Does the World Trade Organization Actually Promote World Trade?

05/01/2003
Summary of working paper 9347
Featured in print Digest

Very little happens to countries' trade openness upon joining the WTO.

The World Trade Organization (WTO) has become one of the world's most controversial multilateral organizations, perhaps rivaling only the International Monetary Fund as the favorite target for anti-globalization activists, who argue that the body's policies and rules favor trade at the expense of workers and the environment. But, according to the research of NBER Research Associate Andrew Rose, maybe everyone should stop worrying so much about the Geneva-based agency. Not only does the WTO not increase trade by member countries; it doesn't even produce more open trade policies among member states.

In Do We Really Know that the WTO Increases Trade? (NBER Working Paper No. 9273) and Do WTO Members Have a More Liberal Trade Policy? (NBER Working Paper No. 9347) Rose offers compelling counterpoints to the common wisdom regarding the global impact and effectiveness of the WTO and of its predecessor organization, the General Agreement on Tariffs and Trade (GATT). In the first paper, Rose estimates the impact of multilateral trade agreements on international trade using a standard "gravity" model of bilateral trade, which explains trade with the distance between countries and their joint income. He takes into account various factors that can affect trade, including culture (whether a pair of countries shares a common language), geography (whether the countries are landlocked), and history (whether one nation colonized another). He also adds the key variable of GATT/WTO membership: as of April 2003, membership had risen from 23 original founding members to 146 countries. Rose concludes that, taking all such factors into account, members in the trade body do not display significantly different trading patterns from countries outside the agency. "Countries joining or belonging to the GATT/WTO do not have significantly different trade from non-members" he writes.

Rose finds that traditional linkages among countries -- such as belonging to the same regional trade pact or sharing languages, borders, and colonial histories -- account for nearly two-thirds of the variations in trade. Above and beyond these gravity effects, membership in the GATT/WTO actually has a slightly negative (and statistically insignificant) impact on trade. "No reasonable person believes that membership in the GATT or WTO actually reduces trade," explains the author, "so I prefer to interpret the negative coefficient as a mystery rather than an indictment."

Rose seeks to solve the mystery in the second paper, which examines whether WTO members indeed display more liberal and open trade policies than non-members. In order to skirt the massive debate over how to measure openness in trade policy, Rose decides against choosing a particular indicator and instead looks at more than 60 different measures of trade policy. The measures can be broken down into seven groups: outcome-based measures of openness (such as the ratio of trade to GDP); trade flows adjusted for country characteristics; tariffs; non-tariff barriers (NTBs); informal or qualitative measures (such as World Bank measures of "trade orientation"; composite indexes, such as the Heritage Foundation's index of economic freedom; and measures based on price outcomes. The different measures cover periods ranging from 1950 to 1998.

Rose finds that very little happens to countries' trade openness upon joining the WTO. A typical country acceding to the WTO has an openness ratio (imports plus exports/GDP) of 73.1 percent five years before joining the organization. Five years after accession, the joining countries display openness ratios of only 70.4 percent. By the same token, tariffs actually rise (insignificantly) from 12.5 percent to 13.1 percent. One example: When Mexico joined the GATT in 1986, its tariffs averaged 6.4 percent of imports. Five years later, tariffs stood at 7.1percent. (Indeed, Mexico's tariff rates did not really start dropping until it joined the North American Free Trade Agreement in the 1990s.)

"It seems that none of the 64 measures of trade policy [openness] is strongly and consistently tied to GATT/WTO membership," explains Rose, with the exception of the Heritage Foundation's index of economic freedom. (WTO members tend to enjoy more economic freedom, as measured by that index.) The author also finds that the GATT repeatedly admitted countries that are relatively closed to trade, and allowed them to remain so for long periods after joining.

So, if the WTO neither increases trade by member states nor produces more liberal trade policy among members, then why does the organization exist? Rose cites other researchers who highlight various secondary WTO functions, such as "coordinating" trade policy among members (without necessarily liberalizing it) or serving as a disseminator of information. Finally, Rose leaves his readers with an intriguing possible explanation for the WTO's ineffectiveness. "Of course," he writes, "a weak international institution may be the deliberate result of its members."

-- Carlos Lozada