Risk sharing by financial markets in federal systems : what do we really measure?

This paper criticises the standard methodology used to measure the importance of different channels of risk sharing in federal states such as the one used in Asdrubali et al.'s (1996) seminal contribution. It argues that the methodology chosen in these papers systematically underestimates the role federal governments play in stabilizing the business cycle in its member states (and overstates the role of financial markets in stabilization) as it a) ignores the possibility of direct spending by the federal government in a single state stabilizing state GDP, b) strips out effects of transfers and grants in national recessions, c) counts smoothing of distributed profits by domestic firms as "smoothing by capital markets" and d) counts a normal variation of households' savings to smooth consumption as "smoothing by credit markets".