the geographical effects of the traditional intra-zone specialization, despite the generally positive effects of sectoral structure. These measures of competitiveness correlate to nominal and real effective exchange rates, factors that are commonly perceived as important determinants of a country’s export competitiveness.