What it means
Governments pay for social programmes whether they work or not. A social impact bond reverses the order: private investors fund the programme first, and repayment depends on measured success.
The structure chains four parties: a government defines an outcome, fewer reoffenders, say, investors fund a charity to pursue it, an independent evaluator measures the result, and government pays only against the evidence. Despite the name it is not a bond in the investor's sense: there is no coupon and no guaranteed principal, and the return is a success fee dressed in fixed-income clothing.
The UK government pioneered the model, and its published guidance on social impact bonds sets out the rationale: transferring performance risk to investors while paying for outcomes rather than services. The first experiment ran at Peterborough prison from 2010: investors funded rehabilitation work with short-sentenced prisoners, and repayment hinged on reoffending falling enough against a matched comparison.
The appeal is alignment: charities get multi-year funding, governments pay only for what works, and investors take the risk that good intentions do not survive measurement. The criticism is overhead and gaming: evaluation costs, complex contracts, and the temptation to serve the easiest cases have kept the market small relative to its publicity.
For a non-finance reader, a social impact bond is pay-for-results with borrowed money: the public pays for the bridge after someone walks across it, and the lender bears the chance nobody does. The market's geography spread faster than its volume: dozens of countries launched pilots, and development impact bonds extended the model to programmes funded by donors rather than domestic treasuries.
Evaluation is the hidden protagonist: without a credible counterfactual, the payment is politics, so randomised or matched comparisons became the model's load-bearing wall. The rebrand tells a story: practitioners increasingly say social outcomes contracts, shedding the bond label that promised a safety the instrument never had.
In practice
Real-world examples.
Example
A treasury pays $5.9 million against verified year-long job placements, repayment scaled to measured outcomes. Evidence set the price. Had the placements fallen short, the payment would have fallen with them.
Example
Contract design consumes nine months on outcome definition, comparators and pricing before delivery starts. Lawyers, the evaluator and the charity all need to agree what counts as success. That up-front cost is one reason the model suits only problems where failure has been expensive.
Example
Five-year funding with pay-on-evidence lets the charity drop methods the evaluator shows are failing. It does not need to wait for a funder's annual review to change course. The investors accept that its changes may help or hurt their return.
Formula
Calculation
Outcome payment = verified outcomes x agreed price per outcome. Outcome targets are fixed contractually, an independent evaluator measures results against a comparator, and government repayment plus return scales with achieved outcomes, falling to zero if targets are missed.
Worked example. Investors commit $5 million to a training charity, and the government agrees to pay $5,900 for each participant verified in work for a full year. If 1,000 participants qualify, the payment is 1,000 x $5,900 = $5.9 million, so investors recover their $5 million plus $0.9 million, a total return of $0.9 million / $5 million = 18% over the life of the contract. If only 700 qualify, the payment is 700 x $5,900 = $4.13 million and investors lose $0.87 million, and if the evaluator verifies none, nothing is paid and the $5 million is lost.Case study
Seen in the real world.
This case study is fictional and illustrative. A made-up regional government wants to cut youth unemployment and hates paying for programmes that photograph well and fail quietly. Its treasury signs a social impact bond: investors commit $5 million to a training charity, and repayment scales with verified job placements sustained for a year. The contract negotiation teaches the design's real difficulty: defining the outcome, sustained employment, not interviews, choosing the comparison group, and pricing each result consumed nine months and three law firms.
The delivery years vindicate the alignment argument: the charity, funded for five years and paid on evidence, iterates its method ruthlessly, dropping the classroom curriculum that evaluation shows is dead weight. The final evaluator's report triggers a payment of $5.9 million, an outcome the treasury estimates would have cost more as a conventional programme with typical failure rates. The post-mortem is honest about scale: the structure's discipline is real and its transaction costs are heavy, so the region reserves it for problems where measurement is possible and failure has been expensive. The charity's director frames the souvenir, a coffee mug reading paid on results, and admits the phrase changed how every programme she runs is designed.
Watch out
Common mistakes.
- Treating it as a bond; investors risk principal and earn a success fee, with no coupon, so it behaves like outcome-linked equity in a program.
- Assuming it suits every service; outcomes must be measurable and attributable, which excludes much of what governments fund.
- Ignoring the overheads; evaluation and contracting costs are substantial, so the model earns its keep only where failure is expensive.
Questions
People also ask.
What is a social impact bond?
A pay-for-success contract: private investors fund a social program, and government repays with a return only if independently measured outcomes are achieved.
Is it a real bond?
No; principal is at risk and there is no coupon. The return is a success payment tied to outcomes, not a debt obligation. The name oversells its safety.
Where did it start?
The UK pioneered the model, with the first at Peterborough prison in 2010, funding rehabilitation repayable against reduced reoffending.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related