Skip to content

Grant or Subsidy Program

Outcome-conditioned funding policy — instantiates Public Goods Provision

Funds a good that private incentives would underprovide by paying an external provider against defined outcomes, then measuring whether the public value actually appeared.

A Grant or Subsidy Program provides a shared good by paying someone else to produce it — routing money from a funder (a government, foundation, or agency) to independent providers whose private incentives alone would leave the good underproduced. Its defining feature is the conditional, outcome-oriented transfer: the funder does not use or govern the good, so it substitutes measurement for control, tying disbursement to defined targets and demanding evidence that public value was created. A subsidy lowers the cost of a desirable activity; a grant funds a specific deliverable — but both share the same core problem this mechanism solves, which is how a distant funder gets a good produced by parties it doesn't run and verifies it was worth the money.

Example

A state wants broadband in sparsely-settled rural counties that no carrier will wire because the per-household cost never pays back. It stands up a subsidy program: private and cooperative internet providers can apply for funding to build in eligible areas, awarded competitively. Crucially, the money is tied to outcomes, not intentions — providers are paid in tranches as they hit verified milestones (households actually passed, minimum speeds actually delivered), and awardees must report coverage and performance data the agency audits. A provider that wires 1,200 homes to the promised standard gets its subsidy; one that builds slower or thinner gets a proportionally smaller draw. The state never operates a network; it buys a measured public benefit from those who can build it.

How it works

The mechanism substitutes conditions and evidence for direct control:

  • Ring-fenced fund with eligibility. A defined pot of money and rules for who and what qualifies channel funding to the underprovided good and away from what the market already supplies.
  • Outcome-conditioned disbursement. Payment is tied to defined targets — milestones, coverage, deliverables — so money follows results, not promises.
  • Measurement of realized benefit. The program instruments whether the intended public value actually appeared, since the funder can't see it by using the good itself.[1]
  • Reporting and audit. Recipients report against terms and the program verifies, closing the loop between money out and value produced.

Tuning parameters

  • Input vs. outcome conditioning — paying for activity (inputs) is easy to administer but funds effort not results; paying for outcomes aligns incentives but pushes measurement cost and risk onto providers.
  • Competitive vs. formula allocation — competition drives quality and value for money but favors sophisticated applicants; formula funding is fairer and simpler but rewards eligibility over performance.
  • Match requirement — requiring recipients to co-fund screens for commitment and stretches the pool, but excludes worthy providers who can't put up a match.
  • Measurement burden — rigorous outcome verification protects against waste but adds cost and can crowd out small providers; light-touch reporting is cheap but invites drift.
  • Funding horizon — multi-year commitments let recipients build durable capacity; short cycles keep control but breed dependency and stop-start provision.

When it helps, and when it misleads

It is the natural tool when the good is best produced by independent parties the funder can't or shouldn't run — research, public-interest infrastructure, environmental or open goods — and when the funder is willing to define and measure what success looks like. Its central risks are additionality (paying for what would have happened anyway), capture by skilled grant-seekers whose real product is the proposal, and dependency on soft, short-term funding that vanishes and strands the good. The classic misuse is measuring outputs that are easy to count instead of the outcomes that matter, so a program reports success while the public value quietly fails to materialize. The discipline is to condition money on realized benefit, test for additionality, and pair the grant with a plan for what sustains the good when the program ends.

How it implements the components

  • funding_pool — the program's ring-fenced budget is the source of provision capacity, disbursed to external providers.
  • provision_level_target — eligibility and milestone criteria encode what counts as enough (coverage, standard, deliverable) to earn funding.
  • benefit_measurement_model — its signature: because the funder doesn't use the good, it must model and measure whether the intended public value actually appeared.
  • accountability_and_reporting — recipient reporting and audit close the loop between money spent and value delivered.

It does not raise the money it distributes — pooling contributions from the public is Public Funding or Taxation's job (contribution_rule) — and it does not govern the good's ongoing operation, which for shared-authority cases is Public–Private or Multi-Stakeholder Partnership's (provision_responsibility).

References

[1] Additionality asks whether a funded outcome would have occurred without the funding. A grant that pays for activity that would have happened anyway produces no additional public value, so credible programs test additionality rather than crediting themselves with every outcome they observe. ↩