“The entity that would pay for prevention is often not the entity that benefits from it — and this mismatch kills upstream thinking.” — Dan Heath
Upstream thinking fails not because people don’t see the value of prevention — most people understand intuitively that prevention is better than cure. It fails because the financial incentives in most systems are structured to reward downstream action and to make upstream investment structurally irrational.
The core problem is a mismatch: the organization that would pay for an upstream intervention is usually not the same organization that would benefit from its success. This misalignment is not accidental — it is baked into how governments, healthcare systems, insurance markets, and corporations are structured.
Consider healthcare. A health insurance company could invest in programs to help its members lose weight, quit smoking, and manage chronic conditions — classic upstream interventions that would reduce expensive hospitalizations and procedures downstream.
But here’s the problem: the average American changes health insurance plans every few years. If a health insurer invests in a five-year weight loss program today, the financial benefits — reduced heart disease, diabetes, and cancer treatments — will largely accrue to whatever health insurer the member has when those conditions would have developed. The insurer that invested in prevention pays the cost; the insurer that the member happened to have a decade later captures the savings.
This structural problem makes upstream investment irrational from a pure financial perspective, even when it is obviously beneficial from a social perspective.
One innovative solution to the funding mismatch is the “Pay for Success” model — sometimes implemented as a “Social Impact Bond” (SIB), though the name is slightly misleading (they are not actually bonds in the traditional financial sense).
The basic structure:
This structure forces upfront clarity about what “success” means, aligns financial incentives with social outcomes, and transfers risk from the government to the investor. It also creates strong incentives for rigorous measurement and program quality.
The Peterborough Prison experiment in the UK — one of the first SIBs — showed meaningful reductions in reoffending rates among short-sentence prisoners, demonstrating that the model could work in practice.
In theory, insurance companies are ideally positioned to fund upstream interventions — they bear the downstream costs of the problems that prevention could address. But as the healthcare example above shows, customer mobility undermines this incentive.
Medicaid — the US government health insurance program for low-income individuals — has a different structure. State governments largely administer Medicaid, and state populations are relatively stable. If a state government invests in upstream maternal health programs today, the savings on NICU costs, developmental disabilities, and childhood health problems will largely accrue to that same state’s Medicaid budget.
Several states have discovered this: investments in prenatal care, smoking cessation for pregnant women, and treatment of maternal depression produce substantial downstream savings for state Medicaid budgets. North Carolina’s Medicaid transformation program invested in social determinants of health — food, housing, transportation — and documented significant reductions in ER visits and hospitalizations.
When the entity paying for prevention is also the entity that captures the savings, upstream investment makes financial as well as social sense.
The funding mismatch doesn’t only operate at organizational level — it operates in individual decision-making too. People invest less in their health, their education, and their financial planning than would benefit them in the long run, in part because the costs are immediate and the benefits are distant.
Behavioral economists have shown that “commitment devices” — mechanisms that lock people into their own good intentions — can dramatically increase investment in upstream personal behaviors. Automatic enrollment in 401(k) plans increased retirement savings rates from ~30% to ~90% in many US companies. Pre-commitment to health behaviors has similar effects.
The design of default options in systems matters enormously for upstream personal behavior. Making the healthy, future-oriented choice the default — rather than requiring active effort to access it — is one of the most powerful upstream interventions available.
Identify an upstream intervention you believe would be beneficial. Map out who would pay for it and who would receive the financial benefit if it succeeded. Is there a mismatch? If so, what structural change or incentive alignment could fix it?