Usual argument in favor of a fiscal contraction in response to an adverse fiscal shock goes along the lines of the Expansionary Fiscal Contraction - or structural - argument. There are many who agree/disagree with this proposition, but there's plenty of literature covering it.
A new argument in favour of 'austerity' response to a fiscal shock is presented in the recent paper from the Bank of Italy.
Optimal Fiscal Policy When Agents Fear Government Default, by Francesco Caprioli, Pietro Rizza and Pietro Tommasino (March 2012), link here, argues that under optimal fiscal policy, when a government is facing with investors who fear a sovereign default, and assuming investors learn over time so as to gradually correct from the initially over-pessimistic view of the default probability, "a frontloaded fiscal consolidation after an adverse fiscal shock is optimal". In other words, 'austerity' can work when it facilitates learning process to support investors' discovery of the 'true' lower probability of default.
In a summary, the findings are:
- When agents fear government default, a fiscal consolidation after an adverse fiscal shock becomes optimal. The intuition is that the interest rate on government debt is too high due to distorted expectations about government default; therefore the marginal cost of higher distortionary taxes today is more than compensated by the expected future marginal benefits of lower distortionary taxes tomorrow.
- The incentive to reduce debt is stronger: a) the more pessimistic agents are about government solvency and b) for a given degree of pessimism, the higher the post-crisis debt level.
- The state of agents initial beliefs has an effect on the long-run mean value of the tax rate and debt. In particular, the more pessimistic agents initial beliefs, the lower the long-run mean value of debt. The intuition is that the more pessimistic the agents are, the stronger the incentive to change their expectations.