The price of a prescription drug is not set by the FDA, not by an insurer, and not by any publicly accountable body. It is set by the manufacturer at launch, based on internal calculations of market potential, anticipated competition, and the price the market will bear. To understand why prices are set as they are, and to evaluate the competing claims made about the relationship between development costs and pricing, it helps to understand how drugs move from scientific hypothesis to pharmacy shelf.
The Development Pipeline
Drug development follows a roughly sequential path, though the stages overlap and the timeline varies considerably by therapeutic area and drug type.
Discovery and target identification: Researchers identify biological targets — proteins, enzymes, or genes — implicated in a disease process, then search for compounds that interact with those targets. This process, which once relied on chemical synthesis and screening of compound libraries, now increasingly incorporates computational modeling and genomic data. Thousands of candidate compounds may be evaluated before a small number emerge as worth pursuing further. This stage typically takes three to six years.
Preclinical research: Promising candidates are tested in laboratory and animal models to assess safety, toxicity, efficacy, and pharmacokinetics — how the compound is absorbed, distributed, metabolized, and eliminated by the body. Preclinical testing must comply with FDA’s Good Laboratory Practice standards and produces the data required for an Investigational New Drug (IND) application. The IND is filed with the FDA to request permission to begin human testing. Preclinical research typically takes one to three years.
Clinical trials: Human testing proceeds in three standard phases. Phase I trials, involving 20 to 100 participants (typically healthy volunteers or, for serious diseases, patients), focus primarily on safety and dosing. Phase II trials, involving up to several hundred patients with the disease or condition, assess early evidence of efficacy and refine dosing. Phase III trials, typically involving hundreds to thousands of patients across multiple clinical sites, are the pivotal studies required for approval — they generate the large-scale evidence of safety and efficacy that supports an NDA or BLA submission. Clinical development as a whole typically requires six to seven years, though timelines vary widely.
FDA review: After completing Phase III, the manufacturer submits a New Drug Application (NDA) for small-molecule drugs or a Biologics License Application (BLA) for biologics. The FDA has a standard review timeline of 10 months for most applications after acceptance for filing; priority review — available for drugs treating serious conditions with unmet medical needs — runs six months. FDA guidance provides that the agency must determine within 60 days of submission whether to accept the application for filing.
Post-market surveillance: Approval is not the end of monitoring. Phase IV studies and ongoing pharmacovigilance are required to detect rare adverse events that Phase III trials, even with thousands of participants, may not have captured.
The Attrition Problem
Clinical development is a high-failure process. Industry data compiled by Norstella shows that the average likelihood of approval for a new Phase I drug is approximately 6.7 percent — meaning that for every 100 compounds entering Phase I, roughly 7 eventually receive FDA approval. Phase I to Phase II transition succeeds roughly 47 percent of the time; Phase II to Phase III only 28 percent; Phase III to approval about 55 percent.
The high failure rate has two implications for cost accounting. First, the cost of failed programs must be included in any estimate of what it costs to produce a single approved drug — a company bears the expense of every failed compound along the path to each success. Second, the high-attrition framework means that clinical development expenditures are heavily front-loaded in risk and back-loaded in return.
The $2.6 Billion Estimate: What It Claims and How It Is Calculated
The most widely cited figure for pharmaceutical R&D costs is the estimate produced by the Tufts Center for the Study of Drug Development (CSDD). Published in 2014, the Tufts estimate placed the average cost of developing a new approved drug at $2.588 billion — a figure that the Médecins Sans Frontières Access Campaign and other commentators noted was 2.5 times the prior Tufts estimate of $802 million published in 2003.
The $2.6 billion figure includes two components: approximately $1.4 billion in out-of-pocket costs (direct research expenditures) and approximately $1.2 billion in “time costs” — representing the opportunity cost of capital tied up during the development period, calculated as the returns that could have been earned if those funds had been invested elsewhere during the years of development. The time cost component is a theoretical figure, not an expenditure.
Several methodological critiques have been raised about the Tufts estimate.
As Forbes noted at the time of the estimate’s release, five specific criticisms recurred: the study relies on proprietary, non-peer-reviewable data provided by pharmaceutical companies; the sample of 106 drugs was not representative of the full range of drug development; the capital cost component includes theoretical returns rather than actual expenditures; R&D expenses are partially tax-deductible, reducing the after-tax cash outlay; and the Tufts Center receives funding from pharmaceutical industry sources.
Public Citizen has argued at length that after accounting for the tax deductibility of R&D costs and the exclusion of publicly supported drugs from the Tufts sample, the actual after-tax cash outlay for the average new drug is substantially lower — its earlier analyses placed the figure at roughly $110 million (in year 2000 dollars). The pharmaceutical industry disputes these alternative calculations.
The debate is not resolvable from public data, because the underlying cost data are proprietary and not independently verifiable. What is agreed upon is that the $2.6 billion figure includes substantial opportunity-cost components that represent theoretical foregone returns, not cash expenditures, and that the sample was limited to drugs developed entirely without government support — which excludes most commercially significant drugs.
The Role of Publicly Funded Research
A separate and partly parallel debate concerns the contribution of public investment — primarily through the NIH — to the drugs that ultimately reach the market.
A 2018 study published in PNAS found that NIH funding contributed to published research associated with every one of the 210 new drugs approved by the FDA from 2010 to 2016, collectively involving more than 200,000 project years of grant funding totaling more than $100 billion. The researchers noted that more than 90 percent of this funding involved basic research on the biological targets of the drugs, rather than on the drug molecules themselves. NIH-funded research on drug targets and mechanisms provides the scientific foundation on which manufacturers build specific drug candidates.
A 2023 JAMA Health Forum study examined NIH funding for phased clinical trials specifically — the development research that most directly parallels industry activity — and found that NIH funding contributed to phased clinical trials involving 62 percent of the 387 drugs approved from 2010 to 2019. Total NIH funding for those clinical trials was $8.1 billion, representing approximately 10 percent of reported industry spending on clinical development. NIH funding was most concentrated in Phase 1 and Phase 2 trials (covering about 25 percent and 22 percent of those costs, respectively) and much smaller in Phase 3 (about 4 percent).
Taken together, these findings suggest that public funding plays a substantial role in the basic science underlying new drugs and a more limited but still significant role in early-phase clinical development. Industry funding dominates Phase 3 trials, the most expensive stage, and most of the regulatory submission and commercialization costs.
Public Citizen has noted that the 10 drugs in the first round of Medicare price negotiations collectively benefited from $12 billion in NIH funding for foundational research on drug targets and applied research prior to FDA approval.
The Relationship Between R&D Costs and Pricing Decisions
Industry representatives consistently link R&D costs to the justification for high drug prices: high prices are necessary to recoup development costs and fund future research. Academic analysts have questioned this link on two grounds.
First, the relationship between a specific drug’s development cost and its price is not documented publicly and may not reflect actual development economics. Drug prices are set based on market conditions, anticipated patient volume, competing products, insurance coverage practices, and willingness-to-pay assessments — not as a direct markup over identifiable costs. A drug that treats a large patient population can generate substantial returns at a moderate price; a drug for a rare condition may be priced to recover costs from a small number of patients. The pricing logic differs substantially across these scenarios.
Second, a substantial fraction of research and development expenditures by major pharmaceutical companies — as reported in their own financial statements — is allocated to developing variants of existing drugs (new formulations, dosages, delivery mechanisms) and marketing, rather than to genuinely novel therapeutic mechanisms. The aggregate industry R&D spending figures cited in policy debates do not disaggregate between these categories.
None of this resolves the underlying policy questions about what level of R&D investment is socially optimal, how it should be funded, or what pricing systems best allocate costs and returns. What the evidence does establish is that the $2.6 billion figure used in industry arguments is a contested methodological construct, not an audited accounting of typical drug development expenditures, and that public investment plays a substantial role in generating the scientific foundation on which commercial drug development proceeds.
This article was researched and drafted with AI assistance under human review. See our full AI and editorial practices.