Biotech Valuation: How to Make “Pre-Revenue” Cool Again

Biotech Valuation Definition: In a biotech valuation, you project the revenue, expenses, and cash flows for an early-stage startup based on “far in the future” forecasts over 20 – 30 years, assume that sales grow to a “peak” and then decline to ~$0, leaving no Terminal Value, and you probability-weight the cash flows or vary the Discount Rate over time to reflect the risk; biotech valuation for “platform companies” is closer to the standard DCF approach with a Terminal Value.

Early-stage biotech investing has long been known as “speculative,” but in the current market environment, with meme stocks, unprofitable AI companies, and rocket-ship firms launching into orbit based on vibes, the valuation of biotech firms is far more grounded.

The key difference vs. the approach used for tech startups is that many of the parameters are known in advance:

• If you know the company’s drugs, you know about how many potential patients there are.
• There have probably been similar drugs treating similar patient populations, so you can estimate the pricing and market share information from those.
• And you can even retrieve data for the associated R&D and operating expenses based on companies that have produced similar drugs.

By contrast, almost none of this is known in advance for the average tech startup.

Who knows how many users Starlink, Uber, or Snapchat can capture? You might be able to guesstimate the Total Addressable Market (TAM), but the market share and pricing are far more difficult.

This explains why biotech valuation is largely based on the far-in-the-future DCF, with valuation multiples based on “several years forward revenue” for additional support.

Files & Resources:

• Simplified Biotech Valuation/DCF for Antios Therapeutics (XL)
• Biotech Valuation – Presentation Slides (PDF)

Biotech Valuation: The Basic Sketch of a DCF for a Pre-Revenue Company

The standard approach is to assume that a drug gets released after a series of clinical trials to test its safety and efficacy, its sales grow to a peak level (“peak sales”), they start to decline, and then “generic” competitors enter the market, which reduces the company’s sales by 80 – 90% until they eventually reach $0.

This assumes that it’s a startup developing a new, patent-protected drug.

Since the government regulators grant the company a time-limited monopoly, it can sell its drug at high prices for a period. Government-run healthcare plans or insurance companies pay for it rather than individuals (in most cases).

Eventually, the exclusivity period ends, and “generic” competitors that make biosimilar versions of the drug at much lower prices enter the market.

You start the forecast by researching the number of people with the disease or medical condition the company’s drugs are treating, forecasting how many of them are “diagnosed,” and estimating what the company’s eventual market share will be.

For example, for this company (Antios Therapeutics) that was working on a drug for chronic hepatitis B, we assumed there were roughly 250 – 300 million adults worldwide with the infection, that 10% were diagnosed, and that the company’s eventual market share would be just under 1%:

You can easily find estimates for the potential patient count online or in the company’s presentations.

The “% diagnosed” assumption is quite low (10%) because many of these adults are in places like Africa, where the level of medical care and overall access are much lower.

In big, broad markets like this one, the final market share at peak sales tends to be quite low (under 10%, and often below 1%) because it’s very difficult to penetrate a market with tens of millions of potential patients.

We could look at previously released hepatitis drugs and assess their market penetration numbers, plus the company’s “cumulative revenue” estimates, to come up with a guesstimate here:

We assume that it takes about 6 years to reach “peak market penetration” because that’s about average for biotech; typical numbers range from 4 to 8 years, depending on the drug.

For example, Entyvio for Crohn’s disease, which had some technical/market similarities, took about 5 – 6 years to reach its peak sales.

The exact timing is adjustable in the assumptions at the top of the Excel file if you disagree.

The initial pricing of $6,000 per patient per year here is based on what similar, patent-protected HBV drugs have sold at historically.

There are modest price increases each year until generics enter the market in Year 15 of the model, or Year 11 of the commercial period:

Revenue equals the Total Patient Count * Average Price per Patient per Year, and we label this “Unadjusted” to indicate that it has not yet been probability-weighted or otherwise risk-adjusted.

The expense and cash flow numbers are based on the company’s estimates during the development period; in the commercial period, they’re linked to comparable public companies with similar drugs.

After a drug finishes its development and gets released, operating margins are normally very high, such as in the 50 – 70%+ range.

There may be CapEx and Working Capital requirements in this commercial period, but they tend to be modest for single-product biotech companies without in-house manufacturing:

All biotech startups lose money in their early years and accumulate Net Operating Loss (NOL) balances, which they can use to reduce their Taxable Income in later years.

This follows our standard NOL setup, with Losses increasing the balance and positive EBIT potentially being offset by the balance, up to the 80% allowance here:

The Unlevered Free Cash Flow (UFCF) equals NOPAT + Depreciation & Amortization +/- Change in Working Capital – CapEx.

NOPAT is defined as EBIT minus Cash Taxes, so we can avoid an additional line for the Deferred Taxes and skip the Book vs. Cash Tax difference.

At this point in a full model, we might select comparable public companies that have similar drugs and calculate a range of WACC figures manually.

For example, we might calculate WACC for “risky, pre-revenue” companies and WACC for “mature, revenue-generating, cash flow-positive” companies and make our company move closer to the latter figures over time.

But we could also use one Discount Rate and then probability-adjust all the UFCF figures.

Broadly speaking, you must pick one of these approaches and stick with it in the analysis:

1. Vary the Discount Rate over time (higher to lower) as the drug moves through clinical trials; OR
2. Use a constant Discount Rate but probability-adjust the UFCF, based on increasingly lower percentages through the stages.

We use the second approach here.

The “probability weightings” are based on the chances of similar drugs advancing from Phase II to Phase III clinical trials, Phase III to regulatory review, and regulatory review to commercial launch.

So, we multiply all the Phase III UFCF numbers by ~38% (Phase II to Phase III probability), and we multiply all the Commercial UFCF numbers by 38% * 64% * 93% = 23%:

There is no Terminal Value because the probability-adjusted UFCF is almost $0 by the end of the 20-year forecast period. So, we simply use the NPV function to discount and sum up all the probability-adjusted UFCFs to determine the Implied Enterprise Value.

Then, we add/subtract the standard “bridge” items (only Cash and Debt since the NOLs are used up in the forecasts) to determine the Implied Equity Value.

We can then sensitize these results and calculate the Implied Equity Value under different scenarios, such as different Discount Rates, margins, market shares, and pricing assumptions:

This is the approach for a simple biotech DCF, but in real life, it always gets more complicated (see below).

Biotech Valuation Multiples

You can also use public comps and precedent transactions to value biotech firms, but the mechanics and multiples differ.

Typically, you use multi-year forward revenue multiples, based on metrics such as “L + 5 Revenue” (i.e., Revenue 5 Years into the Future After the Launch Date) or “T + 4 Revenue” (i.e., Revenue 4 Years into the Future After the Transaction Date).

It’s also common to see multiples based on “Peak Sales,” as this metric is often used to benchmark early-stage biotech firms.

You can see a few examples of these types of multiples used in real-life presentations and Fairness Opinions issued by banks below:

The same approach is used in both public comps and precedent transactions, but with the transactions, everything is based on the deal announcement date.

One issue is that it’s not always clear whether the revenue projections are unadjusted or probability-weighted, making it tricky to rely on external data sources for this.

For the screening criteria, the main issue is that you should not mix and match biotech firms at different stages.

For example, you can’t use L + 5 Revenue multiples from both pre-revenue biotech firms and ones that are already in the commercial stage.

Pick one set and use L + 5 Revenue for pre-revenue firms, but more conventional 1-year forward metrics for those that already generate revenue.

What Makes Biotech Valuation Hard: Sum-of-the-Parts and Platform Companies

Based on everything above, biotech valuation might seem simple.

Sure, it takes some time and effort to find the data, but the concepts behinds the forecasts are not that difficult.

In real life, it becomes more difficult for two main reasons:

• Many companies have multiple products at different development stages, which means the DCF setup is more involved.
• Many firms operate as “platform companies,” so they always have a mix of existing, revenue-generating products and new ones in development.

With the first issue, the usual approach is to set up a separate DCF for each drug so you can reflect different success probabilities, revenue projections, and expense profiles, and then aggregate everything at the end:

With the second issue, you’ll often split a company’s drugs into “new/experimental” vs. “mature” and set up separate forecasts for the revenue.

You still probability-adjust the new drugs currently under development, but you do not apply any adjustments to existing ones that are in the commercial stage:

Picking the Terminal Value assumptions is tricky for this type of company, but one simple solution is to assume a modest negative long-term growth rate.

The logic is that the company’s revenue will still decline over the long term, but it won’t necessarily “drop to $0” as it would for a company with a single drug.

Biotech Valuation Examples from Banks

If you want to learn even more about biotech valuation, check out these bank-issued presentations and Fairness Opinions:

• Evercore – Biogen / Apellis (Search for “Certain Financial Projections”)
• Moore Financial Consulting – MIRA Pharmaceuticals / SKNY Pharmaceuticals (Search for “Schedule A to Agreement and Plan of Merger”)
• M3 Partners – Serranto (Detailed Business Plan Out of Chapter 11 Bankruptcy)
• Lazard – Project Salamander (Example Public Comps and Precedent Transactions)
• Centerview – Fairness Opinion for Chinook Therapeutics (No comparable companies used; only a DCF)
• Cowen – Project Avalanche (Example of TEV / Peak Sales Multiple in the Precedent Transactions)

If you want even more, search for terms like “peak sales” and “probability-adjusted” on the sec.gov site and see what turns up.

Do it right, and you might even understand the power of “pre-revenue” more than Russ Hanneman.