News & Views

As economists, we are often faced with the difficult choice of identifying the highest investment priority from a disparate, long list of options. Some would say that we could simply fall back on a cost-benefit analysis, using economic metrics to make decisions. But that is not always possible, in situations of data poverty, or where project ideas are not well developed. In any case, cost-benefit analysis is not always appropriate. What about situations where nothing is economically viable in the traditional sense? These are not reasons to delay or avoid making decisions, but a reason to make them differently.

What is Multi-Criteria Analysis?

One option is multi-criteria analysis (MCA). An MCA does away with the need to prove that investment is economically viable but, instead, is a tool that can be used to identify the best way to achieve an objective when it is accepted that the objective is a priority.

MCA has a wide range of applications from small islands to remote villages, or anywhere where data is scarce. It is particularly helpful when working in data-poor environments, and when you are trying to compare apples with oranges, as it enables the quantification of, often difficult to define, variables. We use MCA when trying to compare investments across different sectors, or in situations where the data to allow the quantification of beneficiaries or benefits doesn’t exist.

Building a Robust MCA Framework

A good MCA will have several key attributes, but the most important is always that the criteria used are linked to higher-level documents, such as a national development strategy or local plans, which establish overarching goals, which should be largely apolitical. Using these can help to overcome two of the most difficult issues when identifying priorities – political interference where politicians want to ensure that projects in their constituency or ward get priority, and stakeholders disagreeing with the results of an assessment because ‘their scheme’ isn’t high enough in the priority list. Showing direct links to national or local strategic objectives increases the chance that people will agree with both the criteria and the results of the prioritisation.

The figure below shows the mapping of criteria and National Development Strategy (NDS) goals from one of our recent projects. As can be seen, not all NDS goals need to be included and some goals can be linked to more than 1 criteria, but each criterion needs to be linked to at least 1 goal.

To weight or not to weight?

MCA is a very flexible tool, allowing a wide range of criteria to be incorporated. At the same time, it is possible to reflect different priorities by including additional criteria matching a particular goal. Alternatively, it is possible to add a weighting system to the framework, allowing a greater emphasis on a particular aspect or criterion. However, care needs to be taken when doing this. It is very difficult to achieve agreement across all stakeholders when applying weights, as the principle of weights is often seen as downgrading those aspects which are given lesser weights. Extensive stakeholder consultation and workshopping is usually required to achieve any level of agreement on the weights, which will be on top of getting agreement on the criteria themselves. This can be costly in terms of time and money.

In any case, our experience is that the application of weights rarely changes the results of the assessment significantly. A good project will always be a priority project no matter what weights are used. Our practice is therefore to not apply weights, but instead to undertake sensitivity testing to show that different weighting would make little difference in the end.

How to Use the Results

What an MCA framework can do is reflect a hierarchy of needs. We are all familiar with the concept that humans have basic needs (shelter, water, food) that need to be met before more advanced desires become important. Using an MCA framework can build this in. When comparing different infrastructure sectors, it is possible to reflect that clean water will always be more important than say a basic transport network, which is always more important than leisure facilities.

A good example of this hierarchy was seen in developing the Niue National Transport Strategy, where we were able to explain, using an MCA, why the port was more important than the airport. All fuel comes in through the port, including the fuel for aircraft, meaning that planes would not be able to serve the island, if the port was non-functional.

MCA cannot, however, do everything, and in some circumstances, it can be positively unhelpful. If you want to know more about when not to use MCA, let’s have a chat.

To summarise, Multi-Criteria Analysis is a great tool for infrastructure economists and planners but cannot be used in every situation and will often need to be implemented by experienced practitioners. For another example of how we deploy MCA frameworks, see our COMBO tool.

If you would like to learn more about how ITP has applied MCA in a range of situations, or more about ITP's Economics and Finance practice, please contact james.reeves@itp.rhdhv.com.