Yesterday, I made a pitch to a foundation to fund the development of an adaptation of the Social Impact Bond idea. During the discussion, one of the selection committee asked me, “this bond will lock government in to producing these outcomes for ten years. Do you think that is a good idea?”
I wasn’t expecting such a question, but the role of commitment on the outcomes of government policy has been a concern of mine (see this report for more details). So while I was unprepared, I had definite views on commitment and government policy. So, I replied, “as someone who has worked in government for a while, one of the main problems is that government lacks credible commitment. So, yes, I think it is a good idea for government to be locked in to produce social and environmental outcomes. Otherwise, their priorities will change every three years with a change in government. I think impact investment is a great commitment device. It ties the hands of the King.”
My time ended soon after. But I was still thinking about this exchange. My point was that if you can lock government in to generating outcomes that we know are beneficial (i.e. reduce Indigenous disadvantage), that will ensure any change of government wouldn’t affect the work to create these impacts. In fact, that is one of the rationale for Social Impact Bonds. It creates a long-term stream of funding for social service organisations without being affected by political uncertainty. Furthermore, it provides the operational freedom for service organisations to choose the best strategies to create impact. All that matters to government, is that the impact sought is generated.
I think my questioner’s concerns were two-fold. First, locking-in outcomes could result in the generation of the wrong outcomes. Second, how do you know the Social Impact Bond is measuring the right outcome anyway?
These concerns are valid. But it depends on the choice of the outcomes and the metrics. The same could be said of government policy in general. In fact, at least Social Impact Bonds have a finite life. With government policy, the impacts could self-perpetuate as bureaucracies grow to administer the policy. Furthermore, outcomes are not transparent in general so policy-makers and stakeholders may not know if the policy has been effective. So, this is a generalised concern for any form of government policy. With Social Impact Bonds, you can clearly see if it has been a mistake. This doesn’t stop government from prematurely breaking the contract if this is the case.
But how can we be sure that the choice of the metrics will be the right one? I think it is hard to have a productive discussion about specific metrics unless you have the data before you. Often, in social and environmental policy, it hasn’t been collected or at the right level of disaggregation. Until you have done the hard work to find out what you know and don’t know, the discussion of metrics is a bit academic.
Impact investment could indeed lock governments in to outcomes that may have unintended consequences. But impacts are also transparent under impact investment. This would produce useful information that would help build policy-makers’ and researchers’ understanding of the drivers of impact. It would also be useful, if we selected the right outcomes and metrics to show that there is a solution to a deep and persistent social problem. Surely more evidence and consistency is better than less.