Development finance rhetoric is built on the need to fill yawning financing gaps. Trillions of dollars of investment are needed for the Sustainable Development Goals. These numbers motivate the current policy fixation on using blended finance to mobilise private capital and turn billions into trillions.
But these are not financing gaps, in the sense of an unmet demand for finance. They are wish lists — estimates of the amount of investment it would take to produce everything that the SDGs need, compared to the level of investment that is actually happening. If asset managers turned up with trillions to invest, they would not necessarily find anywhere to put it. Africa might need huge investment in renewable energy, for instance, but there is not a great deal happening on the ground.
It is not even obvious that an enormous increase in the pace of investment would be a good idea. In a forthcoming paper Peter Blair Henry and Camille Gardner (GH) propose a superior approach to wish lists for sizing investment requirements, based on trying to estimate the sum of investment where the benefits look like exceeding the costs. The authors compare the estimated social return on public infrastructure investment against two hurdle rates – one domestic and one international. Looking at investment in roads and electricity, the authors find only 7 of 53 developing countries in which the social returns to public investment clear both hurdles.
Why wouldn’t investments needed to achieve the SDGs have high social returns? Because the social return on investing in schools depends on whether you have enough well-trained teachers, and whether the economy provides jobs that make good use of well-educated people. The social return on transportation infrastructure depends on what it ends up transporting. And so on. Just because you calculate that giving children a good education would require 1000 new schools does not tell you what will happen if you build 1000 schools. The social returns that GH estimate amounts to asking how well things turned out, on average, in the past.
Their paper begins with a sobering reminder that historical episodes of high investment in public capital are widely thought to be partially responsible for previous debt crises. There is not much sign that high levels of public investment tend to result in long-run productivity improvements, and short run booms are sometimes followed by slumps.
As an exercise in speculative fiction, it’s interesting to ask what would happen if these wish lists of investment were to actually materialise. Would national incomes be catapulted to the levels needed to cover the costs, or would everyone involved be bankrupted? One would need to have faith in “big push” development theories, in which complementarities across different investments result in a return that is much greater than the sum of the parts, to believe in a happy ending.
GH look at roads and electricity, and there is evidence these are complements, meaning social returns are much higher when the two are well coordinated. Perhaps GH’s results are too pessimistic because future investments will be better chosen than in the past. Even without such optimism, we should not be too dispirited by GHs’ conclusions because we should set more store in their approach than its execution. Quite sensibly, GH make the best use of the available data but that turns out to be about twenty years old. Moreover, social returns are estimated from the historical effect of investment on GDP but any credible estimate of the social return today would need to factor in climate change, which theirs do not.
Climate change introduces the possibility of investments with very large social returns. For example, IMF research recently estimated that replacing coal fired power stations would produce about $80tn of benefits from roughly $30tn of upfront financing. Investments like that should pass the GH test.
But even estimates of positive social returns investments are not estimates of the likely demand for finance. Something needs to turn hypotheticals into reality. That is especially so for private investments, where investors are motivated by private returns not social. Governments can take decisions based on social returns because they can recoup some return via taxation and may also be willing to pay for social benefits that cannot be monetized. Laws and regulations can direct private investment in more socially beneficial directions, but otherwise taxes and subsidies are the main tools.
In the context of international development cooperation, that role is played by blended finance. The idea here is that relatively small amounts of concessional public finance can unlock much larger amounts of private investment with large social returns, as required by the SDGs. But what can a small amount of concessional finance really achieve? It can somewhat increase the risk-adjusted returns available to private investors and somewhat reduce the cost of capital for the project. That will only unlock trillions of investment if there are trillions-worth of potential projects already quite close to commercial viability (a point I first made way back in 2015).
We hear so much hype about blended finance that it’s easy to overlook the fact that there hasn’t been much of it – or at least, the subsidies involved have been small. The World Bank’s IDA Private Sector Window, for example, might have a face value of $2.5bn but its investments tend to be priced for capital preservation, so the subsidy element is not huge. Governments do not like allocating scarce grant resources to subsidising private investment, outside their own borders. The recent $8.5bn Just Energy Transition Partnership with South Africa contained $330m in grants (although that could simply reflect the fact that renewable energy requires little subsidy).
Blended finance is increasingly seen as a disappointment, with relatively small amounts of private finance having been mobilised. We should be unsurprised when small changes in prices produce small change in quantities. We may need more generous subsidies to fully decarbonise the global economy by 2050, to push beyond renewable power into “hard to decarbonise” industries. So far, we have only seen rich countries announce green subsidies for themselves and although the recent World Bank roadmap mentions “increasing the scope and amount of concessional blended finance” which may or may not amount to a green subsidies package for developing countries. It remains to be seen how cost competitive new green industrial technologies will be, and hence how large “viability gaps” will be, but reaching net zero by 2050 could well require subsidies of a magnitude that nobody today is in a position to supply.
