What do 70 years of data on world cycles tell us? The decoupling of prices and quantities

Chart 1: Contribution of world cycles to the variance of annual growth in domestic variables

Since the 2000s, numerous studies have argued that globalisation strengthens world cycles – defined as synchronised co-movements in economic variables across countries. According to this view, fluctuations in open economies are increasingly shaped by global factors – US interest rates, global financial conditions and commodity prices, for example – rather than by domestic shocks.

We revisit this narrative using a new quarterly macro-financial dataset covering more than 40 countries and spanning 1950 to 2019. Constructed using the International Monetary Fund’s paper archives, the dataset allows for a more systematic analysis of global co-movements of key variables (growth in GDP, credit, inflation, bond yields and stock prices) over time.


A disconnect between the synchronisation of prices and of real economies

Our empirical analysis reveals a striking divergence between the synchronisation of price variables (e.g. inflation and stock prices) and the synchronisation of quantities (e.g. GDP and credit). As shown in Chart 1, world cycles explain a much larger share of the variance of growth in prices (inflation, stock prices, bond yields) than that of GDP and credit growth. In other words, asset markets are highly synchronised at the global level, while real economies share a common cycle, but it explains a much smaller share of their fluctuations.

The finding is even more striking when we trace the synchronisation of these variables over time (Chart 2). Since the 1950s, the synchronisation of asset prices has increased steadily, while that of GDP and credit has remained stable – and even diminished after the global financial crisis. A breakdown by sub-periods shows that, during “normal” times, the world cycle explains less than 20% of the variance of domestic GDP growth in the median country.
 
We also examine the role of commercial and financial openness in shaping economies’ sensitivity to world cycles. First, we find that trade integration is associated with a higher contribution of the world cycle to domestic fluctuations in activity and asset prices: economies that are more open to global trade tend to be more synchronised with world economic and financial cycles.

The effects of financial integration are different. Greater financial openness is associated with stronger co-movement of asset prices with the world financial cycle, but weaker co-movement of activity with the world economic cycle. In financially open countries, asset prices are more aligned with global conditions, whereas GDP becomes less correlated with the world cycle.
 

Chart 2: Contribution of the world cycle to the variance of annual growth in domestic variables, by period

What explains the price-quantity divergence?

How can we explain why financial integration synchronises stock market prices without synchronising real economies? We propose a mechanism using a two-country general equilibrium model, in which countries choose their domestic and global asset portfolios and their production technology. Financial openness makes it easier to diversify and spread risk, which tends to align asset prices between countries. However, it also encourages countries to adopt more profitable yet more specialised production technologies, which are exposed to a higher risk. This dual trend – greater risk and greater specialisation – naturally reduces the synchronisation of domestic GDP.

What does this mean for economic policy?

This price-quantity disconnect challenges conventional ideas. First, world cycles do not require full financial openness: there was already significant asset price synchronisation under the Bretton Woods system, despite the existence of capital controls.
Second, although our analysis does not allow us to isolate the precise role of national policies in the observed co-movements of prices and activity, they do shed light on the choices central banks might face. Inflation appears increasingly driven by global factors, whereas activity remains largely shaped by domestic trends.

This divergence can complicate the conduct of monetary policy: simultaneously stabilising traditional targets (inflation and the economic cycle) means dealing with forces that are becoming progressively less correlated. This suggests that monetary policy choices i.e. the relative weight given to stabilising inflation and stabilising activity, must take account of whether the determining factors are domestic or global.