Do sub-programmes really exist

[Martin Johnson]

Perhaps the first and most basic issue that must be addressed in programme budgeting is the question – what is a programme? In many, if not most country cases, the question is addressed reasonably clearly. In Albania, for example, the following definition applies:

“A programme is a group of activities that can be effectively and jointly managed and whose outputs contribute directly (or indirectly as inputs to other activities) to the achievement of policy objectives”

In turn, the terms ‘programme policy statement’, ‘input’, ‘activity’, ‘output’, ‘goal’ and ‘objective’ have specific definitions in Albania to ensure clarity over what is, and what is not, a programme.

But what is a sub-programme?

So far, the only definition I have come across in various budget manuals that I have seen is some variation on the following: ‘A sub-programme is a sub-division of a programme’. Whilst this is not incorrect (it is not incorrect because it is tautological), it is not helpful in determining if and when a programme should be divided into sub-programmes and the criteria for making that sub-division.

This made me consider whether, in fact, there is actually a distinct concept that is a sub-programme, or whether there are only programmes. It then made me think that the answer must surely lie in whether it is possible to specify a clear definition for ‘sub-programme’ that is not merely tautological – a definition that also explains why a sub-programme exists, what it is for and why it is not a ‘programme’. If a clear, workable and useful definition can be specified, sub-programmes must exist. If not, then they do not exist.

After some thought, I have come to the conclusion that sub-programmes could exist according to the following definition (within the framework of an Albanian programme, or according to a similar definition under an alternative programme framework):

“Where a programme policy statement contains separate and distinct sub-policy areas, with discrete goals and objectives that relate exclusively to each sub-policy area respectively, and where management arrangements for sub-policy areas are distinct from one another, sub-programmes may be specified. A sub-programme, therefore, is a group of activities that can be effectively and jointly managed and whose outputs contribute directly (or indirectly as inputs to other activities) to the achievement of policy objectives in the corresponding sub-policy area.”

Does this fly? Do colleagues have better alternatives to offer?

[Napodano]

While your definition of sub-program is fine, I do not see why a ministry should make its program-based structure too complex.

Yes, a Transport program could have as sub-programs the different transportation modes, i.e. roads, railways, waterways and air services. Yet in my opinion each of the latter could be considered as a program in itself. In this way you reduce a layer in the performance framework that should be developed for each program.

This approach would also help in the (often elusive)  link between the program structure and the organizational structure of a Ministry.

[John Short]

So........stick to programmes

[Martin Johnson]

Mauro is probably correct in suggesting that one should eschew sub-programmes and stick to programmes, and that is certainly the view that was taken in Albania. But the fact is that many programme budgeting systems incorporate sub-programmes as an integral component of the architecture (e.g. Rwanda, Mauritius, Jordan, Palestine). This does not necessarily mean, however, that use of sub-programmes is more appropriate than sole reliance on programmes. Which is the reason why I think that a tautological 'rationale' for programmes (i.e. defining them as a sub-division of programmes) is not good enough and why a test for their existence (i.e. a test for whether one should incorporate sub-programmes in a PB structure plus if and when they should be used) should rely on a clear and appropriate definition.

But think on this too.

When a MoF Budget Department carries out its challenge function, a key parameter for a productive process involves the MoF being able to see sufficient detail in a given LM budget submission (in terms of programmes, objectives, outputs, resource allocations, unit costs, etc.) whilst being able to maintain a strategic overview of the link between budgets and policies. Maintaining a strategic overview of a given ministry would involve, among other things, considering a 'manageable' number of programmes - considering potential resource allocations between programmes is a more manageable exercise when there are, say 6 to 10 programmes rather than say 20 to 30 (or more). In the same way, a programme manager can more readily consider shifting resources between competing outputs when there is a relatively modest number of outputs rather than a 'plethora'.

The proof of the pudding, of course, is in the eating. If there is a bewildering number of programmes in a Ministry (or if there is a programme which itself has a number of natural programme areas) it may indicate that some of them should be collected together in some way - either in a separate Ministry perhaps or through the collecting some programmes together as sub-programmes within an overarching programme. Given administrative inertia and political imperative, one can easily imagine instances where sub-programmes are used to avoid the administrative difficulties and potential political problems of carving out new ministries. In fact something similar to this happened in Albania when faced with the difficulties of dealing with the State Police programme, within the Ministry of Interior, which itself had a number of natural programme areas.

So, I broadly agree with Mauro - avoid sub-programmes if at all possible. But recognise that under exceptional circumstances they may be required. Above all though, move away from tautology and define properly what sub-programmes are. That way they will be specified only when they are required. The danger in systems guided by a tautological specification of sub-programme is that sub-programmes end up as outputs and activities.

So back to my question .... does my definition for sub-programme fly .... ?