Trade Openness Reduces Financial Vulnerability

"Economies that trade less with other countries are more prone to sudden stops and to currency crises."

Many a debate over the merits of free trade focuses on whether a country that is more exposed to international markets makes itself more vulnerable to a form of economic whiplash known as a "sudden stop." The term refers to situations in which a country experiences an abrupt cessation of foreign investment, which can precipitate a currency crisis. Some argue that an economy deeply integrated into the global market is at high risk of suffering shocks from abroad. Others contend that free traders actually have an easier time withstanding tremors that occur in the world of international finance. Different economic studies have offered support for both points of view.

NBER Research Associate Jeffrey Frankel and co-author Eduardo Cavallo believe that these studies have failed to resolve the debate because they do not properly isolate the effect of variation in countries' openness to trade. In Does Openness to Trade Make Countries More Vulnerable to Sudden Stops or Less? Using Gravity to Establish Causality (NBER Working Paper No. 10957), they find that a more comprehensive analysis of trade openness clearly shows that trade can lower one's vulnerability to externally induced crises.

"Some may find this counterintuitive: trade protectionism does not 'shield' countries from the volatility of world markets as proponents might hope," they write. "On the contrary, less trade openness leads to greater vulnerability to sudden stops."

They point out that a "conservative estimate yields a surprising result": a country that increases the value of its trade from 20 percent of its Gross Domestic Product (GDP) to 30 percent -- which in the real world would be going from Argentina's situation to Australia's -- "reduces the probability of a sudden stop by 32 percent."

Frankel and Cavallo note that in the past experts have examined the relationship between free trade and vulnerability to economic crises and reached conflicting conclusions. They point out that some studies "find that openness to trade is associated with fewer sudden stops" but others "find that openness helps trigger crises."

Most of these studies, irrespective of their conclusions, identify a country as open to trade through a single measure: the value of a country's trade expressed as a percentage of its GDP. Frankel and Cavallo point out that the ratio of a country's trade to GDP, while a potentially good measure, can be influenced by a variety of internal factors -- such as income and general policy reforms -- and may not necessarily succeed in identifying greater or lesser openness to trade.

Frankel and Cavallo seek to correct for this potential misreading of trade openness by looking at a broader range of trade-related information -- geographic characteristics and country sizes, which determine bilateral trade flows (a set of data technically known as the "gravity instrument") -- to confirm that a country is, in fact, correctly identified as open, or not open, to trade. When trade is a small share of GDP in a country that is landlocked or located far from the rest of the world, for example, one can be pretty confident that its low openness is genuine, and not simply a response to past financial crises.

This more thorough examination yields what the authors believe is a strong argument in favor of free trade as a way to reduce a country's vulnerability to economic setbacks. Frankel and Cavallo wanted to correct for any bias in previous studies that might have led to an overstating of the positive benefits of trade openness. But when they "corrected" for these potential problems, it only strengthened the argument for the insulating effect of openness and the pitfalls of protectionism. "In summary, the evidence appears to be quite robust," they conclude. "Economies that trade less with other countries are more prone to sudden stops and to currency crises.

-- Matthew Davis

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us