Mapping out a novel way to assess the effectiveness of development aid

Assessing the effectiveness of international aid across Indonesia, the world's fourth most populous country and an archipelago of around 17,000 islands, is no easy task. But that has not deterred Singapore Management University Assistant Professor of Political Science (Education), Nathan Peng.

In his paper, "Political Equality and Aid Efficacy: Infrastructure Distribution and Development Aid Effectiveness in Indonesia," which is currently a working draft but has already been presented at several academic events, Professor Peng proposes a novel way to measure political accountability through the distribution of public infrastructure.

"I'm using infrastructure distribution as a signal of local politics," he says. "That's the novel approach."

So why focus on a country as large as Indonesia? Professor Peng says it is diverse yet has a common institutional base. "It holds many constants and yet is a plethora of social, political, and economic contexts all in one country."

"The great thing about Indonesia is that the federal government decides the budget and leaves it up to local governments to decide what to do with it. This means that any diverging outcomes we see in infrastructure distribution are because of local politics, not because of different levels of resourcing or differences in national law. So that neatly isolates the local politics that I want to study."

In the paper, he cites previous research that highlights how political elites are more likely to use resources such as roads, schools, and health facilities in the public interest and foster economic growth when they need the support of a greater proportion of their population to stay in power.

The paper examines 'if the same logic can be applied to international aid; that is, whether development assistance fosters more economic progress when allocated to regions with greater political equality."

"After almost a century now, since the Second World War, we've been trying to foster development in countries that we think lag behind. So the answer I eventually landed on was politics: how we organize ourselves, what we believe in, who makes the decisions, and who follows through with those decisions and why? That's what drives economic growth."

The paper focuses on 'a specific aspect of local politics: constituency breadth," that is, the breadth of constituents you're accountable to, to stay in power.

The World Bank, which has been providing development aid for decades, has not "figured out how to compare different political contexts objectively," Professor Peng says. "What I'm doing here is really to come up with a way to begin doing that and the approach I took was to look at outcomes that reflect local politics, the distribution of roads."

"The prevailing research generally looks at public spending and how well public funds are utilized," Professor Peng says. "And the general consensus is that if you give it to an elite which has a narrow support base that it needs to stay in power, then generally that elite is going to squander the money just to please that narrow support base and leave the rest to languish."

'Aid's overall impact is ambiguous'

Professor Peng believes the World Bank's strategy of giving to the poorest regions in Indonesia makes sense even though this makes the impact of aid somewhat ambiguous and uncertain.

The World Bank's mandate is poverty alleviation and so it makes sense to help poor regions, he says. "Unfortunately, the areas that might best use the aid, where you might see more growth in absolute terms, is not where the aid is going and I think that's a good thing because you want to bring everyone up to the same level rather than helping areas with the most potential and then you're creating greater inequality within the country. That might not be desirable."

"My other finding is that I expected that giving to more politically equal regions would lead to better use of aid money and resources, but that is not what I found. Ironically, when you give aid to politically more unequal regions, you see more growth. And that's the result I wanted to warn against as well. Just because you see a better result doesn't mean you should do it. They really do need to take local politics into account to avoid doing more harm than good."

As someone who had previously worked for Singapore's Ministry of Trade and Industry, as well as the Ministry of Social and Family Development, Professor Peng believes it would be hard to design systems that can work around this 'conundrum," where one might have to give up on KPIs "to do real good" that cannot be measured in numbers, at least not quickly.

The scope of his study was ambitious, given the vast expanse of the country. In terms of methodology, Professor Peng took both a qualitative and quantitative approach, interviewing a wide range of experts and then examining satellite data and imagery. "Given that the data quality in developing countries is generally pretty bad, I thought using satellite data would help work around many of the problems we have as practitioners."

"When I spoke to civil engineers, infrastructural and political experts in Indonesia, I wanted to confirm two things: if roads disappear after a few years from satellite imagery and if local political leaders had sway over infrastructure decisions. So, I wanted to make sure it was politics driving the results and not anything else. That's why I took both a qualitative and quantitative approach."

It might sound odd to see if roads 'disappear," but it is important to understand whose decisions are responsible for existing roads. The World Bank acknowledges that corruption is a problem in Indonesia and if corruption had been involved—and for Professor Peng "corruption has a very strong political element"—roads may not have been built in the first place as the money may have been siphoned off. And even if roads had been built, they may not have been maintained properly and subsequently became overgrown due a lack of funding.

Based on what Professor Peng calls a 'proof in the pudding approach, focusing on the outcome of political machinations," the paper presents a 'starkly different picture from what the rich literature on constituency breadth and winning coalitions would lead us to expect. If aid dollars behaved like government funds, they should then be more efficiently used when allocated to politicians who are accountable to more of their constituents, as opposed to being squandered as political pork for the few. Yet that is not what we observe."

The paper goes on to say, "What is unexpected is the finding that more politically unequal regions are associated with higher growth rates after aid is disbursed."

The findings of the study are, the paper states, 'complex." First, with aid directed to poorer regions rather than regions which could use the aid more efficiently, the paper points out that 'outcomes that at first appear to be economically sub-optimal might in fact represent genuine progress."

Second, while giving aid to regions with politically repressive leaders might 'do more to unlock economic potential' and lay the foundation 'for their eventual removal by empowering the politically marginalized," the relationship between economic growth and political representation is 'far from linear."

Finally, while aid might be able to unlock economic potential, this is 'by no means guaranteed. Aid's overall impact is still ambiguous."

The paper concludes, "Overall, these results suggest a potential conundrum for donors, where giving money to more politically unequal regions might generate more economic development in the short term, but with the potential downside of solidifying the positions of more repressive political elites. What is clear, however, is that measuring the proportion of constituents elites believe to be required for maintaining power will help us make more precise estimates of how development aid fosters economic growth."