Risk based approach to a challenge fund

We are now in the final stages of putting together the investment strategy for ABIF. Two of the biggest challenges facing us in defining the investment strategy for the fund were:

1. How to ensure that we incentivise without subsidising; and
2. How to ensure that we do not displace commercial finance providers.

I think that we have come up with a solution to both problems by applying some relatively simple tools from the world of private sector corporate finance.


Incentive or subsidy?
The operational objective of the fund is to incentivise commercially viable investment projects that will achieve our strategic objective of achieving market change that will accelerate inclusive economic growth. The incentive we offer is co-investment in the form of a grant to the private sector partner. Incentivising investment is all well and good, but in the absence of some rational mechanism to target and ration funds, a well meant grant can so easily become a market distorting and value destroying subsidy.

The original design for ABIF suggested that we should limit the ABIF grant by fixing a maximum grant level of 50% of the total investment project when co-investing with individual companies and 70% when co-investing with business associations, and then rely on competition between potential grantees to maximise leverage of donor funds. This strategy was intended to limit the risk of subsidy and has the advantages (from the donor project's point of view) of being the established way that challenge funds operate (so it is the "safe option") and administratively simple (so it is easy to understand).

However, it is a very blunt instrument, which fails to take into account the market circumstances in which the fund is operating and could exclude high risk/high impact projects from applying to the fund.

First, where do the pre-determined numbers 50% or 70% (or any other number that different challenge funds use) come from, why not 25% or 75% or anything in between? Answer, any number pre-determined in this way would be completely arbitrary. Secondly, why should funding two identical projects in different places and different times impose the same limit? Answer, none at all.

In the imperfect market of donor funding (where gaming is rife), there is just as much risk of subsidising a project by giving a 30% grant as there is a 50% grant or a 70% grant if you don't understand the investment risk associated with the investment project. The riskier the location (for ABIF, places like Helmand), the higher the grant has to be to offset that risk and incentivise investment that will impact some of our least accessible target beneficiaries. The riskier the innovation (say a business model completely untried in Afghanistan), the higher the grant has to be to offset the risk and accelerate the introduction of new products or services that will impact the incomes of our target beneficiaries. And of course, the converse is true, the less risky the location and the less risky the innovation, the lower our grant should be. Thus the risk based approach is a vital value for money tool for donors.

Adopting a risk-variable investment strategy not only provides a basis for opening up the competition to more risky investment projects and for comparing projects on a like-for-like basis, it also provides a mechanism to avoid subsidy. However imperfect the challenge fund competition may be, by having a risk benchmark for each project will allow us to identify the level of grant that would cross the line from incentive into subsidy and negotiate the ABIF contribution accordingly.


Avoiding displacement
Afghanistan has a very limited commercial investment finance market. Bank corporate lending is minimal and there are very few investment finance products available on the market. However, there is something, however inadequate, going on. The last thing that a donor funded grant fund should do is to nip the market in the bud by blundering in and distributing grants to private sector investors. The danger of displacement is very real.

To try to avoid displacement, the original ABIF design had as an eligibility criteria that the investment project should have failed to attract commercial finance. In other words, the fund would only provide a grant if the banks wouldn't touch the investment project. Sounds sensible at first glance, but two issues:

1. Another blunt instrument; why shouldn't ABIF support projects that banks would fund, but because of high interest rates (about 20% in Afghanistan) the investment is non-viable?
2. The risk of gaming is obvious; get a letter from a bank declining a loan, and go to ABIF where you can get free money!

Again, the risk based investment strategy provides an answer. By understanding the investment project risk and the cashflows associated with the project, it is relatively straightforward to arrive at a weighted average cost of capital (WACC) that turns a commercially viable project into a viable investment opportunity. In any country, the WACC would be derived from the cost of equity and debt to the investor. The justification for ABIF providing grants is that the WACC for Afghan investors is so high that investment doesn't pay, the returns are simply not high enough to cover the cost of the finance.

ABIF cannot change the risk that determines the cost of capital, and it cannot change the future investment project cashflows, but it can reduce the size of the original investment to achieve those cashflows. So ABIF provides an investor with a third financing option; in addition to equity and debt, the investor can apply for a grant. We know the cost of debt (about 20%), we know the cost of grant (0%) and if we can find out the cost of equity (in Afghanistan, retained earnings), we can calculate the appropriate level of grant that would avoid displacement of other sources of finance. Knowing this number would allow us to provide a complimentary source of finance, rather than a competing source of finance. ABIF will be able to offer grants at a level that fits into the realities of the commercial finance market.

The way forward
Having designed and agreed the principles of this risk based investment strategy, the next step is to work out the details and put it into practice. So far as we are aware, this is the first time that a challenge fund has tried to bring this degree of rigour to its grant giving so there is no model to follow from the world of private sector development.

Our plan is to conduct a kind of business confidence survey. By speaking with business leaders around Afghanistan, we can understand their perceptions of risk and therefore their implied cost of equity. This survey will need careful planning and design, but we do think that the results will allow us to:

1. Arrive at an estimate of the implied cost of equity; and
2. Uunderstand varying risk perceptions around Afghanistan.

Armed with this understanding, the next step will be to interpret the survey findings and incorporate them into the detailed investment modelling tools. Given that applicants have to demonstrate the commercial viability of their proposals (one essential tool being a cashflow forecast), we should have the necessary data to complete the calculations.

It will be extremely interesting to see how this works in practice, but it is particularly pleasing that we now have an investment strategy that will ensure that the project is as rigorous about controlling its inputs as it is about measuring its results.