Useful things are sometimes found in unexpected places. Just as you can learn, from Chapter 35 of “Anna Karenina”:http://www.online-literature.com/tolstoy/anna_karenina/159/ how to make strawberry jam without water, so you can get a pretty good introduction to the static economic analysis of trade policies from the WTO’s latest annual World Trade report. It includes several “thematic essays”, one of which poses the question whether you can solve a dispute over WTO rights with a spreadsheet. I’m late in my recognition of the Report’s appearance: “Daniel Drezner”:http://www.danieldrezner.com/archives/002149.html, for example, praised the essay on “outsourcing” a couple of days back. But, emerging from the compilation of my own essay on “trade advocacy”:http://www.inquit.com/article/440/e‑book-on-trade-advocacy-nearly-complete, I took the opportunity to browse through the chapter on the use of economic models in WTO disputes. Well worth reading. The conclusions on the use of quantitative analysis are pretty conservative: * WTO dispute settlement is “above all about determining well-reasoned outcomes on the basis of agreed legal texts” * But “quantitative analysis may strengthen parties’ argumentation before panels and increase the comfort level of arbitrators in making an award … ” This is a view I share. There is just a whiff, however, of the exasperation that economists must feel when faced with the intuitions of economically naive lawyers and other “practical men”:http://en.wikiquote.org/wiki/John_Maynard_Keynes about the markets. “Quantitative economics”, the essayist adds “can help to avoid misinterpretation when economic rationality is counter-intuitive and less than obvious, but nevertheless pertinent to the substance or direction of legal reasoning.” Those intuitions have, after all, given us economic abominations of “fairness” and “due process” such as anti-dumping (and the infamous “Byrd amendment”:http://www.inquit.com/article/47/the-cost-of-the-byrd-amendment), mercantilist “concession”-bargaining, and the theory of Special and Differential treatment, that holds the equitable way to deal with trade adjustment costs for developing countries is to eliminate both the costs and the benefits of adjustment. You don’t need to read too closely between the lines of this essay to discover that the WTO Economics Division has less interest in analyzing the history of disputes than in shaping their future course. Their explanation of what “cross price elasticities” really mean—and under what conditions they mean anything—and their illustration of the use of regression analysis with detailed warnings about multicollinearity and autocorrelation are intended, I suspect, as a warning to trade policy makers who believe that you can build the new jerusalem in Geneva with a spreadsheet and access to the GTAP database. This may be a fond hope but it’s an important message. In the past decade, the dramatic improvement in the availability of trade data (on merchandise trade, at least) and the fall in the cost of computing power has led to a rapid escalation in the number and scale of economic models of world trade. Fascinating as they are, and valuable in their place, there is a terrible temptation to use economic models to develop counterfactual contentions in dispute cases: “had X complied with the Agreement, then it’s share of the market would have been no larger than Y …” This weak, and frequently misleading, use is actually encouraged by the provisions of e.g. Art.XVI:3 on subsidies (the serious prejudice rule).