Infrastructure and IDA19: A Priority or Not?

The grim picture for SDG-related infrastructure finance in low-income countries (LICs) is by now familiar. But let’s review some salient evidence:

• The IMF estimates that the average LIC needs to spend an additional 15 percentage points of GDP annually on SDG-related investments—7 percent for SDG-related infrastructure and 8 percent for SDG-related health and education spending. That leaves yawning gaps, even if LICs are able to meet the ambitious goal of boosting domestic resource mobilization by 5 percentage points of GDP.
• Private finance for infrastructure for LICs is falling, not rising. According to the World Bank, in 2018, the IDA (poorest) countries received their lowest level of private infrastructure finance in 10 years, 47 percent lower than the average for the past five years.
• Public debt is rising rapidly in LICs. The IMF finds that 40 percent are in, or at risk of, debt distress. The World Bank reports that average public debt/GDP in sub-Saharan Africa rose rapidly from 37 to 56 percent from 2012 to 2016.

In short, huge needs are pushing up against rapidly tightening public borrowing constraints and declining private finance (probably for reasons of both demand and supply).  

It is no surprise that LIC governments have turned to China—which, according to the AfDB, is the source of an annual average of 15 percent of infrastructure finance in Africa. But, as Chinese lending to Africa has grown tenfold from 2012–2017, the public debt build-up has been strikingly rapid.  An estimated 20 percent of African government external debt is owed to China. Moody’s estimated in 2018 that interest payments to Chinese creditors account for more than 20 percent of revenue in Angola, Ghana, Zambia, and Nigeria.

Growing signs of borrowers’ remorse, especially regarding debts for projects of questionable economic value, along with signs that China itself is beginning to worry about debt sustainability, point to a slowing of the debt build-up. Yet in 2018 President Xi pledged another $60 billion in finance for Africa following on a similar pledge three years earlier.

Clearly, a more sustainable way to address this gap would be a large increase in MDB concessional finance for SDG-related infrastructure in poor countries. One would think, in this context, it should be easy to find rapidly rising concessional lending for infrastructure in IDA annual commitments and in the priorities being set in current IDA19 replenishment negotiations. (The International Development Association is the World Bank’s window for concessional loans and grants to the world’s poorest countries. IDA donors meet every three years to replenish IDA resources. IDA19 is the nineteenth such replenishment negotiation.)  Oddly, we see little evidence of either.

Image

IDA infrastructure commitments (to governments) show no marked trends, including in transport and water--sectors where it is hard to attract private finance in LICs. Consumer ability and willingness to pay often fall short of levels necessary for profitability, and technological change has not lowered the cost of service delivery as much as it has for renewable energy. But these sectors are also critical for connecting consumers, producers, and workers to markets, for human development, and for mitigating and adapting to climate change. So we would have expected that IDA sovereign lending would have been increasing in these areas, even before the expanded lending capacity under IDA18 fully kicks in.

What about IDA19 and future trends? Do we see a strong emphasis on expanding infrastructure finance to the public or private sector there?  

Not clear.

Catalyzing private infrastructure investment through IDA

Let’s take IDA19 consideration of finance to catalyze private infrastructure investment. World Bank management documents released prior to the June IDA replenishment negotiation have some encouraging language: “IDA19 will focus on quality infrastructure and enabling reforms that create conditions for private investments to catalyze economic transformation.” This suggests a welcome emphasis on making better use of one of the most powerful comparative advantages that the World Bank and other MDBs have—their ability to bring together support for policy and institutional reform with project finance, technical assistance, and project development support.

The emphasis on breaking down the internal Bank silos that have undercut this advantage is a promising development. In fact, the challenges to private infrastructure investment that the Bank itself highlights could all be addressed by better intra-bank collaboration: more upstream work to develop bankable private investments, coordination across Bank entities on the pace of sector reforms, and more capacity to bring together complementary World Bank Group instruments in a way that is customized to support individual transactions.

In this regard, turning to the IDA19 proposals for the $2.5 billion IDA Private Sector Window (PSW), one would expect to see a corresponding increase in risk-sharing infrastructure project finance for the private sector. Unfortunately, this is not apparent for the part of the IDA PSW specifically designed to mobilize private finance for infrastructure: the Risk Mitigation Facility (RMF). It has struggled since its inception, with no Board approvals yet. Bank management’s proposed solution is to make it smaller going forward, cutting its allocation of PSW resources from $1 billion to $500–$600 million, a 40–50 percent decrease, and reallocating the resources to other PSW facilities. One hopes that the other facilities (the blended finance, local currency, and Multilateral Investment Guarantee Agency facilities) can be directed more toward infrastructure. So far, only four of the fourteen PSW Board-approved projects have been for infrastructure—two for telecoms, where most investment is already privately financed, and two for renewable energy.

IDA infrastructure lending to the public sector

What about IDA infrastructure lending to governments? Do we see any proposed IDA19 quantitative commitments to expand the volume of total infrastructure commitments, or, better still, to meet targets for infrastructure access? Management documents so far do not include such proposals among IDA policy commitments, except for the case of broadband penetration (“reaching 30 percent broadband penetration in at least 20 IDA countries in the African continent by 2023.”) Rather, there is only a general, nonquantitative proposed commitment that hardly appears to be much of a course change: “IDA commits to support in 10 IDA countries the modernization of infrastructure systems (e.g., power, transportation).”  

To be clear, the issue may be as much on the IDA country demand side as on the supply side. Most IDA finance is intentionally unearmarked so that recipient governments can make allocation decisions that meet their development objectives. These governments are certainly confronted with pressing needs for concessional finance for social as well as infrastructure spending. As in any government, short-term consumption needs often trump investments with longer payoffs (which includes investments in health and education delivery systems as well as in infrastructure).

But IDA donors set overall thematic priorities and the growth in the total concessional finance envelope. These donor decisions are key for determining whether IDA provides the kind of infrastructure product offerings and support needed to make a surge in infrastructure investment possible, as well as enough of an increase in concessional finance to help make it possible to pursue both critical SDG-related investment spending and consumption spending.

Some may perhaps argue that regional development banks should take the lead in infrastructure while IDA resources are used for other purposes. But for Africa, which has the most LICs, IDA offers a far larger source of concessional finance than the AfDB, with a lending capacity in Africa roughly six times greater than that of the AfDB concessional window.

Heavy reliance on commercial private lending or non-concessional Chinese lending to supplement scarce LIC government fiscal space for infrastructure is both risky and unrealistic. LIC domestic resource mobilization is improving but that takes time. Bilateral development aid is stagnating in total volume and falling in relation to GDP for the median LIC. I don’t see other alternatives at scale on the development landscape. IDA donors should stifle feelings of déjà vu, balance new priorities with still-urgent old ones, and set ambitious financial or outcome targets for IDA infrastructure investment.