The economics of a drug: R&D cost, risk, and pricing

The stages, and the jargon, matter here:

• Preclinical: lab and animal testing, before humans.
• Phase 1: small trials, mainly checking safety.
• Phase 2: does the drug actually work, and at what dose?
• Phase 3: large trials proving efficacy and safety at scale. The most expensive stage.
• Regulatory review: submission to an agency like the FDA (US Food and Drug Administration) or the EMA (European Medicines Agency) for approval.

Each stage is a gate. Money spent at Phase 1 buys you the *option* to spend more at Phase 2, but only if the data justify it. This is why pharma finance borrows heavily from options thinking: you pay a small amount now to keep the right (not the obligation) to invest more later.

The probability of a candidate surviving all the way from Phase 1 to approval is low. Estimates vary by disease area and are genuinely uncertain, but the widely cited figure is roughly in the low double digits as a percentage. Oncology (cancer) tends to be lower; some other areas higher. The BIO industry association publishes recurring analyses of these clinical success rates, available here.

Why cost of capital drives everything

Here is the part non-finance people underestimate: time and risk are expensive.

A drug can take well over a decade from discovery to market. Every dollar spent in year one on a program that pays off (maybe) in year twelve has to compete against every other use of that dollar, including simply returning cash to shareholders.

That is the cost of capital: the minimum return investors demand given the risk. Pharma R&D is risky, illiquid, and long-dated, so its cost of capital is high. A high discount rate punishes distant cash flows severely.

Two consequences follow directly:

1. Distant payoffs get heavily discounted. A billion dollars of revenue twelve years out is worth a fraction of that today.

2. Speed has enormous financial value. Shaving even a year off development timelines, or reaching the market faster than a competitor, materially changes the math.

This is why companies pay up for anything that de-risks or accelerates a program: better trial design, biomarkers to select the right patients, or acquiring an asset that is already through Phase 2.

Risk-adjusted NPVNPVNet Present Value is the sum of an investment's future cash flows discounted to today, minus the initial outlay. A positive NPV signals value creation.Voir la définition complète →: the decision tool

The workhorse metric is risk-adjusted net present value (rNPV), sometimes called eNPV (expected NPVNPVNet Present Value is the sum of an investment's future cash flows discounted to today, minus the initial outlay. A positive NPV signals value creation.Voir la définition complète →).

Plain-English version:

> Take the future cash flows a drug might generate. Multiply them by the probability the drug actually reaches the market. Then discount everything back to today using the cost of capital. Subtract the costs.

In shorthand:

Two levers dominate the output:

• Probability of success. A program that clears Phase 2 with strong data may see its probability of eventual approval jump sharply. That single event can swing rNPV from negative to strongly positive, which is why good Phase 2 data can trigger huge valuation moves and acquisition interest.
• Peak sales and duration. How large is the eligible patient population, what price can the drug command, and how long before patents expire and generics arrive?

That last point brings us to the revenue side.

The patent cliff and the exclusivity clock

A drug's economics live and die by exclusivity.

Patents and regulatory exclusivity give a limited window of protection. When they expire, generics (chemically identical copies) or biosimilars (highly similar versions of biologic drugs) enter, and prices can collapse. This sudden revenue drop is the patent cliff.

So the clock is always ticking. Much of the patent life is consumed *before* the drug ever launches, eaten up by the long development timeline. The commercial window to earn a return can be surprisingly short. This is a major reason companies push hard for speed and for premium pricing while exclusivity lasts.

Pricing and reimbursement: where the return is really decided

You can invent a brilliant medicine and still earn a poor return if you cannot get paid for it.

In most markets the patient is not the payer. The real customer is the payer: an insurer, a government health system, or a pharmacy benefit manager. Getting a drug reimbursed (covered so patients can access it affordably) is a separate battle from getting it approved.

Different systems, different logic:

• United States: fragmented, negotiated pricing. Insurers and PBMs (pharmacy benefit managers, middlemen who negotiate on behalf of health plans) drive hard bargains and control formulary placement (whether and how a drug is covered). Recent policy, including Medicare's new authority to negotiate prices on certain high-spend drugs, is reshaping expected returns for affected products.
• Europe and many other markets: government bodies assess value before agreeing to pay. In England, for example, NICE (the National Institute for Health and Care Excellence) evaluates whether a drug's benefit justifies its cost, often using cost-effectiveness thresholds. You can browse NICE guidance here.

The concept underneath both is health technology assessment (HTA): does this drug deliver enough clinical benefit to justify its price versus existing options? If a new drug is only marginally better than a cheap generic, payers will resist a premium price, no matter how much the company spent developing it.

Sunk R&D cost does not justify price. Payers price to *value delivered*, not to *cost incurred*. This is one of the most misunderstood points in the whole sector. The billion dollars you spent is irrelevant to a payer deciding what an extra few months of survival is worth.

Putting it together: how a program lives or dies

Trace one candidate through the finance lens:

1. Early stage. Low probability of success, large future costs still ahead. rNPV is fragile and highly sensitive to assumptions. The company funds it as one small bet among many.

2. Positive Phase 2. Probability of success jumps. rNPV can turn sharply positive. This is often when partnerships and acquisitions happen, because the asset just got dramatically de-risked.

3. Phase 3 and approval. Big spend, then the exclusivity clock starts ticking in earnest.

4. Launch and reimbursement. The real test. Can the company secure coverage at a price that reflects the drug's value before generics arrive?

At each step, finance is asking the same question: given updated probabilities, costs, and the expected price we can actually get paid, is the risk-adjusted return still above our cost of capital? If not, the rational move is to kill the program and redeploy the capital, even after spending heavily. Sunk costs are sunk.

That discipline (killing losers fast, doubling down on winners) is what makes the portfolio math work.

Key Takeaways

• Pharma R&D is a portfolio of high-failure, high-payoff bets. Most candidates fail; a few winners must fund everything, including the failures.
• Cost of capital and long timelines make speed and de-risking extremely valuable. Distant cash flows are heavily discounted, so accelerating development directly raises value.
• Risk-adjusted NPV is the core decision tool. It weights future cash flows by probability of success and discounts them. Positive Phase 2 data is often the pivotal value inflection point.
• Exclusivity is finite, and the patent cliff is brutal. Much of the commercial return must be earned in a short window before generics or biosimilars arrive.
• Payers, not patients, decide the return. Reimbursement and health technology assessment price to value delivered, not to R&D spent. This lesson is educational and not investment advice.