Impact and duration
I have been thinking about the relationship between impact and the duration of investments. When should an impact investor tie their money up for many years, as opposed to recycling their capital more quickly and making more (impactful) investments over time?
This is a complicated question. We might want to get into the difference between social returns (impact) and private returns, and ask whether some projects with high social returns might be long-term in nature.
But here, I think, is one aspect of the question. It has to do with the distinction between the time that the underlying real investment (“project”) takes to generate returns and survive, and the tenor of the investment instruments. A long term project can be financed by short term instruments. I am sorry if this is perfectly obvious, but it’s only just struck me!
Start with the simplest possible example in which there is no difference between private and social returns, cash flows generated by the underlying real investment are constant, and “impact” equals the lifetime returns of the project.
Consider a real investment that takes 10 years to pay for itself. You could finance that with one 10-year loan, or five 2-year loans, one after another.
So here the question of whether the long or short term debt has more impact is really a question of attribution. The 10 year loan is responsible for the whole project, and its whole impact. The 2 year loan is only responsible for one fifth of it, so it has less impact.
Our objective is to maximise impact. In this case moving to shorter tenor instruments so we can recycle our capital more quickly and deploy more each year would do nothing to raise our impact. We could deploy 5 times more, but each dollar deployed has 1/5th of the impact.
If short terms loans actually had underlying investments that paid for themselves quickly, you’d choose them over 10-year loans all day long. But that option is usually not on the table - investment projects that pay for themselves in a few years are very rare, because that’s a very high return project.
I suspect this could hold more generally. Short-term instruments might often be responsible, in an attribution sense, for only part of the impact of underlying project, whereas longer-tenor instruments are more often likely to be responsible for the lot (or at least, more of it). Of course you can split attribution up between multiple investors co-investing at the same point in time too, but that’s is true regardless of the duration of the instrument. The point here is that sometimes attribution between different investors needs to happen over time too.
Trade finance is a short term instrument that needs to be refreshed or its impact stops. Other investors might take up the baton if you exit, but if they do, they get credit for subsequent impact. Short term working capital loans are similar. They need to be rolled over to sustain the impact. Some short term investments imply that other investors would be subsequently needed if you exit to sustain the impact of the impact-generating activity. If you don’t exit, you are really make a long-term investment via a series of short-term investments.
But this attribution of impact between investors over time contradicts how I think about equity (by which I mean primary fund raising or growth capital) — or anything else where it seems intuitive that the impact is created by those who are in at the beginning. With equity we might say the impact is created by the initial ticket that pays for the underlying real investment, and if you can sell your equity two months later then you can swan off taking credit for the lifetime impact of the thing. At BII, we don’t give ourselves credit for impact when we buy equity in a secondary transaction, unless we are doing something else too, to change the impact of the underlying enterprise. With project finance, we might say the impact is created when you originate and close the project, and if the long-term debt you provide is refinanced once the construction phase is over and you exit, your impact is not diminished.
In these cases, the attribution of impact across investors looks like giving all the credit to the lender who supplied the first two-year loan in the original example, and gives no credit to whomever supplies the next four 2-year loans. So which approach is right — if many investors are involved over time, is the impact created by the first mover or should it be divided up between the initial and subsequent investors?
Perhaps the answer is to look at things from the project’s point of view. With equity, for example, it is the initial growth capital that does the work, and who owns the equity subsequently is immaterial (ignoring potential non-financial contributions of shareholders). But a 10-year payback project cannot happen with just one 2-year loan. It needs those subsequent 2-year loans.
Sometimes you may hold an investment for only a short period, but the impact was created at the start and whether you exit or stay invested does not affect impact. Sometimes an instrument might be short-term, and it’s one of a series of investments that are necessary to sustain the activity.
At least, in theory. This post has been something of a thought experiment to make the point that with some instruments the attribution of impact might need to be split between different investors over time. Hence switching from long duration instruments to short duration can diminish your impact per transaction. But that’s not necessarily what’s actually happening when DFIs shorten the duration of their investments. Because sometimes that could mean exiting equity or debt positions sooner, in secondary transactions that do not leave the project needing to raise more money to survive.
