Public Economics
Austerity programmes

Conventional models of aggregate demand suggest that a reduction in the government deficit should depress consumption and output, but their theoretical foundations are often unclear and their empirical predictions sometimes inconsistent with reality. The Danish and Irish governments enacted particularly sharp expenditure cuts in the 1980s. Denmark's primary government deficit fell by 10% of GDP during 1982-6, while its consumption/GDP ratio rose by several percentage points. In Ireland, the fiscal contraction in 1982 led to a sharp fall in private consumption, but a second attempt in 1987 had a strongly expansionary effect.
In Discussion Paper No. 599, Research Fellow Giuseppe Bertola and Allan Drazen develop a model of government spending in which significant cuts take place only when the ratio of government spending to output hits a `trigger point', so sharp policy changes occur only infrequently. When current fiscal innovations are expected to persist over consumers' planning horizons, government spending tends to crowd out private consumption almost one-for-one; the public reduces consumption to meet current or expected future tax obligations. Since a current increase in government spending increases the likelihood that it will be stabilized (to meet the government's solvency constraint), policy innovations that would be contractionary in a static model may be expansionary in this dynamic model if they induce strong enough expectations of future policy changes. If private agents believe that levels of public spending are too high to be sustainable and will therefore soon be cut, current private spending will increase accordingly.
Bertola and Drazen note that ruling out discrete stabilizations implies that movements in the ratio of government spending to output will be matched by equal movements in the consumption/output ratio. When discrete cuts in spending are included, however, increases in current government expenditure only partially crowd out consumption. Such increases raise the probability of a future discrete expenditure cut, so agents expect a smaller future tax increase than in the case of no intervention. The authors examine the stabilization experiences of Denmark and Ireland, which provide support for their model: a pattern typical of the relationship between private and government consumption in many countries emerges quite strongly. When government spending is a small fraction of GDP, the ratios of private and public consumption to GDP exhibit an inverse relationship, which flattens out as the ratio of government spending to output increases.

Trigger Points and Budget Cuts: Explaining the Effects of Fiscal Austerity
Giuseppe Bertola and Allan Drazen

Discussion Paper No. 599, December 1991 (IM)