amplitude either of economic difficulties or of surges in prosperity of individual states. This is
both the product of, and the source of the sense of national solidarity which all relevant
economic and monetary unions share.”
Although they were not designed explicitly for that purpose, transfer mechanisms of
this kind can be regarded as an insurance mechanism against asymmetric cyclical
fluctuations. Regions in a more favorable cyclical position than the federation on average pay
transfers to regions in a less favorable position. This dampens the relative boom in the former
and the relative recession in the latter. If each region had its own currency and exchange rate
were flexible, exchange rate adjustments would provide a similar stabilizing function, as
regions in a relative boom would experience an appreciation of their real exchange rates and a
worsening of their current accounts, while regions in a relative recession would experience
the opposite.12 This consideration is the basis for Kenen’s (1969) suggestion that fiscal
transfers among the member states of a monetary union could replace the adjustment to
asymmetric cyclical shocks otherwise provided by the exchange rate.
Empirical research in the 1990s has sought to estimate how important the transfer
mechanisms in existing federations and large unitary states are in this regard. This discussion
was spurred by Sachs and Sala-i-Martin’s (1991) estimates that the US federal budget
smoothes around 33 - 40 percent of asymmetric shocks to regions in the US. Subsequent
research has shown, however, that, for a number of data and conceptual reasons, these authors
grossly overestimate the smoothing function of the federal budget. Estimates of this kind are
sensitive to the use of different national accounting concepts (Mélitz and Zumer, 1999) and
must distinguish between permanent redistribution among regions and the response to
transitory shocks (von Hagen, 1992). The consensus estimate today is that the federal budget
smoothes about 10-15 percent of asymmetric shocks in the US.13 Estimates for Canada yield
similar results.
One might argue that such a mechanism is not required in the European Monetary Union (or
elsewhere), as markets can fulfill the same function. Free trade and mobility of capital and
labor generate opportunities for the citizens of the member states to protect themselves
against asymmetric shocks. Indeed, empirical studies for the US suggest that financial
markets smooth 30 to 50 percent of state-specific income shocks (Asdrubali et al, 1996;
Athanasoulis and van Wincoop, 1998; Mélitz and Zumer, 1999). Mélitz and Zumer report
similar results for Canada, suggesting that markets are more important in providing insurance
than fiscal mechanisms. But, even if monetary integration promotes financial market
integration and, thereby, the scope for protection against asymmetric shocks, one may argue
that markets provide less insurance than citizens demand. Thus, the question remains whether
EMU needs a fiscal insurance mechanism against asymmetric shocks.
To explore the principles of fiscal insurance, consider a monetary union consisting of i=1,.. .,n
states or regions. Each region is endowed in each period with a stochastic per-capita income
yit with expected value y in all regions. Let yt be the average income in the monetary union.
There is a fiscal transfer mechanism in the monetary union paying transfers τit to consumers
in region i and period t. Households in each region are risk averse and have linear-quadratic
utility functions in consumption, cit,
12 See Stockman (1998) for a broad review of the theoretical and empirical research on the stabilizing
function of the real exchange rate.
13 See von Hagen (2006) for a review of the empirical literature.