Biotech bubble? Not if you look at sell-side price targets.
Consider the publicly-traded biotechnology company, Science Project Inc (SPI) (a real company, but not its real name or ticker symbol), and a recent quarterly result. The five investment bank research analysts who cover the company each published a research report interpreting the announcement and each reaffirmed twelve-month price targets, which ranged from $8-14 a share, significantly higher than $5.35 a share market price at the time of the announcement.
SPI has a proprietary drug discovery platform, and it has partnerships with five larger pharmaceutical companies; it is also plans to develop drug candidates independently. None of the partnered or internal drug candidates has been tested in humans yet and none will generate sales until 2018, at the earliest (the pharmaceutical alliances will generate “milestone payments” to the company during the development stage.)
The analysts’ near-term price targets are derived by estimating the future cash generation for each product in development, and then calculating what that flow of money would be worth as a lump sum today. This “present-value” calculation adjusts the future revenue estimates (derived by estimating market size and the pace of penetration and taking a best guess at pricing) by guessing a peak sales year, multiplying that year’s sales by 15 to 20x (at the analysts’ discretion), and then discounting this figure by 35-50% (at the analyst’s discretion) to reflect the risk that the predictions will be inaccurate. These present values are then added together for a sum-of-the-parts target price.
When performed on a portfolio of low-risk bonds, this exercise, formally called a discounted cash-flow analysis or DCF, is both precise and accurate. When applied to company earnings, the analysis, while still precise, becomes increasingly inaccurate as inputs into the equation become more variable. DCF is most helpful when analyzing companies with fixed prices, steady earnings and predictable growth. The opposite is true for companies with uncertain pricing, and difficult-to-predict markets. Companies with long development stages and huge optionality (failed oil wells, for example) are the most problematic.
Development-stage biotechnology is arguably the most difficult sector for this type of analysis. Returning to SPI: between now and the predicted arrival of sales revenue in 2018, each of the company’s products faces a series of clinical and regulatory tests. Failure during any of these stages can drop the value of that drug candidate to zero. Given the high failure rate in drug development, SPI’s product pipeline could ultimately be worth nothing.
DCF price targets trivialize the complexity of placing a value on this type of company. The sum of estimates, guesses and discretionary variables that need to be force-fed into otherwise straight-forward math gives the analyst a tool to justify any number he or she wishes, confident that even the most arbitrarily chosen figure can be supported by a seemingly logical process.
As a life sciences portfolio manager, I know many biotechnology equity analysts. I frequently rely on their reports to answer the who?, what?, when? and where? in my own due diligence and they sometimes help me with the why? They work tirelessly, addressing the often-conflicting pulls from their colleagues in banking, their clients on the buy-side, and management teams from the companies they cover. They are superbly trained, many with PhD’s in the sciences or engineering. All are good at what they do and many are outstanding. And they are all smart enough to recognize the futility of building a seemingly precise model on guesstimates and almost random variables.
Price targets on companies like SPI are exercises in precise inaccuracy, diluting the integrity of the value-added main body of the analysts’ work.