This is partially how we pay for African Infrastructure
Synthetic Credit History/Guarantee Derivatives
Last week, I asked a billion-dollar question – How do we pay for Africa’s infrastructure? In this post, I present two examples from Liberia that explore possible options.
A lot has been written about the difficulty of infrastructure as an asset class. An illustrative example of this is the case of Jim Fitterling, the CEO of Dow. In an interview with the FT, he remarked that “There was plenty of private capital available to invest in infrastructure projects [in the United States] but you have to make them investable.” This is coming from the CEO of a company that has identified infrastructure investments ranging from modernized electrical grids to 5G wireless networks as among the areas most likely to drive its growth. If this is Dow’s posture toward the US infrastructure market, imagine what it is like for low-income countries. Compounding all this, low-income countries in sub-Saharan Africa bring unique complications and confounding factors to the prospect of private sector investment in their infrastructure.
As of February 2020, thirty-three of the thirty-nine countries that have qualified for, are eligible or potentially eligible, and may wish to receive HIPC (High Indebted Poor Countries) Initiative assistance are in Sub-Saharan Africa. It is difficult to get excited about lending to a country that qualifies for or is eligible for assistance under an initiative called the Highly Indebted Poor Countries Initiative. Coming out on the other side of HIPC, with debt waived and a new lease on life, these countries start off with limited credit history. What credit history they previously had could not have been great in the first place since it resulted in their needing or qualifying for assistance under HIPC. Building a history of consistent commitment takes time.
Many of these countries are also fragile states. They have either recently emerged from violent conflict or still have low-intensity violent conflicts ongoing. There are high levels of institutional and social fragility. Institutional fragility is a red flag for political risk. Infrastructure financing is usually long-term and expected to survive multiple political cycles. This kind of financing requires predictability which again, is undermined by institutional fragility.
All of these conditions applied to Liberia when we sought to finance a road project of between $100 and $125 million with private money in 2017. As Minister of Works, the project was exciting for the promise it held for the rest of the continent. If, with all of Liberia’s weaknesses, we could structure a project that could attract private sector financing, I felt it could be done in most places on the continent. So, we set out to design a project that could a) deliver a reasonable return with reduced/hedged risks, b) consistently deliver availability payments, and c) ensure a predictable policy environment over the long-term.
Our first attempt did not go very well. We were approached by a private sector firm for a PPP bridge project that would connect two sections of the capital, significantly reduce transport costs and travel times. With limited fiscal space, the government could not provide a sovereign guarantee since it would effectively be a contingent liability and impinge on the [aforementioned], non-existent fiscal space. Our prospective private sector partners assured us that from their preliminary traffic count, the project would pay for itself through tolls. We were excited and proceeded to calculate the cost of land acquisition, resettlement action, and other permits that the government would be required to provide.
However, the Ministry of Works did not have the requisite competence in house to negotiate a PPP contract, so we applied for and received a grant from the Africa Legal Support Facility at the African Development Bank. With the grant we hired Norton Rose Fulbright, who brought along the engineering firm, Roughton. This was probably one of those few times where a low-income country government came to the negotiations more prepared than the private sector actor. Roughton raised questions about the methodology and reliability of the traffic and Norton Rose Fulbright discovered a clause in the draft agreement that placed contingent liability on the government even though we had repeatedly noted that we could not. In the end, the private company got into a shouting match with our lawyers and it went downhill from there. Ultimately, our transaction advisor concluded that there was insufficient, underlying data made it impossible for the government to agree to move forward with the project. Such projects are still possible with sufficient traffic for tolls, or the government’s ability to underwrite the shortfall. The HKB bridge in Cote d’Ivoire is a similar project.
The picture below shows an artist’s rendition (from a different contractor) of the bridge in question. It looks beautiful, but I digress.
A Second Go:
In 2017, with World Bank support, we concluded a spatial analysis of the Liberian road network and found that 49 percent of the population lacked access to an all-weather road. The estimated cost of connecting all our 15 county capitals by a two-lane paved road would be about $1.19 billion dollar. There was no fiscal space to borrow such an amount of money just for transport infrastructure when only about 25 percent of the country was connected to the grid and the availability of power still irregular. Then there was the water and sanitation deficit. We needed to find a way to convince the private sector that investing in Liberian roads was worth it.
We had one thing going for us. We were either the first or the second country in Africa to attempt building infrastructure on an output and performance-based contract. We paid the contractor for service-level performance of the road, over 10 to 12 years. The Output and Performance-based Road Contract (OPRC) was introduced by the World Bank and it aligned the incentives of owner and contractor. Knowing that they would be responsible for the road for a decade after construction and only be paid if the road performed at pre-determined service levels, the contractor has all the incentives to build the best possible road. For a country with a weak and yet developing maintenance culture, it ensured a functional road over the medium-term.
For our private sector finance road, we agreed it would be OPRC (over 15 years) – two years for construction and thirteen years of routine and periodic maintenance. The entire project would be placed under a Special Purpose Vehicle as the “owner” with binding obligations from the government. The project was designed for the contractor to organize financing, the SPV, with resources from an IDA loan, would make the mobilization/advance payment and subsequently make milestone-linked payments over the course of the 13-years after a 2-year construction period. The government could then use a portion of its IDA allocation to obtain a guarantee from the World Bank. The guarantee would go either directly to commercial lenders to mitigate debt service risks or backstop the ongoing availability payments. In 2016, we had passed a law creating a Road Fund that was financed through fuel levies and other fees. Forty percent of the Fund’s resources were dedicated to rehabilitation and improvement works. The government’s milestone-linked availability payments were expected to be drawn primarily from the Road Fund, backed, as noted above, by the World Bank guarantee.
Because Liberia maintains a consolidated account for all public revenues, the Minister of Finance had issued an irrevocable authorization for monies collected for the Road Fund to be moved into the Fund account, which had clear governance rules on how it could be withdrawn.
With the pieces in place, we took our project on the road. We met with contractors in Nairobi and banks in Paris seeking feedback on any blind spots or any weakness in our structure. Both the banks and the contractors seemed pleased. At that point, I had to leave the delegation and return to Monrovia because it was now December 2017, and in January 2018 a new administration was set to be inaugurated. The clock had run out…
About a year later, on December 18, 2018, the World Bank approved the structure. The World Bank board approved credit for the road project and an additional $48 million payment guarantee to leverage funding from the private sector.
What this tells us is that it is the same kind of financial innovation that leads to the introduction new asset classes can be applied to low trust, no credit history, low-income countries. Just as synthetic assets derive their value from underlying financial instruments or assets, it is possible to build derivatives for sovereign guarantees in low-income countries using either World Bank, African Development Bank, or other multilateral finance institutions as the underlying source of value. It is such a shame that development finance institutions have been so conservative in financial engineering for low-income countries.
Many low-income countries have either road funds or general infrastructure funds, but they are usually hobbled by governance issues especially financial management. As Liberia came out of its civil war, building state capacity wholesale was out of the question. A way around this in Liberia and other states with weak institutional capacity is a governance innovation, the “implementation unit.” These were small islands of competence in oceans of institutional fragility. At the Ministry of Finance, we had a Public Financial Management Unit (PFMU) that handled multilateral and bilateral funds. At the Ministry of Public Works, we had an Infrastructure Implementation Unit, that oversaw multilateral and bilateral financed infra projects.
As we attempt to build predictability, credibility, and institutional capacity where it does not exist, a best practice would be creating a similar institutional mechanism – a National Infrastructure Trust Fund (NITF). The NITF would be a coordinating institution to plan, design, and supervise the construction of infrastructure projects. It would be a central clearing house for feasibility studies, balancing economic viability and social impact. Such an institution cannot be created by executive action. It must be established by law and its top officials professionally recruited and tenured. For at least the first decade of its existence, the MDBs which would provide synthetic guarantees must have an oversight role in its governance. The NITF will be responsible for designing projects to attract private sector participation. There is an example of this in Mexico with Fondo Nacional de Infraestructura (FONADIN). The use of “implementation units” was a governance innovation out of the World Bank and other MDBs. The idea can now be extended to build permanent institutions to oversee infrastructure financing.
The scale of the Africa’s infrastructure deficit and underlying weakness of both fiscal and institutional capacity across the region require innovative interventions that create synthetic versions of those crucial requirements for private participation in infrastructure financing. Unfortunately, development finance has chosen to be conservative and eschew innovation and creativity. Speaking of the Liberian project, the World Bank task team leaders noted that it was the first of its kind in IDA’s portfolio to deploys a combination of IDA Credit, IDA Guarantees, and Grant resources to leverage private financing for a road Public Private Partnership (PPP) in an affordable and fiscally sustainable manner. Glad it was done, but disappointed that this was the first time it has been attempted.
Such an interesting article - the case studies, in particular, are fascinating! But I have a couple of qualms: (1) synthetic versions of fiscal/ organisational capacities might be sufficient for get 'spades in the ground' but it's perhaps less clear that they can manage the risks of infrastructure over the life-cycle - and most of the costs of infrastructure relate to capex not opex. How can we ensure that decisions made in year 0 take into account the costs and risks generated by those decisions in years 5, 10 and 15? (2) I worry that a focus on synthetic capacity may impede the work of those who are trying to build fiscal/ organisational capacities locally - a process that only happens, in my view, when local people/organisations face up to the critical problems of their society, and figure out and develop what they need to solve them.
Mr. Moore, I didn't get the part where other African countries can learn from Liberia's experience since the clock ran out. To be honest you're incredible