(Originally published 22 April 2010)
I have previously criticized the assumption made by some that MT budgeting necessarily requires that ministries be given fixed MT ceilings – that is, ceilings which represent commitments from government on the level of funding each ministry will receive in the next, say, three years. A system of indicative ceilings is, I have argued, preferable in many countries. There is an important point to add to this debate: namely, that there can sometimes be serious tension between fixed medium-term ministry expenditure ceilings and the aggregate fiscal policy or fiscal rules which guide budget policy.
By committing to fixed MT ministry expenditure ceilings, government is, broadly speaking, committing to certain levels of aggregate government expenditure in the MT. And if it then also commits to a specific deficit outcome via a fiscal rule, aggregate revenue becomes a residual. This means that any shocks to public finances which not catered for by the deficit rule itself will need to be fully absorbed by adjusting aggregate revenue. Put simply, stability on the expenditure side means corresponding instability on the revenue side. But tax policy volatility is no more a good thing than is expenditure volatility, and to insist that all of the adjustment to shocks come from the revenue side is clearly bad policy.
Take a concrete example of a fiscal rule which is quite justifiably attracting a lot of attention these days – the Swiss debt brake rule. The core of this rule is a simple structural budget deficit objective. To the rule adds an “error-correction” mechanism which addresses the danger of debt ratcheting as a results of any actual deficits being higher than those projected in budgets. When the cumulative excess of the actual budget deficit over the target deficit exceeds a certain threshold level, the rule requires immediate corrective actions to bring the deficit down. This action may take the form of either expenditure cuts or revenue increases.
Clearly, if one were to impose fixed medium-term ceilings on top of the Swiss debt brake rule, the effect would be that whenever the rule demands immediate corrective actions, this would necessary have to take the form entirely of tax increases. The same is true in respect to a range of other fiscal rules.
Sure, fixed medium-term ministry ceilings are perfectly compatible with good aggregate fiscal policy in countries which take a different approach to fiscal policy and rules. For example, t hey will work fine in a country like Sweden which sets fixed MT aggregate expenditure ceilings which are broadly derived from a medium-term deficit objective (in Sweden’s case, a one percent surplus target), but in which this deficit target is not a fixed rule.
The general point is clear: in choosing between fixed and indicative medium-term ministry expenditure ceilings, it is essential to consider the compatibility of fixed ceilings with the government’s aggregate fiscal policy and any fiscal rules which might be in place.
The above analysis simplifies somewhat to make the point. For example, a system of fixed medium-term ministry expenditure ceilings does not in fact literally fix aggregate public spending in the medium term, because the fixed ministry ceilings would typically not include all public spending. (Expenditure like social security benefits, which is demand and formula-driven would typically be left out.) But taking these types of complications into account doesn’t in any way affect the conclusion.
This should not be understood as an attack on the fixed medium-term ministry ceilings model. In countries which have good medium-term fiscal forecasting capacity and well-developed spending prioritization mechanisms, such an approach can be excellent. But it is not appropriate for everyone.