Outcome-Based Funding for Social Development: A Theoretical Critique
A Discussion between Shantanu Ghosh (SG) and V Santhakumar (VS)
This discussion follows V Santhakumar’s article on the topic that is published in two parts here:
Outcome-Based Funding for Social Development: A Theoretical Critique Part I
Outcome-Based Funding for Social Development: A Theoretical Critique Part II
SG: I really liked reading the paper and the logical flow of the arguments and propositions that you have made. I also noted that you have attempted to keep an objective, evidence- and model-based view in your critique. Your critique and observations and the benefit and applicability model and equation all resonate.
Outcome-based Funding (OBF) or Pay for Success (PFS) constructs are additive to traditional developmental spending, not a substitute. It is fit for specific purposes and circumstances and cannot and should not be imposed on all situations. The necessary (but not sufficient) condition is the ability for objective measurement of outcomes. Outcomes can be in many forms (measures of learning ability in primary education is a very valid objective outcome). The sufficiency conditions are varied according to the situation. So, the question is not whether OBF/PFS is generally better or worse than traditional development spending – the more apt question is: Are there situations where that model provides (mostly through behavioural economics but also through microeconomic logic) a higher chance of better outcomes and at what cost? The real challenge is determining that incremental cost/benefit trade-off curve and I would submit that in many cases, an attempt at accurate modelling will fail compared to setting ambitious goals and acceptable delta on costs. In fact, in the commercial world today, especially in transformation, it is empirically established that scenario modelling, as opposed to exact transfer function determination, serves a much better purpose.
VS: Completely agree; there can be a certain improvement in the effectiveness in certain cases and some increase in capital coming to social financing, but excessive claims are not good; there is a need for context-specific understanding and elaboration of possible benefits.
SG: I agree that excessive claims and positioning as a panacea will do more harm than good. We need to understand it is another way, not the only other way.
Commercial investors who participate must come from a background of commercial expectation but with an added lens on social good. This is not an esoteric concept with the growth of ESG[i] investing. What it means in financial structure terms is that we need to create a more moderate return expectation as a unique asset class combining the two objectives as opposed to a pure market supply-driven arbitrage model between all financing options. I would also submit that the market has not deepened enough to establish the right pricing. There are capital loss examples in some global structures as well as outsized returns in some others. And the market is too shallow to take either as a norm. Over a period, a unique return model would be established for a relevant segment of investors.
VS: True. There can be some measurement or elaboration to enable these investors to understand how they have gained in terms of these two objectives. In that sense, innovations like OBF can be seen as instruments that enable the transition of normal investors towards a valuation on the basis of these two objectives.
SG: That has to be the focus on a lot of economic studies – as Ronald Cohen says we moved from a singular focus on return to risk and return and now it needs to be – risk, return and impact. of course, a third dimension complicates every model geometrically.
The key is behaviour and approach to the process. You have made the reference to alignment and agency problems which are all relevant. I would comment on the four dimensions which play a big role. One is the activity/budget-based approach of the most traditional grant-based spending with an inherent minimization (of cost and especially, non-direct beneficiary expense) versus a more outcome-based, incentive model to shape and modify the intervention within guard rails. This has got nothing to do with intrinsic motivation but just the realities of two different models.
VS: I agree that the alignment in conventional models is an issue. I will talk about some of the problems at the end. However, my concern is: won’t there be an implicit cost-minimisation in OBF? In that sense, are these two models different?
SG: I actually do not think most OBFs operate from a cost minimization focused philosophy. Of course, cost consciousness is always a base criterion, but it is probably not as centre stage as in traditional grant models. And in some ways, one of the defining features of good OBFs should be an explicit demonstration that the model has sought for value maximization versus cost.
Number two is the time dimension; again, grant activity is seldom locked in for the medium- to long-term and varies dramatically from other external stimuli (e.g., COVID response) whereas OBF models lock in commitments. Again, the question here is not whether an efficient market should automatically repurpose spending to the highest ROI (all forms of return) interventions (as may be well justified by the COVID situation) but whether for that specific intervention, OBF with longer-term commitment provides a better chance of enhanced outcomes.
VS: I think there is a certain realisation of the issue of time horizons for outcomes in conventional models. I see a change in philanthropic organisations in this regard. However, there is no assurance that such a change would lead to outcomes.
SG: I agree. Again, the two models need not be at the opposite ends of the spectrum and over a period both should imbibe better practices of the other and converge towards the centre.
The third is the rigour in definition – scoping and measurement – the standard of which, by definition, needs to be equal or higher for OBF.
VS: I think the incentive structure of OBF ensures that, and that is not my concern. There could be a self-selection of projects. There could be a non-selection of projects where the change is holistic, and the measurement has to be complex.
SG: Number four, the more ambitious OBF programs should make a few providers collaborate and the way to drive coherence in that kind of ecosystem leverage model is through an objective outcome defined structure and incentives than a budget-driven mandate or best-effort basis model.
VS: Agree; what you suggest may enable learning and course correction too if I have understood the argument correctly.
SG: These four factors – maximization of outcome approach vs task execution approach; pre-determined time horizon and certainty of investment and better rigour in definition, scope and measurement and ecosystem leverage have, at least in the business world, proven to be critical enhancers of performance and outcome. Exceptions have typically proved the rule. And while the social sector is different from a complexity, core motivation and categorization of outcome perspective, I am sure that these fundamental factors that shape the human approach to problem-solving, executing, collaborating and adapting approaches are still applicable.
General Observations
VS: I don’t think that the conventional funding models are very effective. You may see some articles on these links:
Philanthropic Foundations and the Government: Challenges in the Relationship
Philanthropic Foundations and NGOs – Challenges in the Relationship
Challenges in Managing Employees of an Altruistic Organisation
I think what is needed is innovation for social finance enhancement and effectiveness. All these efforts [OBF] are part of this, and we should learn correct lessons. Each attempt is useful but has limitations too. For example, RCT or Impact Evaluation in the conventional sense is useful but has limitations. There should be a continuous effort to learn and address limitations and challenges.
SG: Absolutely! In fact, one of the best practices should be to have a robust debate on the right evaluation model. We recently did that for our Haryana structure and I thought the process forced a very robust dialogue with a bunch of experts. I am sure there will be questionable choices, but a good process must ensure that dialogue and then, the chances of better metrics will be significantly higher.
In summary, I agree with the way you have approached the critique and hopefully, some of the common-sense reflections from my three decades experience of performance management (with variable results) would resonate with you. I think the OBF market is way too nascent to have its past paint the future in terms of applicability – our lens needs to be whether there should be a logical justification for part of the developmental spend move towards OBF and studies and I hope this would strengthen the design of such structures.
Shantanu Ghosh, after 30 years of service in the corporate sector, moved to the development sector and is currently CEO, Social Finance India.
V Santhakumar is Professor, Azim Premji University, Bangalore.
[i] Environmental, Social, and Governance
