Synchronisation Between South Africa and the U.S.: A Structural Dynamic Factor Analysis

Abstract

This paper studies the synchronisation of the South African and the US cycles and transmission channels through which supply and demand shocks from the US affect economic activity in South Africa in a structural dynamic factor model framework. We find, using the full-sample period, US supply shocks are transmitted to South Africa through business confidence and imports of goods and services; while US demand shocks are transmitted "via" interest rates, stock prices, exports of goods and services, and real effective exchange rates. Second, there is a decrease in integration over time translated by a drop in synchronisation of cycles. The impact of an increase in comovement of GDP is outweighed by the structural reforms initiated by the government after the end of apartheid. Finally, the idiosyncratic component still plays an important role in the South African economy. Copyright (c) 2009 The Author. Journal compilation (c) 2009 Economic Society of South Africa.

A large factor model for forecasting macroeconomic variables in South Africa

“…From a purely economic point of view, this paper, in conjunction with the work of Kabundi (2009), highlights the fact that not only can the FM be used to successfully analyze the degree of synchronization of South Africa with the US, but also the framework has tremendous potential for use as a forecasting tool relative to small-scale models, given its ability to handle large amounts of data on the wide range of variables that tend to affect a small open developing economy like that of South Africa, and specifically in our context the three key macroeconomic variables, namely the per capita growth rate, inflation, and the short-term interest rate. The fact that there exists a FM that tends to outperform both the naïve RW model and small-scale models that only account for the role of a particular variable or the interaction amongst the variables of interest, clearly highlights the possible model misspecification, in the sense that the latter set of models fail to utilize the effect of a large number of other variables which are not used in their estimation.…”

“…The forecasting performances of the FMs, estimated under alternative assumptions with regard to the interaction between the factors and the variables of interest, are evaluated and compared with the performances of three other alternative models, namely an unrestricted classical VAR, an optimal Bayesian VAR 2 (BVAR) and a New-Keynesian Dynamic Stochastic General Equilibrium (NKDSGE) model, on the basis of the Root Mean Squared Error (RMSE) of the out-of-sample forecasts. Although Kabundi (2009) used the DFM to assess the synchronization of South Africa and the US, and the channels through which the US supply and demand shocks are transmitted, to the best of our knowledge this is the first attempt to use a large FM to forecast key macroeconomic variables in South Africa. Moreover, it must be noted that, with the exception of Wang (2009), comparisons between a FM and a DSGE model are rare, but worthy of discussion, especially in the context of a developing economy like that of South Africa.…”

Regional Business Cycle Phases in Taiwan

“…Studies on regional business cycles developed in the 1920s and were pioneered by McLaughlin (1930), Vining (1945, 1946a, 1946b, 1949) and Isard (1949). In the last decade, many studies on the dynamics of the business cycle in different regions or countries have been conducted (Still, 1997; Clark and Eric, 1999; Rissman, 1999; Carlino and DeFina, 2004; Selover et al ., 2005; Owyang et al ., 2005, 2009; Hall and McDermott, 2007; Kabundi and Loots, 2007, 2010; Norman and Walker, 2007; Kabundi, 2009).…”

“…They found strong and significant evidence of co‐movement in the South African business cycle with that of Swaziland, Botswana, Zimbabwe, the Democratic Republic of the Congo, Lesotho and Angola. Recently, Kabundi (2009) used a structural dynamic factor model framework to investigate the synchronisation of South African and U.S. cycles and transmission channels through which supply and demand shocks from the United States affect economic activity in South Africa. Kabundi (2009) found that U.S. supply shocks are transmitted to South Africa through business confidence and imports of goods and services.…”

“…Recently, Kabundi (2009) used a structural dynamic factor model framework to investigate the synchronisation of South African and U.S. cycles and transmission channels through which supply and demand shocks from the United States affect economic activity in South Africa. Kabundi (2009) found that U.S. supply shocks are transmitted to South Africa through business confidence and imports of goods and services. U.S. demand shocks are transmitted via interest rates, stock prices, exports of goods and services, and real effective exchange rates.…”