Searching for the sources of stabilisation in output growth areas: evidence from the G-7 economies

In his influential paper, “A new approach to the economic analysis of nonstationary time series and the business cycle”, Hamilton (1989) proposed a regime switching model in which output growth switches between two different states according to a first order Markov process. Applying this model to the U.S., he showed that shifts between positive and negative output growth accord well with the NBER’s chronology of business cycle peaks and troughs. In the wake of this paper, a large number of researchers have explored various aspects of the business cycle, such as asymmetry and the duration of economic fluctuations, using the framework of the Markov switching model. Lam (1990), Sichel (1993), Durland and McCurdy (1994), Kim (1994), and Kim and Nelson (1998, 1999a, 1999b) are examples of papers that have further analysed U.S. output. Simpson, Osborn and Sensier (1999) have modelled U.K. data, while Goodwin (1993) and Mills and Wang (2000) have analysed output from the G-7 countries. McConnell and Quiros (1999) have recently documented a structural break in the volatility of U.S. output growth, finding a rather dramatic reduction in output volatility in the most recent two decades relative to the previous three decades. Using yet a further extension of the Markov switching model, Kim and Nelson (1999b) propose a model that includes a separate state variable to capture an unknown structural break point. They use this model to investigate further the sources of stabilisation in recent U.S. output, focusing on both the decline in volatility and on the narrowing gap between mean growth rates during recessions and expansions. They find that both sources of stabilisation have a role to play, but with stronger evidence in favour of a narrowing gap between growth rates during expansions and recessions. Within the context of searching for structural change, our main objective in this paper is to answer the important question of whether this observed stabilisation in output is unique to the U.S. We thus adopt the Kim and Nelson (1999b) model to extend the empirical analysis to the G-7 countries and to present cross-country comparisons.