The Risk Spectrum

The simplest way to organize the zoo of value-based models is by one question: how much financial risk does the provider hold? Everything from quality bonuses to full capitation is a waypoint on that spectrum.

The spectrum at a glance

Model	Provider risk	Core mechanic
Pay-for-performance	None	Bonuses on fee-for-service for quality targets
Shared savings (one-sided)	Upside only	Keep a share of savings vs. benchmark; losses forgiven
Shared savings (two-sided)	Both directions	Share savings and owe a share of overruns
Bundled payments	Episode-level	One target price per episode of care
Capitation / global budgets	Full	Fixed payment per member or per institution

What changes as risk deepens

• Pay-for-performance is easy to join and administer, but when 95 percent of revenue still comes from volume, a 5 percent bonus rarely changes how an organization behaves.
• One-sided shared savings is the standard on-ramp: bill fee-for-service as usual, keep roughly half of any savings against a spending benchmark if quality holds. But upside-only is a lottery ticket, not a stake.
• Two-sided risk is where models start to bite. Facing losses, organizations invest in care management, post-acute review, and analytics that find avoidable spending before year end. Protections like stop-loss caps and risk corridors keep small organizations from being ruined by bad luck.
• Bundles concentrate risk on an episode a provider directly controls (Lesson 3).
• Capitation and global budgets remove fee-for-service entirely: the provider becomes the insurer of its own efficiency (Lesson 4).

Three patterns to carry forward

1. Incentive strength rises with risk. More revenue at stake means more actual change.
2. So does the need for guardrails. Quality floors, risk adjustment, and stop-loss all scale up alongside accountability.
3. The spectrum is a journey. Successful risk-bearing organizations climbed it over years, building capabilities at each step.

Worth remembering: weak incentives produce weak results. The evaluation literature is consistent that meaningful effects appear once providers face downside risk.

Key takeaways

• Classify any model by asking how much risk the provider holds.
• Greater risk requires greater protection: adjustment, stop-loss, quality gates.
• Organizations move along the spectrum gradually, not in one leap.