Under such a reform, the government’s financial position is clearly unchanged - the loss
in tax revenues is exactly matched in present value by a reduction in future benefit
commitments. However, if a truncated estimate of government’s fiscal imbalance were
used, it would show a worsening in the government’s financial position: Revenue losses
during the short term would be counted when evaluating the government’s position, but
reductions in benefit payments accruing beyond the truncated horizon would be excluded.
Hence, a truncated calculation of fiscal imbalance would bias policymakers toward
rejecting a personal accounts reform of public pensions even though it would leave the
government’s true financial position unchanged.
This example could be carried one step further. If the public pension system is initially
financially unsustainable, it could be improved through the reform described earlier: The
government could offer a personal accounts reform wherein future benefits are reduced
by more than dollar-for-dollar in present value. Some individuals may agree to the
exchange of smaller future benefits for personal accounts that they would own and
control. Implementing such an exchange would imply a larger reduction in future
government outlay commitments compared to the immediate reduction in wage tax
receipts. However, again, the reduced revenues in the short term would be included under
a truncated projection horizon, but the larger decline in future obligations would be
excluded. Thus, focusing exclusively on a truncated FI measure would bias policymakers
to reject a reform that could potentially improve the government’s financial condition.
This example also provides the final rationale for adopting an infinite-horizon fiscal
imbalance measure in preference to truncated-horizon measures. The FI measure
facilitates an apples-to-apples comparison of different policy options. If two budget
reform options are financially equivalent in present value terms but one (option A)
involves higher costs in the short term compared to the other (option B), policymaking
would be biased in favor of option B if the evaluation were based on a truncated
projection horizon. Alternatively, reform option A might be financially sounder than
option B, but the latter might involve larger short-term financial gains and larger long-
term costs. If the long-term costs remain hidden under a truncated projection horizon,
policymakers may be biased in favor of option B. These considerations are likely to be
quite important in the EU and EMU context because of the many differences in member
countries’ endowments, fiscal policies, and demographic profiles. Hence, long-term fiscal
surveillance should be based on comprehensive and policy-neutral fiscal measures, and
evaluations of fiscal options should use metrics that allow apples-to-apples comparisons
among available choices.
Generational Imbalance. In general, the generational imbalance measure can be
calculated only for programs not involving pure public goods and that are fully or
partially financed out of dedicated government receipts. Provided that it can be
implemented, the generational imbalance measure shows the amount of transfers that past
and living generations may expect to receive under current policies in excess of their past
and future tax payments toward funding them. Thus, excluding the contribution of future
generations to the fiscal imbalance yields the generational imbalance measure (see
Gokhale et al., 2003 for details).
Most European countries have pension and health care programs that are partly financed
out of dedicated taxes. Excess outlays over dedicated revenues are financed out of
transfers from the general budget account. Although such programs are usually
considered to be “in balance” by definition, the fiscal imbalance measure can be used to
72