FISCAL CONSOLIDATION AND DECENTRALISATION: A TALE OF TWO TIERS



To better capture relative success, and to discriminate across a range of alternative outcomes
following a consolidation attempt, we make use of the more refined success index in
definition 3. In the remainder of the paper we present our results based on both definitions 2
and 3.

Definition 3:

An index of success/failure, SI, is classified as follows:

SI = 3

if the level of the debt to GDP ratio falls by at least 5 percentage points in the
three years following a consolidation attempt (equivalent to “success” in
definition 2).

SI = 2

if the level of the debt to GDP ratio stabilises (within half a percentage point
tolerance of the initial level rate) or if it decreases by less than 5 percentage
points over the three years following a consolidation attempt.

SI = 1

if the growth rate of the debt to GDP ratio over the three years following a
consolidation attempt is less than that observed in the three years prior to the
consolidation attempt (here we use a tolerance of 10% of the initial growth
rate).

SI = 0

if none of the above apply the attempt is classified as a failed consolidation
attempt.

3.4 Consolidation Attempts of Sub-Central Governments

It seems likely that the majority of consolidation episodes are instigated at the level of central
government. However, our dataset also allows us to identify consolidation attempts made by
the sub-central tier of government. Given the smaller scale of sub-central budgets it is
necessary to adapt definition 1 for use at the sub-central level:

Definition 4:

A period of sub-central fiscal consolidation is deemed to have occurred in a given year if the
weighted discretionary fiscal impulse is greater than or equal to +1.5 percent of GDP or if the
weighted discretionary fiscal impulse exceeds 1.25 percent of GDP for two consecutive years.

The weight applied in each case is the sub-central tier’s percentage share of general
government expenditure.

We allow for two-year consolidations at the sub-central level given the relatively smaller size
of this tier and the likelihood that in practice such consolidations are likely to be more gradual
than at the centre11.

11 We have allowed for only single year sub-central consolidations and full results are available on request.
Degrees of freedom become relatively small and though our results remain robust, obtaining statistical inference
with such limited observations is more difficult.



More intriguing information

1. Are Public Investment Efficient in Creating Capital Stocks in Developing Countries?
2. Testing for One-Factor Models versus Stochastic Volatility Models
3. DIVERSITY OF RURAL PLACES - TEXAS
4. Tariff Escalation and Invasive Species Risk
5. MANAGEMENT PRACTICES ON VIRGINIA DAIRY FARMS
6. The quick and the dead: when reaction beats intention
7. he Effect of Phosphorylation on the Electron Capture Dissociation of Peptide Ions
8. Government spending composition, technical change and wage inequality
9. ISO 9000 -- A MARKETING TOOL FOR U.S. AGRIBUSINESS
10. Bidding for Envy-Freeness: A Procedural Approach to n-Player Fair Division Problems