International Synchronization of Growth Rate Cycles: An Analysis in Frequency Domain

This study examines international synchronization of growth rate cycles using spectral techniques in the frequency domain. In particular, we look at synchronization of growth rate cycles between bilateral country pairs, US, UK, Germany, Japan and India over the period from January 1974 to December 2018. We examine two aspects of the synchronization process—one, strength of co-movement across countries’ growth rate cycles, and two, sequencing in terms of leads and lags of these cycles vis-à-vis each other. The strength of co-movements is analyzed using coherences of growth rate cycles between bilateral country pairs across frequency bands and over time. The lead–lag structure between growth rate cycles of countries is obtained from the spectral phase shift parameter. This is evaluated against the lead–lag structure in the time domain, as inferred from the reference chronology given by the Economic Cycle Research Institute (ECRI). Based on the growth rate of the coincident index obtained from ECRI, we infer the sequencing of growth rate cycles in one country vis-à-vis the other in terms of the relative timing of their peaks and troughs. This comparative analysis across the time and frequency domains highlights both the pattern of lead–lag in terms of timing of peaks and troughs (time domain) as well as the lead–lag in terms of all phases of the cycle (frequency domain). For analyzing these patterns over time, we undertake the exercise over two subsamples: January 1974–December 1990 and January 1991–December 2018. We find that the coherence between developed country pairs is, in general, higher than that between developed–emerging economy pairs. We also find evidence of greater co-movement of country cycles post-1990, as compared with that in 1974–1990. The magnitude of leads–lags shows that the synchronization process is faster in the latter time period. The leads–lags obtained from the spectral phase shift estimates are found to be in line with those inferred from economic indicator analysis (EIA).