- Atlas
- Medicine & Health
- Payment And Incentives
Payment And Incentives
Health care is unusual among markets in that the person who receives a service rarely pays for it directly, and the person who orders it is often not the person who bears its cost.
9 frameworks, from 1963 to today · Health Services Research · Medicine & Health
- Moral Hazard and Demand Response 1963-Present
- Supplier-induced demand 1970-Present
- Capitation 1973-Present
- Bundled payment 1985-Present
- Quasi-Experimental Program Evaluation 1985-Present
- Incentive-centered design 1990-Present
- Output-based aid 2000-Present
- Pay-for-Performance 2000-Present
- Accountable care organization 2012-Present
Overview
Health care is unusual among markets in that the person who receives a service rarely pays for it directly, and the person who orders it is often not the person who bears its cost1. Payment and incentives is the subfield of health services research that studies this triangle: how money flows to doctors, hospitals, and other providers, and how the shape of that flow changes what they do2. Its central insight is that payment is not a neutral bookkeeping arrangement but a set of instructions. Change how a provider is paid and you change which patients they accept, which services they offer, how hard they work, and sometimes whether they tell the truth. The field's history is a long argument about how powerful those instructions are, how cleverly providers respond to them, and whether any payment design can align what is good for the provider with what is good for the patient and the public purse.
The Insurance Problem and the Birth of Moral Hazard
Modern payment research begins with insurance. In the 1960s, economists applied the new tools of information economics to health coverage and identified a structural problem: once a person is insured, the price they face at the point of care falls toward zero, so they consume more care than they would if they paid the full cost. This is moral hazard — not dishonesty, but the ordinary response of any insured person to a lower price3. The term had been used in insurance since the nineteenth century; health economists gave it a precise meaning and a policy edge3.
The decisive empirical test came with the RAND Health Insurance Experiment, a randomized trial run in the United States between the mid-1970s and early 1980s. Families were assigned to insurance plans with different levels of cost-sharing, from free care to a large deductible. The result was unambiguous: people with free care used substantially more services than those with cost-sharing, and the extra care produced little measurable improvement in health for the average person. The experiment established that demand responds to price and that insurance design is itself an incentive instrument. It also sharpened a question that has driven the field ever since: if patients respond to price, how do providers respond to payment?
Supplier-Induced Demand and the Physician as Agent
The physician occupies a strange position: the patient trusts the doctor to recommend care, but the doctor also sells that care. In the 1970s, health economists began modeling this as an agency relationship — the physician acts as the patient's agent, but with their own income at stake. The concept of supplier-induced demand described the possibility that physicians, facing a target income or a fall in their patient base, could shift the care they recommend in ways that serve their income rather than the patient's need.
The evidence was contested from the start. Studies of regions with different physician-to-population ratios found that areas with more surgeons performed more surgery, a pattern consistent with induced demand but also explainable by patient preferences or illness patterns. Researchers developed indirect tests — watching how physicians' practice changed when fees changed, or when new physicians entered a market — and the debate over how much induction actually occurs has never fully closed. What the debate achieved was durable: it made the physician's dual role a permanent feature of payment design, and it gave the field its central vocabulary of agency and conflict of interest. Any payment scheme must reckon with the fact that the person deciding what care is needed may benefit from deciding that more is needed.
Capitation, Managed Care, and the Shift of Risk
If fee-for-service pays for volume, its opposite is capitation: a fixed payment per patient per period, regardless of how much care that patient uses4. Under capitation the provider bears the financial risk of illness, so the incentive flips from doing more to doing less. The idea is old — prepaid group practices existed in the United States in the early twentieth century — but it became a mass phenomenon with the managed care movement of the 1980s and 1990s, when health maintenance organizations (HMOs) enrolled millions of Americans under capitated and gatekept arrangements.
Managed care was a natural experiment in incentive design, and it taught the field two lessons. First, capitation works: it restrained spending and reduced hospitalization. Second, it can work too well. Concerns about underprovision — skimping on necessary care, selecting healthier patients, and restricting access — produced a public backlash in the late 1990s and a retreat toward softer forms of risk-sharing. The episode established the field's recurring trade-off: payment that rewards restraint creates incentives to withhold, just as payment that rewards volume creates incentives to overprovide. The design problem is to find a payment that rewards the right care rather than more or less care.
Pay-for-Performance and the Quality Turn
By the 2000s, attention shifted from volume and cost to quality. Pay-for-performance (P4P) pays providers a bonus for meeting measured targets — vaccination rates, blood-pressure control, timely screening5. The logic was appealing: if payment can shape behavior, pay for the behavior you want. Programs spread through the United States, the United Kingdom, and elsewhere; the UK's Quality and Outcomes Framework, introduced in 2004, paid general practitioners substantially for achieving clinical indicators and produced measurable gains in recorded performance.
But the evidence accumulated cautiously. Reviews of P4P trials found modest average effects, with gains concentrated in areas where measurement was easy and often accompanied by unintended responses — attention shifting toward measured targets and away from unmeasured aspects of care. The field learned that what gets measured gets managed, and that incentive schemes can distort as well as direct. P4P also raised a deeper question: whether financial incentives crowd out the intrinsic motivation that draws people into medicine in the first place.
Behavioral Economics and the Design of Incentives
That question opened the door to behavioral economics, which brought psychology into payment design6. The older model assumed a rational provider weighing marginal income against marginal effort. Behavioral research showed that people respond to how incentives are framed, to loss versus gain, to social comparison, and to defaults. A bonus framed as a potential loss may motivate more than the same sum framed as a gain; public reporting of performance can move behavior through reputation as much as money. The field began to treat incentive design as a craft informed by how people actually decide, not just by how much they are paid. This strand also fed into audit and feedback and other implementation strategies, where the incentive is information and comparison rather than cash.
Episode-Based and Total-Cost Contracting
The most ambitious turn came with attempts to pay for whole episodes and whole populations rather than individual services. Bundled payment sets a single price for all the care surrounding a procedure or condition — a hip replacement, a cardiac bypass — so that providers keep savings if they coordinate efficiently and absorb losses if they do not7. The idea was piloted in the United States in the 1980s and revived on a large scale in the 2010s7. Results were mixed: some bundles reduced spending modestly, others showed little effect, and the design demanded careful definition of what the episode includes.
Accountable care organizations (ACOs), launched in the United States from 2012 under the Affordable Care Act, went further: a group of providers takes responsibility for the total cost and quality of a defined population, sharing in savings if it meets quality standards. The model descends directly from capitation and managed care but softens the risk with shared-savings arrangements. Evaluations found modest savings in some programs and little in others, and the field continues to debate whether the savings come from genuine care improvement or from favorable patient selection. The ACO experiment is the clearest contemporary expression of the field's oldest ambition: to pay for health rather than for activity.
Results-Based Financing in Global Health
Outside high-income systems, the same logic took a different form. Results-based financing (RBF) — paying for verified results rather than inputs — spread through low- and middle-income countries from the 2000s, supported by the World Bank and others. Programs paid health facilities for the services they actually delivered, or paid for outcomes such as institutional deliveries or vaccination coverage. The approach drew on the same incentive reasoning as P4P but operated in settings where the prior problem was often too little care rather than too much. Evidence was mixed and context-dependent: RBF could raise the volume of priority services, but effects on quality and on health outcomes were uneven, and the schemes raised questions about measurement, gaming, and whether paying for results strengthened or distorted fragile health systems. The global strand made the field's questions genuinely worldwide: the same tension between rewarding activity and rewarding outcomes appears whether the payer is a national insurer or an international donor.
The Present: A Field of Trade-offs
Today the subfield is defined less by a single winning design than by a mature understanding of trade-offs. Fee-for-service rewards volume; capitation rewards restraint; P4P rewards measured quality; bundles and ACOs reward coordination and total-cost discipline; behavioral design shapes how any of these are experienced84579. The dominant research approach combines economic theory with quasi-experimental evaluation — difference-in-differences, regression discontinuity, randomized programs — to ask what each design actually does in practice. The open questions are the ones the history kept producing: how to measure quality without distorting it, how to reward good care without rewarding patient selection, how much risk providers can bear without withdrawing from care for the sickest, and whether financial incentives strengthen or erode the professional commitment that payment is meant to support. The field's history is the record of learning that payment is a powerful lever, that providers respond to it in predictable and unpredictable ways, and that designing it well remains an unsolved craft.
References
- Health economics, Wikipedia
- Health services research, Wikipedia
- Moral hazard (introduction, History), Wikipedia
- Capitation (healthcare), Wikipedia
- Pay for performance (healthcare), Wikipedia
- Behavioral economics, Wikipedia
- Bundled payment (introduction, History), Wikipedia
- Fee-for-service, Wikipedia
- Accountable care organization, Wikipedia
The frameworks of Payment And Incentives
Each framework is a way of working the field has used: when it held, and the question it answers.
- 1963-Present
Moral Hazard and Demand Response
How does insurance change how much care people use?
- 1970-Present
Supplier-induced demand
Do doctors provide more care because they profit from it?
- 1973-Present
Capitation
What happens when providers are paid a fixed amount per patient?
- 1985-Present
- 1985-Present
Quasi-Experimental Program Evaluation
How can we tell whether a program actually caused a change?
- 1990-Present
Incentive-centered design
How do people actually respond to incentives and frames?
- 2000-Present
Output-based aid
Does paying for results improve health services in poor countries?
- 2000-Present
Pay-for-Performance
Does paying for measured quality improve patient care?
- 2012-Present
Accountable care organization
Can providers be paid for keeping a population healthy?
How the ideas connect
Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.
Accountable care organization grew out of Capitation
Accountable care organizations descend directly from capitation and managed care, taking responsibility for a defined population's total cost but softening the risk with shared-savings arrangements. The ACO model is the contemporary refinement of the capitated, population-based contract.
Supplier-induced demand influenced by Moral Hazard and Demand Response
The moral hazard framework established that demand responds to price and that insurance changes behavior. Health economists then asked whether the same logic applied to providers, whose own income is at stake when they recommend care, giving rise to supplier-induced demand.
Capitation influenced by Supplier-induced demand
Concern that fee-for-service payment encourages physicians to induce demand made the opposite arrangement attractive. Capitation was designed to remove the provider's financial gain from extra services, a direct response to the agency problem the supplier-induced demand literature had framed.
Pay-for-Performance influenced by Quasi-Experimental Program Evaluation
Quasi-experimental program evaluation supplied the methods — difference-in-differences, instrumental variables, and careful comparison groups — that pay-for-performance researchers used to estimate whether financial bonuses changed measured quality. The credibility revolution in program evaluation shaped how P4P was tested and judged.
Accountable care organization influenced by Quasi-Experimental Program Evaluation
The evaluation of accountable care organizations relied on quasi-experimental designs to compare participating providers with similar non-participants, since the programs were not randomized. The methods developed for program evaluation became the standard way to judge whether ACO savings were real.
Output-based aid influenced by Pay-for-Performance
Results-based financing in global health drew on the same incentive reasoning as pay-for-performance, paying for verified results rather than inputs. The approach adapted the P4P logic to low- and middle-income settings where the prior problem was underprovision rather than overprovision.
Bundled payment influenced by Capitation
Capitation's lesson that providers respond to bearing risk for a defined set of services informed episode-based payment, which applies the same logic to a procedure or condition rather than a person over time.
Pay-for-Performance reacted against Capitation
Pay-for-performance arose partly as a reaction to the cost-focused incentives of capitation and managed care, which were seen as rewarding restraint at the expense of quality. P4P shifted the target from doing less to doing better, defining itself against the underprovision problem capitation had created.
Beyond the field
Where Payment And Incentives's ideas came from in other fields, and what they led to there: the strongest 12 of 30 links.
Incentive-centered design influenced by Mechanism Design in Information Economics
Mechanism design supplied the formal tools, especially incentive compatibility and revelation principles, that incentive-centered design applies to system and interface design. The later framework is a direct outgrowth of that theory.
Carried by Leonid Hurwicz.
Incentive-centered design influenced by Incentive compatibility and implementation theory in Mechanism Design
Incentive-centered design takes incentive compatibility and implementation theory as its core formal apparatus, designing systems so that self-interested agents reveal information and act as intended. The later framework is an application and extension of that mechanism-design machinery.
Carried by Leonid Hurwicz; Eric Maskin; Roger Myerson.
Moral Hazard and Demand Response influenced Moral Hazard Analysis in Health Insurance
The RAND health insurance experiment and the demand-response literature it generated took the moral hazard concept and measured how cost-sharing changes patients' use of care.
Carried by RAND Health Insurance Experiment; Joseph Newhouse.
Moral Hazard and Demand Response influenced Agency and Incentive Theory in Health Economics
The RAND health-insurance experiment and the demand-response literature it produced supplied the empirical moral-hazard evidence that agency and incentive theory in health economics built on. The measured demand response to cost sharing carried directly into the incentive framework.
Carried by Joseph Newhouse; RAND Health Insurance Experiment.
Moral Hazard and Demand Response influenced by Risk Aversion and Insurance Demand Theory in Insurance Theory
Arrow's work on risk aversion and insurance demand supplied the utility framework in which moral hazard and demand response are modelled. The later analysis takes the expected-utility apparatus of insurance demand as its starting point.
Carried by Kenneth Arrow.
Moral Hazard and Demand Response influenced Insurance Design and Demand-Side Cost-Sharing in Pharmaceutical Economics
The RAND Health Insurance Experiment's demand-response findings on coinsurance and utilisation supplied the empirical basis for demand-side cost-sharing design. Later benefit-design work used its estimates of how patients respond to out-of-pocket prices.
Carried by RAND Health Insurance Experiment; Joseph Newhouse.
Supplier-induced demand influenced by Information asymmetry and agency in Healthcare Markets
Information asymmetry and agency theory gave supplier-induced demand its explanatory mechanism: the physician as imperfect agent of the patient. The SID literature explicitly rests on this agency asymmetry.
Carried by Kenneth Arrow.
Bundled payment influenced Value-Based and Bundled Payment in Health Insurance
Bundled payment, developed from the 1980s in episodes-of-care and DRG-style payment, is one of the two components the value-based payment framework combines with outcome-linked payment. The later framework adopted the bundled episode as a payment unit.
Quasi-Experimental Program Evaluation influenced Voucher Experiments and Evaluation in School Choice
Quasi-experimental program evaluation methods such as difference-in-differences and instrumental variables were carried into voucher research, where lotteries and oversubscription provided natural experiments.
Quasi-Experimental Program Evaluation influenced Program Design Turn in Social Insurance And Redistribution
Quasi-experimental programme evaluation in health, using difference-in-differences and instrumental variables from the mid-1980s, provided the identification strategies that the program design turn applied to social insurance and welfare programmes.
Quasi-Experimental Program Evaluation influenced Mixed Methods Evaluation in Health Policy Evaluation
Quasi-experimental programme evaluation contributed the quantitative arm of mixed methods evaluations, which pair such designs with qualitative inquiry. The mixed methods frameworks of the 1990s were built partly to integrate these quasi-experimental estimates with process evidence.
Quasi-Experimental Program Evaluation influenced by Experimental and Quasi-Experimental Design in Public Health Program Evaluation
The experimental and quasi-experimental design tradition in public health programme evaluation, rooted in Campbell and Stanley's work, supplied the design catalogue the later framework uses. The later evaluation practice takes its validity typology and design templates from that source.
Carried by Donald T. Campbell; Julian C. Stanley.

Payment And Incentives is a planet in Nousfarer
Its city has a tower for each of these 9 frameworks. Land, walk in, and climb one concept at a time.
Land on Payment And IncentivesRelated fields
Fields its ideas are linked to
- Health Insurance 7 links
- Healthcare Markets 4 links
- Health Financing 3 links
- Health Economics 2 links
- Health Policy Evaluation 2 links
- School Choice 1 link
- Pharmaceutical Economics 1 link
- Compensation And Rewards 1 link
- Insurance Theory 1 link
- Life And Health Insurance 1 link
- Nonprofit Management 1 link
- Global Health Metrics And Evaluation 1 link
More in Health Services Research
- Comparative Effectiveness Research
- Health Policy Evaluation
- Health Services Research
- Healthcare Delivery Systems
- Implementation And Dissemination Research
See where Payment And Incentives sits on the Map of Human Knowledge