A quarterly magazine of the IMF

September 2005, Volume 42, Number 3

Exploring the impact trading partners have on each other's growth

Economic integration among countries has increased significantly in recent decades, with the share of world trade in GDP rising from an average of 26 percent in the 1960s to 42 percent in the 1990s. Common wisdom holds that as countries become more integrated, they are increasingly influenced by each other’s economic development. Some economies are even considered to be engines of global or regional growth. Countries that trade a lot also tend to grow faster—a link that has been extensively documented.

But just how large is the impact of external economic conditions on a country’s growth? And as the world economy becomes more integrated, has the importance of growth spillovers increased? We undertook three studies to try to answer these questions. One study analyzed trade-related growth spillover in over 100 industrial and developing countries. The other two studies sought to assess the impact of the United States and South Africa on the growth of other countries.

Our results show that economic conditions in trading partners do in fact matter significantly for growth. After controlling for other growth determinants, we found that a country’s economic growth is positively influenced by both the growth rate and relative income level of its trading partners. Our findings also suggest that countries benefit relatively more if their trading partners grow faster than they themselves do and are richer. And we found evidence that some countries are indeed engines of global or regional growth: the impact of the United States is significant in many countries around the world, and South Africa matters for economic growth in the rest of Africa. In all three cases, we found the estimated impact of growth spillovers to be relatively large. It has been larger in recent decades and for open economies, implying that international spillover effects may increase in importance as globalization continues.

What theory tells us

Economic conditions abroad—including growth rates and income levels—are thought to influence a country’s growth through several channels.

• The most obvious channel is trade linkages: a rise in trading partners’ growth leads to an increase in their demand for imports, which then contributes directly to an increase in the net exports of the home country. And the positive implications of trade for economic growth are not limited to countries that run surpluses, since countries can benefit from technology transfers and other efficiency gains associated with international trade (Coe and Helpman, 1995).
  
  
• With growing foreign direct and portfolio investment, the spillover effects of trading partners may also be transmitted through financial linkages.
  
  
• Finally, there may be indirect effects, with business and consumer confidence in major countries influencing confidence in other countries.

Testing, testing

Using panel data from 101 industrial and developing countries (Arora and Vamvakidis, 2005a), we sought to test the proposition that trading partners affect each other’s growth, and to quantify this effect.

The analysis comprised a panel regression, both with and without fixed effects, using five-year averages for 1960–1999. The dependent variable was the average per capita real GDP growth for each country. The independent variables comprised variables that have traditionally been considered important determinants of growth, such as measures of convergence, investment in physical capital, human capital, macroeconomic stability (inflation), and openness to international trade. To test the importance of trading partners, the independent variables also included trading partners’ per capita real GDP growth and the ratio of the level of domestic and trading partners’ per capita real GDP. The results capture forces beyond short-term demand and business cycle effects, such as technology spillovers, since they are robust when we consider longer period averages (such as 10 or 20 years).

Analyses such as this one are often criticized on two accounts. First, for confusing the effects of global or regional shocks on countries’ growth with the effect of a trading partner economy. Second, for merely reflecting the "gravity model" of trade, according to which a country’s trade and growth is influenced by the size and distance of neighboring economies. To respond to these criticisms, some specifications of our model included world real per capita GDP growth, real per capita GDP growth in nontrading partner countries, and distance-weighted real per capita GDP growth.

We found that:

• Even after controlling for other growth determinants, a 1 percentage point increase in growth in a country’s trading partners is correlated with an 0.8 percentage point rise in domestic growth (see Table 1). This finding is not driven by common global or regional shocks, since the coefficient remains sizable and significant even after controlling for world growth, for growth in countries that are not trading partners, and for growth in neighboring countries.
  
  
• The positive impact on growth from faster growth in trading partners increases with openness, although this effect is relatively small.
  
  
• The ratio of a country’s GDP per capita to the average GDP per capita of its trading partners has a negative and statistically significant coefficient, suggesting that a country’s growth depends positively on how advanced its trading partners are relative to itself. That is, as a country closes the income gap with its trading partners, it will grow more slowly. In a similar fashion, a country that trades with relatively richer countries will grow faster. Spillover effects are generally assumed to be larger for developing countries because they benefit from the large knowledge stock of their more developed trading partners. One explanation could be that exporting to a more advanced country contributes to specialization in more sophisticated sectors, which may result in positive spillovers to other sectors of the economy. The relatively high content of advanced technological input in such sectors may also contribute to spillover effects.
  
  
• The importance of growth spillovers has increased with the expansion of world trade. Estimates for 1980–99 result in larger estimated coefficients compared with estimates for 1960–99 for both trading partners’ growth and relative GDP per capita. The results also suggest that both rich and poor countries benefit from trading with fast-growing trading partners.
  
  
• The construction of export weights for individual countries revealed that the relative importance of countries’ trading partners tends not to change much over time: the correlation of trade weights over time is relatively high and the set of most important trading partners also remains relatively stable (see Table 2). In addition, countries that trade with relatively rich countries in one decade also trade with relatively rich countries in the next decade. The same holds true for relatively poor countries.

Our results are robust to the criticisms of growth analysis mentioned earlier. The significant estimated impact of trading partners’ growth on domestic growth does not reflect trends in the regional or global economy that affect all countries. In addition, although a country’s growth is indeed positively correlated with growth in countries that are geographically close—which may reflect regional trends or the fact that countries simply trade more with countries that are nearby—the impact of trading partners on a country’s growth goes beyond these regional effects.

Let us now take a closer look at two economies believed to play the role of growth engine.

Global engine: the United States

Economists usually see the United States as an engine of the world economy: U.S. and world output are closely correlated, and movements in U.S. economic growth appear to influence growth in other countries to a significant degree. Certainly, given its size and close links with the rest of the world, the United States could be expected to have a significant influence on growth in other countries. In 2004, U.S. GDP accounted for over one-fifth of world GDP on a purchasing power parity (PPP) basis and for nearly 30 percent of world nominal GDP at market exchange rates. The United States accounted for nearly a quarter of the expansion in world real GDP during the 1990s. World and U.S. growth have moved closely together in recent decades, with a correlation coefficient of over 80 percent. Trade with the United States accounts for a substantial share of total trade in a large number of countries.

Estimates of the overall impact of U.S. growth on growth in other countries during the past two decades, in the context of a standard growth model, suggest that U.S. growth is a significant determinant of growth in a large panel of industrial and developing countries, with an effect as large as one-for-one in some cases (Arora and Vamvakidis, 2004). The impact of U.S. growth turns out to be higher than the impact of growth in the rest of the world. This could be explained by the role of the United States as a major global trading partner. The results are robust to changes in the sample, the period considered, and the inclusion of other growth determinants, including common drivers of growth in both the United States and other countries. We also found the impact of U.S. growth on growth in other countries to be larger than that of other major trading partners. For example, the impact of EU growth on the rest of the world is significant but smaller than the impact of U.S. growth.

Regional engine: South Africa

South Africa’s economic growth is believed to have a substantial impact on growth in other African countries because of the country’s relatively large economic size and its growing links with other African economies. In 2003, South African GDP was equivalent to nearly one-third of African GDP on a purchasing power parity basis and to 38 percent of African nominal GDP at market exchange rates. South Africa accounted for 30 percent of the expansion in African GDP (PPP basis) during 1980–2003, and African and South African growth moved closely together, with a correlation coefficient of over 80 percent. However, South Africa still accounts for a relatively small share of other African countries’ trade, reflecting the country’s isolation during the apartheid period before 1994, although South African financial flows are more significant.

Estimates of the spillover impact suggest that South African growth has a significant positive impact on growth in other African countries. A 1 percentage point increase in South African growth is associated with a 1/2–3/4 percentage point increase in the rest of Africa’s growth (Arora and Vamvakidis, 2005b). The results are robust to the inclusion of global and regional factors and other growth determinants, and to changes in the sample and the period considered. Moreover, estimating growth spillover effects for all other African countries, South Africa is found to be the only African country that plays the role of growth engine for the continent. The growth impact of other countries is either not statistically significant, or small and not robust.

In sum

With growing economic integration across countries, one would expect economic developments in a country to be influenced by developments in other countries. Our research has sought to estimate the extent of this influence. Our findings suggest that trading partners do indeed matter significantly for growth. In particular, it pays for countries to trade with relatively fast-growing and rich trading partners. Developing countries thus benefit from trading with industrial countries that are relatively rich, while industrial countries benefit from trading with developing countries, which tend to grow more rapidly. The results suggest that developments in countries’ major trading partners should be taken into account in growth forecasts and policy design.