Measuring impacts of cash transfers vs infrastructure investments, and 'sins of omission'

In an insightful paper that is a must-read for all methodological purists among academics, George Akerlof draws the distinction between 'hard' (read quantitative or rigorous) and 'soft' (read qualitative) methodologies to explain economic and other phenomena. He claims that economists’ preference for the latter creates ‘sins of omission’. These omissions include arguments and explanations which are not amenable to the ‘hard’ approaches. For example, ‘stories’ and ‘anecdotal accounts’, even with their often-misleading portraits, carry important insights which are missed by the ‘hard’ approaches.

I had blogged earlier about the paper here.

A very good example of this is the debate surrounding cash transfers, and something like say, the Millennium Villages Project, which focused on capital and inputs intensive integrated village development approaches. The 'hard' methodologists find no evidence from investments in physical infrastructure like roads and electricity, and instead argue in favour of the likes of cash transfers ("transformed the economy"!!) as a better use of aid money.

Consider two development strategies followed in backward rural settings. Plan A - investments are made in physical infrastructure - school and hospital facilities, transportation connectivity, and electricity. Plan B - conditional or unconditional cash transfers of equivalent amounts. Which is likely to be more impactful?

Using Akerlof's framework, clearly the 'hard' methodologies will invariably favour the latter over the former. The former will have long-term and path dependency effects - sustainable economic growth is not possible without capital accumulation, of which infrastructure is a sine-qua-non - but limited medium-term effect on quantitative measures like income and on human development. 

What if we qualify impact by making the distinction between poverty alleviation and economic growth? 

Then the former comes with the certainty of positive partial equilibrium effects but limited or no economic trajectory shifting general equilibrium effects. The latter is a sine-qua-non for economic growth, though its likely general equilibrium effects are also dependent on other factors. It takes the village to the starting line in being able to engage meaningfully with the world of economic opportunities outside. But it is unlikely to generate significant immediate poverty alleviation effects. 

In this framework, cash transfers are, especially when compared to investments in basic infrastructure, something like the Cheshire cats - the impacts disappear with time, without even having expanded the production possibility frontiers (PPFs). 

The path dependency part is important. An all-weather road and transport connectivity expands the PPF by opening up the local society and economy with outside and all associated net benefits (I hope we don't dispute that integration of any type of hinterland with the mainland is, on the net, beneficial). Similar is the likely general equilibrium impact of reliable three-phase electricity supply. The combined effects of both in terms of opening up new economic opportunities and markets, and facilitating better access to public services is undeniable. 

There is another important consideration, that of the quality of implementation at scale. A new study on the much derided Millennium Villages Project from Northern Ghana writes,

The project improved some MDG indicators but, with few exceptions, impacts were small and core welfare indicators, such as monetary poverty, undernutrition and child mortality, remained unaffected. We found no spillover effects of the project to neighbouring areas and no displacements of development expenditure by local government and NGOs. We assessed the cost-effectiveness of the intervention and concluded that MVP did not produce the expected cost-saving synergies. We attribute the lack of impact to poor project design, redundancy of the interventions, and excessively high expectations.

In other words, the details of the design and implementation, both intimately tied to state capacity, were the primary reasons for the lack of impact. But we also know that in such countries, state capacity is acutely weak.

So, if state capacity is weak, and it detracts from the state ability to deliver on pretty much anything, then it is only appropriate that we focus on improving state's capacity.

The poverty alleviation-economic growth framework and the weak state capacity (and the inability to incorporate such considerations) are examples of the 'sins of omission' associated with 'hard' approaches. 

That we are even seriously debating these non-issues should be a damning indictment of the global discourse on development. This is also despite reasonably 'hard' evidence that the beneficiaries themselves favour investments over cash transfers. 

In fact, a UKAid report on the latest MVP study unwittingly points to the deficiencies of these evaluations,

Of course, in the long run, the MVP may produce welfare gains. For example, health care service improvements during the MVP period may improve health later on; or other considerable investments in infrastructure (roads, health and school facilities) may have an impact on future outcomes.

There is another contributor to the 'sins of omission', that arising from the marginalisation of experience and latent knowledge and the de novo search for evidence. I have blogged about it here.

As to cash transfers, this blog's views are summarised here. The case for cash transfers as a substitute for in-kind public goods of any kind (infrastructure to vouchers for education and health care) is questionable on multiple grounds, especially in countries where state is present (even if capacity is weak, the objective then being to strengthen state's capacity). Despite experimental evidence on cash transfers not being diverted to wasteful expenditures, scale-effects can be unpredictable. However, a cautious case for some form of cash transfers can be made in certain settings - refugee camps, civil war torn regions, and where state is completely absent.