Output versus outcome budgeting

[NinHaj]

I believe that many practitioners come across a desire to structure budgets around outcomes in many countries. The arguments around the advantages and disadvantages of outcome vs output-based budgeting are a common topic in many performance budgeting reform contexts.

From this perspective I believe that New Zealand model is one of the best ones, as it provides a good balance between outputs and outcomes. Though the NZ model often gets criticized for losing the sight of outcomes, it is important things to understand that the output focus in NZ framework is not at the expense of outcomes, but these two perspectives are intended to be mutually reinforcing. If there have been some gaps in considering the outcomes, this has not been because of some inherent fault of the model but because of weakening of focus over time and more relaxed government practices with Labour Party coming to the Government. NZ is trying to fix this by implementing performance audit and I believe no performance system will work well without such audit.

Any system, I believe, has its advantages and disadvantages, and I concur with those who think that the output focus and performance contracts between ministers and departments have more advantages than disadvantages. One of the key advantages, I believe, is that it is the pragmatic approach that works in practice, and it is difficult to fully appreciate that without having actually worked on those issues. In Armenia, for example, the Ministry of Health was insisting on structuring their appropriations around outcomes/objectives. They were so persistent, that we had to give in and let them try. When the Ministry had a chance to think about it in a practical way they decided to abandon the idea, as they came to the conclusion that this was impracticable.

Many countries that claim to operate outcome-based budget systems, in fact, use outputs as the basis of their appropriations (such as, e.g. Australia). If one looks at the appropriations closely it will be easy to see that they are formulated as outputs or functions although they may be labeled as outcomes. I have also come across a lot of cases, when people call outcomes what are actually inputs, e.g. by 20XX the funding for health will reach x% of total budget. There are also many cases when output quality indicators (such as improvement in the quality of services delivered) are confused with outcome indicators.

One of the strengths of the NZ model is its conceptual coherence, as well as being very clear and simple. It tends to call things the right names and does not confuse different concepts.

Some ministers' and politicians’ dissatisfaction with the output focus is partially due to distaste for focus on accountability - as it is a natural human instinct to try to avoid accountability if that is possible. It is always possible to claim that the outcomes were not achieved because some external factor out of one's control has interfered. It is much more difficult to blame on external factors in case of outputs - as their delivery is much more under the control of the relevant manager/entity.

This being said, it is also important to recognize, that in the end, a country needs to be managed - and managed well. So far, the best way to achieve this appears to be managing outputs while keeping track of outcomes, understanding the linkages and using this information in the planning process. This is what the NZ framework allows you to do.

[Napodano]

NinHaj,

I fully agree with your point of view. Output budgeting is in my opinion the best way of going in developing and transition countries, of course without completely loosing sight of outcomes. This was the experience of Albania. 

What I would be ready to consider is a furhter simplification, that is an activity-based budgeting for countries just starting with the budget reform, especially when civil servants are not sufficiently qualified. South Korea had it this way initially and has then added the ouptut and outcome layers.

[petagny]

It would be interesting to have NinHaj's thoughts on the Albanian model compared to the New Zealand model. My sense is that the 'output classes' of New Zealand (and other similar models) give rather more flexibility than Albania's very specific and singular outputs.

The interesting thing that many people overlook is that the NZ model was introduced at a time of fiscal crisis in order to get a sense of what public sector resources were actually delivering (or not) in terms of services and make cuts accordingly. Elsewhere, there has been a tendency to weaken the performance orientation in the face of a fiscal squeeze. The UK's recent defence review might have been rather more coherent with more of a focus on outputs measures (including quality indicators) instead of inputs (e.g., aircraft carriers vs. jets): now we have a set of inputs that can't actually deliver any outputs!

[STONE]

Outputs and outcomes...  I agree with NinHaj outcomes are too vague and not manageable.  Defining the thing that is an outcome is very difficult, PFM practitioners have managed better with outputs, although in some sectors that too can be hard to do and manage.  Because we have this difficulty so do our clients.  I was interested to see the Armenia experience where the government tried and then saw that it did not work so did the sensible thing, but now it knows exactly what the difficulties are.  Perhaps it is only through PEMPAL type processes that government staffs will find experience that they can accept readily.