Policy analysis excels at modeling first-order effects — the direct, intended consequences of a legislative change. The Congressional Budget Office estimated savings. Industry groups estimated losses. Advocates estimated benefits. Each projection addressed the same narrow question: how much will the Inflation Reduction Act’s drug pricing provisions reduce Medicare spending on the drugs selected for negotiation? The first-order answer is knowable in principle and emerging in practice. The second-order effects — the behavioral responses, strategic adaptations, and market structure changes that radiate outward from the legislation — are where the interesting analysis begins.
The most debated second-order effect concerns R&D investment. The pharmaceutical industry’s core argument against government price negotiation has always been that regulated prices reduce expected returns, which reduces investment in drug development, which reduces the pipeline of future treatments. The argument is economically coherent and empirically uncertain. The relationship between drug pricing policy in a single country and global R&D investment decisions is mediated by so many variables — the global revenue base, tax incentives, capital markets conditions, scientific opportunity, regulatory pathways — that isolating the IRA’s marginal effect is an exercise in assumption-laden counterfactual modeling.
What can be observed, rather than modeled, is how manufacturers are adjusting their strategies in response to the law’s specific provisions. The differential eligibility timelines — small molecules become eligible for Maximum Fair Price negotiation after nine years on market, biologics after thirteen — create a structural incentive to develop biologics rather than small molecules, all else being equal. Four additional years of unregulated pricing represents a significant difference in expected net present value. Whether this incentive is large enough to materially shift portfolio decisions depends on the magnitude of the NPV differential relative to the many other factors that drive development strategy. Early evidence — pipeline disclosures, licensing activity, venture capital allocation — is suggestive but not conclusive.
The launch pricing dynamic is more immediately observable. Manufacturers of drugs approaching market entry must now factor MFP eligibility into their launch pricing calculus. A higher launch price generates more revenue during the unregulated window but may result in a higher AMP, which could paradoxically produce a higher MFP ceiling — since MFP is bounded as a percentage of non-federal AMP. Alternatively, if a manufacturer expects the negotiation to compress revenue below the drug’s break-even point for the Medicare segment, it may choose to maximize commercial revenue through aggressive launch pricing and accept the Medicare ceiling as a constraint on one channel rather than the entire business.
The inflationary rebate provision interacts with these dynamics in ways that compound strategic complexity. Under the IRA, manufacturers of Part D drugs must pay rebates to Medicare if their prices increase faster than inflation. This provision effectively caps annual WAC increases at the CPI rate, at least for the Medicare segment. Manufacturers who previously relied on annual WAC increases to sustain revenue growth — a widespread practice, particularly for drugs with declining volume due to generic or biosimilar competition — must now absorb the volume decline without the offsetting price increase. The effect is a compression of the revenue lifecycle for branded drugs that reduces total lifetime revenue and front-loads the pressure to set a high launch price that does not need subsequent increases to achieve commercial objectives.
The interaction between MFP and the existing 340B program introduces another layer. Drugs selected for MFP negotiation will have a Medicare ceiling price that may be above, below, or approximately equal to their 340B ceiling price, depending on the drug’s AMP and the MFP percentage applied. For covered entities that purchase drugs at 340B prices and dispense them to patients with Medicare coverage, the interaction between MFP reimbursement and 340B acquisition cost will determine margin — and the calculation will differ by drug, by quarter, and by the specific terms of the MFP agreement. The administrative burden of managing these interactions across a portfolio of selected drugs is significant and falls disproportionately on covered entities with limited analytical resources.
Biosimilar market dynamics are affected in unexpected directions. The IRA exempts drugs with generic or biosimilar competition from MFP selection. This creates a scenario in which a reference biologic manufacturer facing imminent MFP selection could benefit from facilitating biosimilar entry — accepting the competitive pressure of a biosimilar in exchange for avoiding the regulatory pressure of a negotiated price ceiling. Whether the biosimilar’s market impact is more or less favorable than the MFP depends on the specifics: the likely MFP percentage, the biosimilar’s expected market share and pricing, and the manufacturer’s ability to retain share through contracting and rebate strategies. The calculus is drug-specific and may produce counterintuitive outcomes.
The geographic effects are worth noting. MFP applies only to the United States Medicare program. For global pharmaceutical companies, the share of revenue derived from U.S. Medicare determines the MFP’s materiality. A drug with eighty percent of its revenue from U.S. Medicare faces a fundamentally different MFP impact than a drug with twenty percent Medicare exposure. Manufacturers with diversified global revenue bases are better insulated. Manufacturers with U.S.-centric commercial profiles — including many rare disease and specialty companies — bear disproportionate exposure.
The political economy of MFP selection introduces its own dynamics. CMS’s selection criteria prioritize high-expenditure drugs without generic or biosimilar competition. The drugs selected for negotiation in each annual cohort send a signal to the market about which therapeutic categories and price points attract regulatory attention. Manufacturers of drugs that narrowly avoid selection in one year must anticipate potential selection in subsequent years and adjust pricing and contracting strategies accordingly. The anticipatory effect — manufacturers moderating prices or adjusting strategies to avoid or prepare for selection — may ultimately have a larger aggregate impact on drug spending than the negotiated prices themselves.
These second-order effects are not defects in the legislation. They are the inevitable consequences of intervening in a complex adaptive system. The pharmaceutical market is not a static machine that can be adjusted with a single policy lever. It is a network of interdependent actors — manufacturers, payers, providers, intermediaries, patients, regulators — each of whom responds to policy changes by adapting their strategies, which in turn changes the environment that other actors face. The IRA mapped the terrain of first-order effects with reasonable precision. The edges of the map — the second-order and third-order effects that emerge from strategic adaptation — are where the territory remains uncharted.
