The evolution of development with trade in global value chains

We propose to use canonical correspondence analysis (CCA) as a way to summarize the main trends in the dynamics of trade, global value chains and development over the period 1995 – 2018. CCA is a descriptive method that extends the algorithm (non-canonical correspondence analysis) that is widely used for calculating the economic complexity index. Both techniques (CCA and economic complexity) are aimed at reducing the dimensionality of large cross-country datasets on international trade. CCA has the advantage that the correlation between the derived indicator(s) to a set of underlying economic variables (in our case at the country level) is included in the derivation of the summary indicators. This facilitates the use of >1 dimensions to summarize the trade dataset. We illustrate this by relating the summary trade indicators (CCA dimensions) to a set of variables about integration of countries in global value chains, as well as a number of general indicators about development. The results indicate a trade-off between general GVC integration and a specialization in supplying intermediates to the global economy. We construct dynamic trajectories that show how individual countries or groups of products (such as high-, medium- and low-tech) navigate this trade-off over time.