The myth of decoupling

Sébastien Wälti 27 July 2009

a

A

The decoupling hypothesis is the idea that business cycles in emerging market economies have become more independent of – or decoupled from – business cycles in advanced economies in recent years. Proponents of the decoupling hypothesis argue that emerging markets have achieved stronger growth in domestic demand, lowering the relative contribution of net exports to economic growth. Emerging markets have allegedly also managed to strengthen domestic policy frameworks and reduce external vulnerabilities, thereby increasing the scope for counter-cyclical policies to mitigate the impact of external shocks. Together, these developments would imply that emerging markets have decoupled from advanced economies.

The decoupling hypothesis runs against the idea that globalisation, through stronger trade and financial linkages across countries, facilitates the international transmission of country-specific shocks, thus leading to greater business cycle synchronisation. Since emerging markets have become more integrated into the world economy, we would expect greater business cycle synchronisation. Not surprisingly, the decoupling debate has been fading as the global crisis unfolded.

Surprisingly, it is now resurfacing. Some argue that emerging markets will emerge from the global crisis sooner and faster than advanced economies, as the former are now the so-called “new engines of global growth”. While it is still too early to get a good picture of the global recovery process, there is enough historical data to assess the decoupling hypothesis for the period before the outbreak of the current crisis.

Research on decoupling

The empirical analysis by Ayhan Kose, Christopher Otrok, and Eswar Prasad (2008) is perhaps the most prominent contribution to the debate. They decompose the real GDP growth rate in over 100 countries into three factors:

  1. a global factor, which picks up fluctuations that are common to all countries,
  2. an economy-type factor, which captures fluctuations that are common to countries within each of the three types of economies advanced, emerging, and developing, and
  3. a country-specific factor.

These factors are estimated over two time periods, the pre-globalisation period (1960-1984) and the globalisation period (1985-2005). The findings offer a nuanced view of the decoupling hypothesis. During the globalisation period, the share of the variation in growth rates explained by the global factor decreased. At the same time, the share explained by the economy-type factor increased for both advanced and emerging market economies. Put simply, there has been decoupling between advanced economies and emerging markets, but stronger convergence of growth rates within both groups.

The results of this factor analysis offer another interpretation, however. Retrieving factors from real GDP growth rates ignores one important development that occurred during the second time period (1985-2005): emerging markets experienced a massive take-off in economic growth. As a result, trend growth rates in advanced economies and emerging markets have diverged significantly (Figure 1).

Figure 1. GDP growth rates and trends

Source: World Economic Outlook Update, IMF, 6 November 2008.

Because trend growth rates diverged during the second time period, it must be the case that the share of the variation in actual growth rates explained by the global factor has decreased, and that the share explained by the economy-type factor has increased. This result obtains regardless of the evolution of business cycle synchronisation. In fact, if we look at deviations of actual growth rates from trend growth rates in Figure 1, they appear to be very similar for both advanced and emerging market economies.

The bottom line is that any assessment of the decoupling hypothesis should not focus on actual growth rates, but rather on deviations from trend. Looking at actual growth rates can be misleading. Saying that China has decoupled from the US because China grows at 5% while the US experience an output decline of 2% is wrong. If the trend growth rate is 9% in China and 2% in the US, both countries are 4 percentage points below trend and their business cycles are therefore perfectly in tune. This is a hypothetical example, but it makes the point.

So, have emerging markets decoupled from advanced economies in recent years? The discussion above provides some guidance for the choice of an empirical approach. First, we should look at deviations of real GDP from trend. Second, decoupling is often thought to have occurred in recent years. Therefore, we should construct annual measures of business cycle synchronisation. Mink, Jacobs and de Haan (2007) have proposed a simple concordance measure of business cycle synchronicity. This measure is computed as the product of the output gap of two countries, or two groups of countries, divided by the absolute value of this product. When two countries have coinciding output gaps, their business cycles are synchronous and the measure is equal to one. When one country has a positive output gap and the other country has a negative output gap, the measure is equal to minus one. The major advantage of this approach relative to factor analysis or correlation coefficients is that the measure varies year to year.

Using real GDP data from the World Economic Outlook database, I assess the degree of synchronicity between 34 emerging market economies and four groups of advanced economies: all advanced economies, the G7 group, the US, and Europe (Wälti 2009). The sample runs from 1980 to 2007. Synchronicity is assessed vis-à-vis all emerging markets together, geographical regions of emerging markets, and individual emerging markets. Figure 2 shows the evolution of the synchronicity measure between the group of all emerging markets vis-à-vis the four groups of advanced economies. Since the concordance indicator exhibits time variation, we also show the time-varying trend of synchronicity.

Figure 2. Business cycle synchronicity

Source: Wälti (2009).

Figure 2 shows that the degree of business cycle synchronicity between emerging markets and advanced economies has not decreased in recent years. The evidence on individual emerging markets shows that there is no country (except for Peru) which reports a general decline in its degree of synchronicity with all four groups of advanced economies.

Conclusion

My analysis, which takes account of the change in emerging markets’ trend growth rates, suggests that decoupling was always a myth. This is not to say that decoupling cannot happen in the future. Indeed, the evidence also shows that synchronicity displays significant time variation. But the results support the view (and is consistent with extensive empirical evidence) that globalisation brings national business cycles closer together.

Note: The views expressed in this column are those of the author and do not necessarily represent those of the Swiss National Bank.

References

Kose, A., Otrok, C. and Prasad, E. (2008), “Global business cycles: convergence or decoupling?”, NBER Working Paper 14292, and “Dissecting the decoupling debate”, VoxEU.org, 4 October 2008.

Mink, M., Jacobs, J. and de Haan, J. (2007), “Measuring synchronicity and comovement of business cycles with an application to the euro area,” CESifo Working Paper 2112.

Wälti, S. (2009), “The myth of decoupling”, manuscript, Swiss National Bank.

a

A

Topics:  Global economy

Tags:  business cycles, emerging markets, decoupling

Economist at the International Monetary Relations Unit of the Swiss National Bank (SNB), and a Research Associate of the Institute for International Integration Studies at Trinity College Dublin