I've had a few initial thoughts:
Do you agree with this assessment?
This is a very good analysis of the situation. I couldn’t agree more regarding the importance of the link between PFM and policy – after all what is a budget if not the main expression of a government’s intent to implement its policies? However, we should be careful with respect to what the role of PEFA is, and what is likely to be feasible during an assessment.
In many areas of PFM assessed by PEFA, it is acknowledged that it is a broad-brush snapshot of the current PFM system. Areas requiring further analysis should be subject to a “drill-down” assessment e.g. procurement. Looking at policy based budgeting in more detail strikes me more as a “drill-up” and concerns me in that we are now potentially assessing government policy (in as much as it exists) – a much more sensitive topic than PFM architecture.
Of course, I would not deny that it is often a good idea to assess how government policy is transformed into a credible budget. It has been the point of PFM reform in many countries! However, we are opening a real can of worms if we are to question the linkage between the government’s policy management systems and the budget.
Why? Well I would posit that much PFM reform has not been endogenous (to use Matt Andrew’s phrase). In the words of Sybi Hyda, commenting on his time as a Budget Officer in Albania “many times I have experienced the situation when high officials at the MoF, including Ministers, have used the expression “we are doing it because it is required from our international advisers”.” Equally, much strategy development has not been endogenous, either at national or sectoral level. This, of course is not surprising, given the priority of donors to increase levels of spending in the social sectors, with MDG targets to fulfil in much of the world – hence the proliferation of Poverty Reduction Strategies. It would appear that most National Development Plans have been either donor led or drawn up in a manner to satisfy donor concerns. Everybody needs donor funds, especially if it is budget support!
Moreover, recent decades have, with few exceptions, appeared to offer little policy choice to governments around the world – the neoliberal paradigm has dominated policy, not only of donors, but also competing political parties in most countries. The best we have had in terms of policy choices seems to have been whether to prioritise roads or the social sector (such as the debate between the donors and Museveni in Uganda in the 1990s in that wonderful documentary: The Bank, The President and the Pearl of Africa). Apart from that, how often do we catch a glimpse of whether the government agrees or not with the “official” policy documents? And if they don’t, what is their policy in that case?
In many (if not most) countries the objective of governments has been to accumulate wealth amongst the major power brokers (for the re-election campaign as well as personal gain), keep the vote bearing civil servants happy, and hopefully be re-elected and/or avoid overthrow. It is not necessarily the aim to provide the services expressed in the “policy” documents.
As a result it will often be the case that the link between policy and budget is weak. There may well be no formal policy management systems to speak of. If they do exist they may not function well. This is of little consequence to many governments. What is of consequence is if the donors expose these truths – hence my concerns about access to assess many of the issues highlighted.
What is your experience in such matters?
Largely that policy is irrelevant in most LICs and MDCs. Governing parties do not distinguish themselves by their policies – hence the willingness to allow donors to ventriloquise their own concerns into strategy documents.
Regarding the PEFA indicators:
PI-11
(i) Existence of and adherence to a fixed budget calendar; This usually scores well, usually reflecting the legal status of the budget code
(ii) Clarity/comprehensiveness of and political involvement in the guidance on the preparation of budget submissions (budget circular or equivalent); This is more erratic in that some countries comply with a donor-suggested timetable, whilst others regard political involvement as superfluous (indeed the budget may well be superfluous!)
(iii) Timely budget approval by the legislature or similarly mandated body (within the last three years); Usually fine, although sometimes disrupted by strife or elections
PI-12
(i) Preparation of multi -year fiscal forecasts and functional allocations; If an MTBP/MTEF has been introduced then this usually scores well
(ii) Scope and frequency of debt sustainability analysis; if the IMF has conducted analyses it scores well. Also if the IMF has trained the Debt Dept. to carry out such analyses.
(iii) Existence of sector strategies with multi-year costing of recurrent & investment expenditure; this usually scores badly. There are often islands of good practice, usually in areas with significant donor funding e.g. education, health, transport.
(iv) Linkages between investment budgets and forward expenditure estimates; invariably poor
My other related experience is that policy can be ignored if political expedience requires. This can occur with little regard to budgetary implications.
Maybe the best example we have seen in that sphere was the U-turn regarding the investment budget in Albania seen in 1999(?).
After establishing reasonable processes (including political buy-in) regarding budget preparation, including the troublesome development budget, all was sacrificed when the political interests of both the Albanian government and the US conspired to build a superhighway through the mountains of Albania to link it with newly independent Kosovo (site of the largest US base in Europe without “friendly” access to the sea). Protestations (by cognoscenti such as consultants) that in design it was over engineered (and therefore needlessly expensive for the Albanian taxpayer) were ignored. Any political opposition to the road was casually swatted away with cries of “Anti-Albanian”. What’s more, although the road was not fully ready for the election campaign, the Prime Minister ensured it was ready enough for him to cut the ribbon and go on to win the election. Bystanders could only admire his political vision (whilst guardedly moaning about the position of his daughter as legal adviser to the main road contractor!).
Having said all of that, of course it is the prerogative of politicians to adapt to circumstances and rewrite policy “on the hoof”. What happened in Albania was proved to be right by the election result! Such policy swings in other countries are often motivated by more naked aggrandisement.
How should this problem be addressed in practice?
I would adopt a gentle approach to assessing the linkage between policy management systems and budget. If any changes were to be made to the indicators,I would either add a dimension or collapse dimensions (i) and (iii) of PI-11 into one dimension and introduce a new dimension on the efficacy of policy management systems where an A score indicated they existed and passed on all (or most?) the following documents to MoF at appropriate times in the budget cycle:
o statements of government priorities;
o medium-term strategic or policy documents;
o annual legislative programs;
o ministry/agency strategic plans;
o existing legislative mandates with expenditure implications;
o individual policy and regulatory proposals; and
o monitoring reports for strategic documents/plans.
A D score would indicate no policy management system.
B & C would be points in between.
Such an approach would not, one hopes, require too much interaction with cabinet or similar, but would indicate if there was such a system in place. I feel also that it has the makings of being objectively measurable.
What type of guidance would be beneficial?
I would leave this section to Mr. Gord Evans. My only comment is that we would have to define what constituted a policy management system. Is a Presidential Decree to be included? A cabinet decision relayed to MoF? Or does it have to be a formal policy institution within the cabinet office or similar?
