spend money without regard to the full cost of taxation. The result would be a tendency to
increase the volume of transfers over time. This tendency could be mitigated by requiring
unanimity for all decision over the payment of transfers, but this would make the system too
rigid to respond quickly to economic shocks as they arise.
A second point is that such a mechanism must avoid the impression of bureaucratic discretion
and that it serves other distributional goals than insurance. This requires transparency and,
therefore, a relatively simple transfer formula. Furthermore, the mechanism must clearly
address cyclical fluctuations. Hammond and von Hagen show that these two requirements
create a trade-off. Identifying cyclical shocks properly calls for the use sophisticated
econometric models which result in fairly complicated formulas to calculate the transfers. If,
however, transfers are based simply on differences in real growth rates across member states,
they generate permanent flows from fast to slow-growing countries and may even destabilize
cyclical movements.
A third point is that, if it aims at offsetting the loss of the exchange rate channel of macro
economic adjustment, a fiscal insurance mechanism would have to target the national
economies of the member states. The different sizes of the national economies then create an
obvious problem if the mechanism is required to be balanced every period, i.e., stabilizing a
negative, asymmetric shock of one percent of GDP in Luxembourg would require a transfer
of small absolute size from the remaining countries, while stabilizing a shock of the same
relative size in Germany would require a payment of very large absolute size from the
remaining countries. This problem could be overcome by allowing the mechanism to run
surpluses and deficits at the aggregate level. In that case, however, one would have to pay
close attention to the risk that the national governments abuse it as a new source of permanent
borrowing circumventing the strictures of the Stability and Growth Pact.
4.3. Assessment of Individual Proposals
An important message from the discussion above is that the design of a fiscal transfer
mechanism for the euro area depends critically on the question whether or not this mechanism
would be required to be balanced financially every period. If not, the main issue that remains
is to identify the shocks properly; once this is done, full insurance against asymmetric demand
shocks and some partial insurance against asymmetric productivity shocks is desirable. A
second, important message is that such an insurance mechanism should not be exposed to
moral hazard. Proposals for fiscal insurance in the euro area should be evaluated on the basis
of these two tenets. We consider two insurance mechanisms frequently found in existing
federal systems. The first is the sharing of tax revenues among the government of euro-area
member states. The second is a euro-area wide unemployment insurance. The former would
pay transfers among the governments, while the latter would pay transfers directly to the
households.
3.1.1. 4.3.1. Tax Revenue Sharing
Insurance through tax revenue sharing can be achieved by having all member governments
pay a fixed proportion of their tax revenues annually into a common euro-area tax fund,
which simultaneously pays out transfers to these governments on a fixed (per capita) basis.
Payments into the fund would then vary with the evolution of the tax base over time, while
payments out of the fund would not follow any cyclical movements. Thus, countries with
temporarily high tax bases would pay more, countries with temporarily low tax bases would
pay less than they receive. As governments adjust their spending accordingly, stabilization is
provided.
The appropriate tax base for such a mechanism would be VAT rather than income or payroll
taxes. The reason is that, first, VAT is closer to demand shocks than income or payroll taxes,
and, second, it reacts faster to cyclical movements in the economy than income taxes or
14