Outcome-Based Funding for Social Development: A Theoretical Critique
Part II

By V Santhakumar

Extending theoretical understanding of OBF

When a philanthropic organization provides money, say, x to a service provider to create a socially desirable outcome, there could be a social gain, even though there is no monetary private gain to the organization. Let us assume the social gain to be s. If there are n projects funded by the organization, then for m (m≤n) projects, s could be greater than x (denoted by S1) and lesser than x for the remaining projects (let that be S2). The lower gains in the second category of projects could be due to the inadequate focus on outcomes. (For simplicity, let us assume that each of these n projects require the same expenditure). When the organization transfers an amount x to the service provider, there is a need for an additional expenditure p*x (where 0≤p≤1) to ensure that the funding leads to the expected outcomes, and this is part of the transaction costs. The total expenditure is nX+ npX or nX(1+p), and the total social gain is mS1+(n-m) S2. The net gain is {mS1+(n-m)S2} – {nX (1+p)}. The ratio of the net gain to the total expenditure is given {mS1+(n-m)S2} – {nX(1+p)}/ {nX(1+p)}, and this is the social return. Let this be `r’.

There are two possible benefits of OBF. There could be a reduction in the cost of achieving the same social gain. In addition, an increase in social gain is possible if investors are able to achieve outcomes in riskier or uncertain projects. For achieving the same social gain through OBF, the maximum that the philanthropic foundation may be willing to pay to investors is nX (1+p). On the other hand, if OBF increases social gains to Si, then the maximum return that the organization would be willing to pay to the investor could be r*{Si-(mS1+ (n-m)S2}. Investors may have to spend nX(1+q) for these projects where q is the share of transaction costs. Hence, the additional gain made by the investors would be the difference between nX(1+p) – nX (1+q) which is equal to the nX (p-q) if there is no increase in social gains due to the OBF. On the other hand, if there is an increase in social gain, the return could be r*{Si-(mS1+ (n-m)S2}+nX(p-q). Let us consider some of the implications of the OBF (based on the description here) in the form of a set of propositions.

Proposition 1: The return to the investor in an OBF program could be the compensation for the reduction in the transaction costs involved in the direct funding of service providers by philanthropic organizations. Given the difficulty in assessing social gains, the investors may not be getting a return on the risk that they take in such investments.

The riskiness in social projects is in terms of achieving social gains and not in monetary gains to investors. Such a social gain cannot be measured easily. Hence, investors may not get back a return on the basis of any possible increase in social gains due to their intervention. This is a known ex-ante for investors. Hence, they may focus on the reduction in transaction costs discussed in the previous section and the associated return. Therefore, the presence of investors in OBF may not enable investments in projects the outcomes of which are uncertain due to external factors (which are beyond the control of the service provider.) This could be the reason for seeing the projects funded by impact bonds as not that risky (Gustafsson-Wright and Boggild-Jones, 2019). Hence, the riskiness inherent in the portfolio of projects (to be) taken up by a philanthropic funder (interested in the enhancement of social welfare) may remain by and large the same with or without the use of OBF. The OBF may not be able to generate additional financial support for experimental projects, and these may be funded only through the direct connection between funders and service providers (and the willingness in this regard on the part of funders may continue to impact the availability of such funding). The limited available empirical evidence too indicates that the major impact of OBF could be the reduction of the transaction costs required for direct funding. This is evident from Gustafsson-Wright and Boggild-Jones (2019), who note that the focus on outcomes driving performance management and developing a culture of monitoring and evaluation are some of the changes which could be visible through the use of impact bonds (one type of OBF).

Proposition 2: OBF may not increase the quantum of resources available for financing social development.

Though this is one of the claims of the proponents of OBF, and as noted by Gustafsson-Wright and Boggild-Jones (2019), it may not materialize since the upfront capital that is given to the service provider by the investor is paid back by the philanthropic funder (after seeing successful outcomes). This is something that can be predicted easily through an ex-ante theoretical analysis of OBF. There could be an increase in the total quantum of resources only when the (transaction) costs involved in using the currently available philanthropic resources is much greater than the reward to be paid back to the investors. Moreover, there could be an increase in the quantum of resources for this purpose, if the current high transaction costs discourage certain philanthropic and other actors from investing in the social sector.

Proposition 3: OBF may increase the availability of funding to for-profit service providers in the social sector.

There are for-profit service providers like fee-charging private schools and hospitals. Being for-profit organizations, these are likely to focus on those sections of society which can pay for their services, and the tangible aspects of social services, such as marks or passing in examinations. These organizations may be able to break even over time, but some of the potential entrepreneurs in this domain may not have enough capital. Though it may not be very difficult for them to access capital from normal financial institutions, OBF can enhance their access to capital. Moreover, these organizations may have an incentive structure and culture to produce outcomes to meet the needs of funders. Philanthropic funders may be interested in funding a part of the cost of these for-profit service providers (or pay back investors who provide capital to such service providers) since through such funding, these for-profit service providers may be able to extend their service to a section of society which cannot pay for it.

Proposition 4: It is not clear what prevents philanthropic organizations from directly providing funds to for-profit service providers of the type mentioned in proposition 3. 

There could be a need for contracting with such service providers since capital is provided upfront. There could be transaction costs before and after the signing of the contract. However, there is no strong reason to believe that the investors who are part of OBF would encounter a significantly lesser cost for this purpose than philanthropic organizations (if they are funding service providers directly). Such a situation may arise if the internal structure (including that of the decision-making process) of the philanthropic organizations significantly enhances the transaction costs in dealing with such service providers. In such a case, the attempts should be to reform the decision-making processes in philanthropic organizations.

Proposition 5: Though there could be a pecuniary interest on the part of all service providers, it is not clear whether not-for-profit organizations would be willing to trade-off other interests just for pecuniary benefits.

A major set of organizations working in the domain of education are not-for-profit organizations. These may have multiple interests, including ideological ones, or may be driven by specific ideas on what should be achieved in education. These may be interested in seeking funds, but it is not clear whether that would encourage them to participate in instruments such as OBF.

Proposition 6: It is not clear how investors would achieve greater savings in transaction costs compared to philanthropic foundations when the former deal with not-for-profit organizations which are interested in achieving both tangible and intangible outcomes.

Not-for-profit organizations may not have a culture of meeting the outcome needs of a financial investor. The intangible outcomes may take time. Given their interest in achieving a holistic improvement of the situation, their intervention could be affected by a number of external factors, which can have a negative impact on the achievement of outcomes. Since the philanthropic foundation is also a not-for-profit organization, probably it may have internal systems which suit the needs of not-for-profit service providers rather than those of financial investors. Hence, we should not be surprised if there is reluctance on the part of the not-for-profit service providers to participate in OBF or if there is a greater likelihood of funders directly dealing with such service providers.

Proposition 7: OBF is less likely to deal with public organizations (such as government schools) as service providers.

There could be a number of reasons for this. Government schools are expected to work on tangible and intangible outcomes and have a public service orientation. They are less likely to have an `outcome’ orientation and may not have an internal incentive structure that can be changed through OBF. Such public organizations must continue to play an important role in the provision of social services since they are generally the main providers of such services to the poor and vulnerable sections of society. There could be scope for OBF to users (like grants to parents indexed to the school attendance of children) but these may be carried out directly by the funders (which could be the government or philanthropic organizations).

Proposition 8: While considering certain insights of behavioural economics or relaxing the assumption of interest to make monetary gains on the part of service providers, it may not be surprising to find that service providers are reluctant to participate in OBF.

The presence and role of intrinsic motivation are noted in not-for-profit organizations, whether these are philanthropic organizations or those which provide various services to people directly.¹ Such an intrinsic motivation may be useful to mitigate (partly) the principal-agent problem and to see that not-so-measurable and intangible aspects of services in education and healthcare are also covered. The possible incompatibility of such a motivation with financial or extrinsic incentives is also noted in literature.² There could be an incompatibility between the intrinsic motivation of the founder or employees of the service provider and the incentive structure of OBF. The fact that their services are bought by an investor who is likely to get a financial return from a philanthropic foundation may demotivate them. They would rather work with the philanthropic foundation directly. This could be one reason for the noted reluctance on the part of many service providers (in the domain of education) to make changes in their organizations to meet the needs of OBF.³

Experiments in behavioural economics have also noted that people are willing to sacrifice for a sense of fairness (Guth et al, 1982;⁴ Kahneman et al, 1986;⁵ McCabe et al, 2003⁶). Such a sense of fairness may make them concerned about the surplus that the investor makes through their actions (especially when OBF does not seem to have a provision to share the surplus with the service provider). All these discussions do not mean that all not-for-profit organizations will not participate in OBF. Some of these may not see a problem in getting money even if that contributes to the financial gains of an investor. They may participate if direct contact with the philanthropic funder is not possible due to various reasons. In such contexts, a reform of the processes that connect funders with service providers would be more appropriate than the use of an instrument like OBF.

Proposition 10: There is no reason to think that OBF would be used effectively by governments in developing countries; even if this is attempted, it may not make a significant impact on the provision of social sector services.

In most cases, governments in developing countries have not made enough investments in social services. This is partly due to the limited availability of resources. There may be a need to have a basic building up of government owned and controlled service provisions in sectors, like education and healthcare, and dependence on funded or subsidized private provision may not be adequate. For example, there may be a need to strengthen government schools in a country like India.⁷ Governments may not have adequate resources to fund non-governmental service providers, and even if this is attempted, it may not enhance the overall quality of such social services. Hence, the funding for OBF may have to come from non-governmental donors.

Proposition 11: OBF need not enhance the discretionary power of service providers in deciding the course of actions while focusing on pre-determined outcomes.

In the case of OBF, there is an intermediary – the investor who is interested in ensuring that the service provider achieves pre-determined outcomes. This actor may influence the plan of action, which may reduce the discretionary power of service providers.

Proposition 12: OBF is less likely to lead to the scaling up of interventions that can achieve planned outcomes.

This is inferred from propositions number 2, 7, 8 and 10. The scaling up of actions in the social sector requires the participation of governments, public service providers, non-governmental organizations, and the availability of large financial resources to be invested in the sector. These are less likely to materialize through the use of OBF.

There could be a tendency to consider OBF as part of social investments (including green investments) which have increased over time (Perez, 2007).⁹). However, the role of the investor in the present form of OBF does not seem to be driven by any of these considerations. On the other hand, philanthropic funders could be driven by all these considerations, but their overriding concern in the participation in OBF seems to be the reduction in the transaction costs using a for-profit intermediary.

Epilogue

What could be the reasons for the use of OBF and an inflated expectation of its possible benefits? It could be that philanthropic organizations are populated and influenced by a set of finance professionals, or they may be extending the framework of profit-motivation into social impact investments. This is not new or unusual since there are arguments that private motives, financial and market incentives can address certain social problems. It is true that these strategies may work in the case of certain aspects of the social sector, for example, making the service of private education providers available to the poorer sections of society. However, it may not address a major part of social issues. These finance professionals may be neglecting two kinds of developments noted in literature. The first is within behavioural economics, which demonstrates the multiple motivations behind investments (some of these are at variance with the profit-motivation described in conventional microeconomics). Secondly, they may seem to think that the way to address the principal-agent problem in not-for-profit settings is the same as that in for-profit organizations. There is a sizable literature on the presence and need for different motivations and the consequent need for different strategies to address the principal-agent problem within, and in dealing with, not-for-profit organizations and these cannot be neglected. Such neglect may lead to the design of alternative instruments which are not so useful.

Author
V Santhakumar, Professor, Azim Premji University, Bangalore  

Outcome-Based Funding for Social Development: A Theoretical Critique Part I
A Discussion between Shantanu Ghosh and V Santhakumar

Featured image by Frank Busch on Unsplash

References

Baron, J. (2000). Thinking and Deciding.  New York: Cambridge University Press

Benartzi, S., & Thaler, R. H. (2004). Save More Tomorrow: Using Behavioural Economics to Increase Employee Saving. Journal of Political Economy, 112, 164-187

Clist, Paul, and Arjan Verschoor. 2014. The Conceptual Basis of Payment by Results. School of International Development, University of East Anglia

Edmiston, D. & Nicholls, A. (2017). Social impact bonds: The role of private capital in outcome-based commissioning. Journal of Social Policy, 47(1), 57-76

Perakis, R. & Savedoff, W. (2015). Does results-based aid change anything? Pecuniary interests, attention, accountability and discretion in four case studies. CGD Policy Paper 053. Washington DC, Centre for Global Development