(Originally published in August 2009 on the IMF's blog)
Countries are often advised to set firm ministry (or sectoral) budget ceilings right at the start of the budget preparation process, as an entirely “top down” process prior to any budget bids or other bottom-up input from spending ministries. This, they are told, should happen immediately following a government decision on the level of aggregate expenditure ceiling.
This raises an important question: is such an early decision on ministry ceilings compatible with good expenditure prioritization?
Good expenditure prioritization is a crucial part of good budgeting. This has never been more true than at the present when in most countries the scope to fund new initiatives from additional revenues is limited or non-existent and when economic recovery will demand fiscal consolidation, with tough choices about what spending is to be cut. Good expenditure prioritization requires flexibility in the allocation of the budget between ministries. It means that ministries running ineffective or inefficient programs have money taken away from them. It also means – although the scope for this is clearly very limited at present – that ministries with new policy initiatives with high political priority should be able, as far as possible, to access additional funding. So we would want to be sure that setting ceilings early does not obstruct such reallocation of resources.
The advice from some in the PFM profession to countries to set early and firm ministry ceilings is based on the proposition that a completely “open ended” budget process, in which ministries are able to bid for whatever they like with no indication of resource constraints, gives ministries no incentives to identify savings, and inevitably results in bids so far in excess of available resources that they are impossibly difficult for the ministry of finance to handle.
The argument is a valid one. There is, however, a real danger that, unless the ceilings are formulated very carefully, setting early and firm ministry ceilings will aggravate allocative rigidity, greatly limiting the possibilities for improved expenditure prioritization. To see this, let’s imagine a system in which the firm ceilings set at the start of the budget preparation process cover all ministry expenditure, whether on existing programs or on new policy initiatives. Ministry ceilings would be set without any prior consideration of “bottom”-up new policy proposals originating from the spending ministries which, if accepted, would require reallocation between ministry ceilings.
In such a system, any decision to increase funding to high-priority sectors would have to be taken in an entirely top-down manner. Some suggest that this is easy: the government just decides that sector X is high priority, and accordingly allocates the ministries concerned more money in the ceilings. All that is supposedly required to do this is a “strategic” meeting of the cabinet at the outset of the budget process at which the government’s priorities are determined.
But this is not the way it works. Even if government recognizes a priority area, it can’t rationally make decisions about how much additional money to allocate without the bottom-up input of concrete new program proposals. Good budgetary decision-making on new spending needs to combine top-down and bottom-up contributions.
Few OECD countries ministry in fact set ministry ceilings before considering bottom-up proposals for new policy. Countries like Canada and Australia manage to avoid an open-ended budget process while permitting considerable scope for consideration of new policy proposals. They do this by requiring spending ministries to submit major new policy proposals separately from their core budget proposals, with the latter based (approximately speaking—see below) on the continuation of pre-existing policies and programs. They then subject the new spending proposals to very rigorous and systematic central scrutiny. Chile, another good practice model, did the same thing with its well-known “Bidding Fund”. It is also worth noting that some of the classic budgeting manuals produced by international organizations (e.g. Schiavo-Campo & Tommasi, 1999, and Potter & Diamond, 1999) which advocate early ceiling-setting in order to avoid an “open-ended” process, make it clear that they do not exclude the modification of these ceilings are a consequence of new policy decisions.
Can we then at least then say that it is a good idea to set early in the budget process ceilings which apply to ongoing spending? Well, here again we have to be careful. The danger is that this will result in the government committing itself prematurely to maintaining funding for ongoing programs and in doing so would greatly reduce the scope for serious expenditure review – scrutiny of existing spending to identify potential efficiency gains or low priority programs which can be cut. By definition, once ministries have been given firm expenditure ceilings, expenditure review aimed at identifying possible cuts to those ceilings becomes irrelevant. Note here that, useful as it certainly is to make use of so-called “planning reserves” (a.k.a. “productivity savings”) when setting the initial ceilings, such reserves don’t in themselves go very far in creating fiscal space for high priority new spending.
Some suggest that the answer to this problem is to undertake all expenditure review work before or very early in the budget process, so that its conclusions can be fully incorporated in early ministry ceilings. Certainly it is a good idea to do some expenditure review work outside of the budget cycle. But to insist on doing it all so early would result in it being rushed and not done sufficiently well. Moreover, it would make it impossible to relate the aggregate value of savings identified via expenditure review to the resources required to fund desired new policy initiatives. This is why countries which integrate expenditure review effectively into the annual budget process – again including Australia and Canada – extend the review process throughout much of the budget preparation period, and have indeed strengthened this is recent years with regular, targeted “strategic review” processes.
One way of resolving this problem is the Danish approach, in which initial budget ceilings are based on current policy but expressly exclude programs which have been targeted for expenditure review (called “special studies”).
All of the above is about expenditure prioritization in the annual budget preparation process. How does this relate to MT budgeting? Doesn’t MT budgeting mean that multi-annual “ceilings” are set? Can’t these serve as the basis for the ministry ceilings set at the start of the budget process?
A common approach to MT budgeting is the “rolling forward estimate” methodology, in which the starting point for formulating this year’s budget is be the (updated) estimate made the previous year of the costs of maintaining ongoing programs. The crucial point here is that using these estimates as the starting point of budget preparation should not mean ratifying them in unadjusted form as ministry ceilings. As occurs in Denmark, they can and should be explicitly adjusted to give scope for expenditure review and full consideration of new policy proposals.
A less widespread form of MT budgeting is that in which the government sets multi-year ceilings which are commitments to spending ministries about the future funding they will receive. The UK is the classic example of this approach, which has now also been adopted in France. Under such a model, a major part of the budget preparation is not annual, but takes place every three years. In this context, the arguments advanced above have a straightforward implication: that it is crucial that, prior to setting the multi-annual ceilings, the triennial budget preparation process undertakes serious expenditure review and gives full consideration to both new spending options. This is precisely what is done in the UK “Expenditure Review” process.
In conclusion, then, setting indicative ceilings early in the budget process is a good idea so long as (1) major new policy proposal may be submitted separately and (2) the ceilings exclude, or are subject to decisions about, the future of major programs which have been selected for expenditure review. Establishing separate and tough processes to consider major new spending proposals can avoid the dangers of “open-ended” budgeting without comprising good expenditure prioritization.