Norad misunderstands the investigation instruction

NORAD Director Ingvar Olsen writing in Panorama to call into question Norad's methods and options assessments, and to suggest that they should be adopted state investigation instructions, shows “a lack of understanding of the objectives of the aid and the various channels and means of the aid apparatus”.

Furthermore, he believes that the proposal of Stronger prioritization of the measures that work best, based on an analysis emphasizing aid measures' effectiveness and opportunity costs, is “simplistic and dangerous”.

It is healthy that we have an engaged debate about how we can have a better and more knowledge-based decision-making process in Norwegian aid management. Assistance is about dealing with complex and large-scale societal challenges, and then it is important to learn from the experiences of others.

Unfortunately, Norad's latest post in the debate shows little interest in it. Calling an aid analysis “dangerous” is not just dumbing down and condescending. It's also on the side of the question. The interesting thing is whether the analysis is correct.

‍

The investigation instruction has considerable room for action

The assessment instruction requires all state administrations to prepare sound decision-making grounds before deciding what action to take.

These decision bases shall at least include answers to six fixed questions, including: What's the problem? What do we want to achieve? What measures are relevant? And what measure is recommended?

Olsen seems to be calling for the use of the assessment instruction in aid management — which until now has invoked an exception (at Doubtful basis) -- oversimplifies the difficult assessments Norad has to make.

He believes it is impossible to assess different aid measures against each other, since not all effects can be measured. Furthermore, Olsen points to the challenge that aid has several different objectives, and that trade-offs must be made between these.

It is somewhat navel-gazing to think that the complexity of one's own sector is so much greater than in any other. Having several different, and sometimes divergent, goals is not unique to aid. The same goes for the fact that not all effects can be measured. This applies to most areas of management, and the assessment instruction takes it into account.

The assessment instruction is used today for decision-making in all other areas of government administration, including health care, transport, culture and defence. In alternative assessments in such areas of society, one must also measure different effects, and with different methods.

In the field of transport, for example, you will be able to quantitatively measure the number of lives saved and increased productivity from the time people save on a better path — so also health care can be quantitatively measured as lives saved and productivity increased.

However, one also estimates other effects that cannot be easily quantified, so-called non-priced effects. This is done qualitatively, which is clearly described in the supervisor of the investigation instruction from DFØ. Other important considerations such as context, long-term development, cultural, geographical and socio-economic conditions and fungibility, can also clearly be part of the assessment instructions.

Again, contrary to what Olsen seems to think: Aid is not such a different field of expertise that one cannot use methods developed in other sectors.

DFØ stresses itself that the assessment instruction has a wide scope for action, both in terms of the methods that can be used and the ways in which knowledge is to be obtained. Elisabeth Aarseth, Head of Department at DFØ, writes: “My experience is that perceived limitations do not follow from the intention of the framework and the actual requirements. Often it is rather the company's own expectations, traditional ways of working, lack of expertise and the somewhat cautious approach of bureaucrats that can seem inhibiting.”

‍

Aid should prioritise the best measures first

To think that Norad should prioritize the best measures first is also not an oversimplification, as Olsen writes. On the contrary, there is a requisite simplification, in a directorate that does not seem to have its mandate clearly enough articulated. When one has a scarcity of resources in relation to the problems one wants to solve, it is obvious that one must prioritize.

In Norads first post in the debate, department director Lisa Sivertsen argues that Norad's ridiculed aid projects may They have a positive effect, and they probably work. She writes: “Every single project that receives support through Norwegian aid is thoroughly assessed in advance, along the way and in the end.”

Still, nowhere does she write that the aforementioned projects have been rated as the best possible use of Norad's scarce funds when they were adopted. Better than the alternatives.

In recent decades, aid has been through a huge knowledge revolution, and we currently have a lot of research on what kind of aid works best. When we have to prioritize what we want to fund anyway, it makes sense to first support what we think works best.

Even when research falls short, no aid should be funded unless it can at least probabilisé and it is sensibly argued — usually based on non-priced effects, context, or cultural, geographical and socio-economic conditions — that it is superior to other alternatives. Good aid is not good enough if we could Prioritize something that is even better.

‍