How not to value biotech.
In this article we focus on analyst’s reports on
public biotech companies that have not yet reached profitability. These
companies often have a market capitalisation below US$ 500 Mio and are therefore
under the radar screen of most analysts. Nevertheless, some analysts accept the
challenge to value unprofitable public biotech companies. Reviewing some of
their reports, we have discovered some original valuation models that merit a
short discussion. We will review in this article the P/E ratio as a valuation
method for nonprofitable biotech companies.
Analysts usually value profitable companies. A good
and fast method to see how expensive a share is is the priceearning ratio (P/E
ratio). If we assume a company with stable earnings, we assume that its value
would be as in the formula below.
_{}[1],
Formula
1: r = discount rate, E = earnings.
If we assume that the market capitalisation of a
company, i.e. its price, is the value the market attributes to this company,
then the P/E ratio corresponds more or less to the inverse of the discount
rate. If the ratio is high, this means that “the market uses a low discount
rate”, i.e. the investment is perceived as rather safe. A low P/E ratio points
at a high discount rate, i.e. a risky investment. If an investor thinks that
the market uses a wrong discount rate, then he can either buy or sell the
company’s shares.
The P/E ratio is
so simple and easy to understand that it is widely popular metric amongst
analysts. Nevertheless, the P/E ratio has two dramatic disadvantages. It
requires the company to be profitable and stable. Biotech companies are mostly
not yet profitable and only at the beginning of their planned life cycle. So,
the P/E ratio is a bad metric to use for biotech companies, you might think.
Have a look at the following valuation of Clinuvel, an
Australian company, performed by an analyst at Louis Capital Markets. Clinuvel
has one product in phase III trials.
Table 1: Sales Analysis

Potential
Peak Sales (US$ Mio) 
Probability 
Probability
Weighted Sales 
Phase III




PLE 
40 
50% 
20 
EPP 
25 
50% 
12.5 




Phase II




SCC/AK 
240 
30% 
72 
SU 
12 
30% 
3.6 
Other 
200 
15% 
30 
TOTAL 
517 

138.1 
Clinuvel develops
the project for various indications. The analyst estimates for each indication
the peak sales and the likelihood that Clinuvel receives marketing approval for
this indication. The overall probability adjusted peak sales are $ 138.1 Mio.
As a next step the analyst calculates the net present value of these
probabilityadjusted sales, assuming that they only occur in 5 years. In order
to get to risk adjusted earnings (=Net in table 2) he must reduce the NPV of
sales by cost of goods, sales, general and administrative expenses, and taxes.
Table 2: Calculation

Parameter 
rNPV 
Sales 
At
15% discount 
68.7 
COGS 
15% 
10.3 
SG&A 
~25% 
17.7 
EBIT 

46.6[2] 
Tax 
35% 
16.3 
Net 

30.3 
The final $ 30.3
Mio represent the risk adjusted net present value of Clinuvel’s earnings in 5
years. On a per share basis this corresponds to (risk adjusted and discounted)
earnings per share (EPS) of $ 0.1. At a share price of A$ 1.07 this corresponds
to a P/E ratio of 9.1 (including the exchange rate A$/US$). The analyst then
compares this to the Industry P/E (BBG World Biotech Index) of 45.05 and
concludes that the share is a bargain.
This method, the socalled dynamic P/E ratio, is much wider spread than one would believe[3]. Some analysts do not even use riskadjusted sales but simply use a much higher discount rate (35%) and factor in the attrition risk this way. But then they all use an industry P/E ratio. These industry P/E ratios differ quite dramatically.
Another analyst from RRS, also valuing Clinuvel, uses a 25 P/E ratio. He mentions, that profitable biotech companies trade at 25 to 40 times their earnings. Other analysts use a multiple of 30.
Dynamic P/E ratio
1. Determine the representative year regarding revenues: T.
2. Assess revenues
for that year: R_{T}
3. Discount and
probabilityadjust earnings: R_{t}=p*R_{T}*(1+r)^{(Tt)}
4. Calculate EBIT: EBIT=R_{t}*(1COGSM&S)
5. Calculate
earnings: E_{t}=EBIT*(1tax)
6. Multiply with selected P/E ratio: P=E_{t}*P/E
P/E ratio in more detailEstablished drug development companies typically exhibit a P/E ratio of about 1015.
Table 3: P/E ratios and
corresponding discount rates
Company 
P/E ratio 
Discount rates 
Amgen 
9.73 
10.3% 
BMS 
14.28 
7.0% 
Lilly 
13.29 
7.5% 
Merck 
11.96 
8.4% 
Pfizer 
9.29 
10.8% 
We can assume that these companies have reached a stable stage of their business cycle. Therefore we can apply the P/E idea and deduct their discount rates, with some caution. A 30x multiple (read price = 30 x earnings) would therefore correspond to a discount rate of 3.3%. Of course, this does not correspond at all to biotech companies that usually exhibit much higher discount rates than pharma companies.
Figure 1: P/E ratio along the
business cycle of a company
Figure 1 exhibits first, that unprofitable companies do not have any P/E ratio at all (due to lack of earnings). Second, companies that are only at the beginning of profitability have naturally a high P/E ratio, because the price is calculated using the future and higher earnings, but are divided by the current and lower earnings. A high P/E ratio is therefore not only a sign of an established and diversified business as is the case for pharma companies, but also for a business that is still in growth.
CritiqueThe described riskadjusted discounted P/E method has
some serious flaws. First, it does not account for all the R&D expenses it
takes to get to a profitable stage. Second, it is completely unclear that the
selected year represents already a stable state of business. Calculating with
the expected P/E ratio in 5 years corresponds to taking the terminal value
already after 5 years. This is quite a short period of time in an industry
where it takes 15 years to commercialise a product that should then stay on the
market for another 10 years. Third, this method is unsuited to correctly
account for the tax value of the company (cf. Avance Newsletter May 2008).
Fourth and most important of all, it makes no sense to compare with an industry
P/E ratio. The P/E ratio depends on the stage of the company and on the
discount rate. Both are particular to the company and the selected year.
We can only speculate why analysts use this method.
One advantage is that the method is relatively insensitive to the discount
rate, one of the most critical parameters. On the other hand the method
supposes that the company is comparable to the industry average and uses an at
best questionable P/E ratio. It is careless to hope that all the flaws of the
method will compensate each other.
With riskadjusted net present value (rNPV) analysts
have a method at hand that accounts for all particularities of the company.
Fortunately, most analysts make use of rNPV, although the described dynamic P/E
method is widely spread.
Dr. Christa Bähr, Dr. Markus Manns: „Life Science am
Kapitalmarkt, Biotechnologie im Fokus“, publication of DVFA, 2005.
www.clinuvel.com/resources/pdf/analyst_reports/
LouisCapitalUpdate12008.pdf
www.clinuvel.com/resources/pdf/analyst_reports/20071120RSS.pdf
[1] If a growth
factor is factored in the equation becomes _{}.
[2]It is unclear why this is not 40.7.
[3] cf reference
Bähr/Manns