Do's and don'ts of valuing a biotech company

Do:

Figure out the cash burn of the company

• Cash burn, as in the amount of cash left if the company's cash flows dry up (or sometimes calculated as the # of months left of operations that could be financed by the company before it has to shut down), is crucial to understanding its financial health. If a company has a high cash burn rate, that's a sign it is using its cash reserves quickly, which could indicate financial instability.
• Cash burn is so important because biotech is more a game of "how long can you last funding your R&D into your pipeline" than one of "how can I make as much shareholder value as quickly as possible". Biotechs, especially small caps lacking revenue, rely off of the mercy of milestone payments from pharma companies they collaborate with and, ahem... public speculators in the stock market.
• To calculate the cash burn rate for a biotech company, you would typically follow these steps:
• Identify the company's cash balance at the start and end of a given period. You can find this information in the company's balance sheet, part of its quarterly or annual financial reports.
• Subtract the ending balance from the starting balance to determine the net change in cash.
• If the company is burning cash, the net change in cash will be a negative number, indicating the amount of cash spent over the period.
• Divide this net change by the number of months in the period to calculate the monthly cash burn rate.
• Note: measuring cash burn is only relevant for companies that are not generating positive cash flows from their operations. If a company generates positive cash flow, then it might be using cash for capital investments or other non-operational expenses.

Find out what drugs they have in their pipeline

⇛ If they already have any approved drugs:

• Calculate the sales revenues (can find through annual report of the company) & estimate patient population size. Make your best estimates to see if these figures will grow or shrink in the next decade– more on this in the next bullet point.
• Are those approved drugs in competitive markets?
• Look at competitor drug pricing, sales, etc. Best data would be this given drug's share of the market: this info is likely in a company's annual report. 
• See if any promising disruptive competitor drugs are soon to be approved / have been recently approved

 ⇛ If they only have unapproved drugs: 

(Note: a lot of these points still apply to companies with feeble revenue streams from their approved drugs and companies with lots of unapproved drugs in their pipeline that eat up a lot of R&D. Simply put, companies with weak positive cash flows or negative cash flow)

• How far are the drugs through the FDA process? Preclinical? Phase II? NDA? The further a drug is in this process, the more likely it is to be successful. Drug development tends to be lindy ;)

• Look at your company's website. They will usually have news about the progress of each drug in their pipeline.

• Eyeball the clinical trial data of the drug on clinicaltrials.gov. Rarely gives valuable info (like... a lot of the time they don't even share the trial data), but it can't hurt do take a look. 

• You can use the takeaways from this report to get a crude idea of whether the drug class (as in if it's first in class, second in class, etc), indication, or form (small molecule? biologic? gene therapy?) make it more or less likely to succeed in further stages of clinical trial. 

• How disruptive is their drug, potentially? I haven't done this myself too much, admittedly. But my intuition would be to start with doing some gonzo research (lol) and looking at reviews of other drugs taken by patient population for the given indication. You could do this by looking at drug reviews for a given indication on websites like drugs.com, or user experiences on Reddit / Twitter.
• Some indications like oncology and rare diseases have a bit of an advantage here. Sometimes the FDA also will grant a special designation if a drug is promising enough. There are a lot of different, distinct designations: breakthrough therapy, fast track, priority review, orphan drug, accelerated approval, RMAT. Some of these are more impressive than others– for example, breakthrough therapy drugs show significant promise over available therapies, while priority review drugs are pushed through quicker because biotechs can pay the FDA to expedite the process.
• Drugs with a new mechanism of action (first in class) also make interesting candidates for this reason.

• Study previous successes of the company's management. Are they experienced with this [drug class, indication, etc] or not?
• Look at how their compensation / equity is structured. I believe this info can be found in their 10-K or annual report. 
• An executive compensation package that's heavily tied to performance (including stock and option grants) can indicate that the management's interests are closely aligned with those of the shareholders. If executives stand to benefit from the company's success, they're likely to be highly motivated to increase shareholder value.
• If a significant portion of executive compensation is in the form of stock options, it can suggest that management is confident about the company's future. That's because the value of these options is directly tied to the company's stock price: the better the company performs, the more the options are worth.
• High levels of compensation in the form of cash or guaranteed bonuses might suggest that the company is less risky, since executives are not relying on stock performance to make up a large part of their compensation. On the other hand, high levels of stock-based compensation can suggest a higher-risk, higher-reward scenario.

Construct a DCF analysis of the company

If you are lazy / strapped for time or your name rhymes with Shmalex Shmesin, you might be able to get away with ignoring the above "do's" and just focusing this step.

Note: DCFs work only for already mature biotechs with actual income streams.

• Ironically, contrary to the suggestion above, DCFs can be time consuming. But they are perhaps the most important step in valuing a company, as it assigns "real" numbers to a company's value.
• The main point of doing a discounted cash flow (DCF) is to remove potential bias and risk of overvaluing a company– a DCF is composed of a set of cold, unemotional formulas used to estimate the current tangible value of a company based in its projected future cash flows.
• Here's a mini "don't" – actively try not to let any bias slip into the inputs of the model. Sometimes there will be wiggle room or educated guesses you will have to make for numbers that you plug in. It's best to assume the most conservative scenario– unnecessarily positive guesstimates can fudge a company's value from being undervalued to overvalued.
• You will need to look at at figures like weighted average cost of capital (WACC), revenue growth, assets vs liabilities, to get the best DCF.
• Here is a great tutorial by Kostadin Ristovski on how to value companies using a DCF analysis (not specifically for biotech). This is the template he uses– likely you will need some editing for your own adjustments.
• Here is a tutorial by Richard Murphy if you're feeling a bit more adventurous. 
• I highly recommend watching both Kostadin and Richard's other videos by the way! 

       

Don't:

Try to trade options on biotech company news or catalysts

• Trading options is extremely risky for beginners– and not even the "experts" get it every time. In fact, the reason you hear about X financial firm or individual blowing up is likely because they were misusing options. When not done with utter caution and proper hedging, it can lead to disaster. See: Nassim Taleb's books.
• The first thing you have to worry about when trading biotech news is whether or not the new information has been "priced in"– as in the possibility that the option price has already adjusted to the news by the time you get around to noticing it. In a cutthroat industry like finance where bigger firms have higher quality and quicker information streams, this is always going to be a concern for the little guy. More on this later. 

• Biotech company options chains are often very illiquid. If you look at the spread between the bid and ask prices of biotech options, you will want to gag a bit. Spreads are so large in biotech because the prices of biotech companies are so volatile– hence, most reasonable don't want to speculate on such high risk, high reward gambling. Less trade volume = less likelihood you will get your desired trade filled, sorry.

• Many important trade determining factors are difficult to predict for catalysts.
• Often, speculators will use float and beta as crude measures to estimate the volatility of a company they want to trade on, and then use this as a measure of risk:reward during catalyst events. Unfortunately, it's probably more nuanced than that– what matters more than either of those is whether a company can withstand the setback of their drug failing a part of the clinical trial process. Whether or not the company have heavy milestone reward payments based on the performance of a given phase, how many other drugs are in the company's pipeline, and many of the points mentioned in the prior do's are vital to measuring its risk. Yes, volatility/risk is loosely and implicitly measured by those indicators, but knowing the real situation behind the scenes is always better than using an imperfect number to trade off of!

• It is almost guaranteed that the implied volatility of the company's options will swell to astronomical heights before a catalyst event. Speculator interest almost always balloons before an important company event, such as run up to the results of a phase III trial. Since the outcomes of said trials are so unpredictable, the volatility that others expect of the outcome will be very high (it could go up, it could go down... who knows!!). This exaggerated volatility will spike the premium on the options– as volatility plays an important factor in the pricing of options. Buyer beware: if you're playing a company on the day of a catalyst, you will probably be eating some loss from the get go via the high premium, especially if you're buying an option that's close to being in the money. 

• It is difficult to even guess the timing of catalyst events. For an entire month this summer, I wasted time messing around with data from Biopharmcatalyst, a database that claims to have accurate dates for upcoming catalysts, only to realize that all of my analysis was useless if I didn't know when the news would come (the data they collected about the timing of a catalyst was either really vague like Q2 2023, the date had already passed, or the date was straight up inaccurate). For reasons unbeknownst to us mere mortals, it is really difficult to find exact dates for catalysts that match up with the timing of a given company's option chain. So if you can't even time a trade correctly, how are you supposed to even gamble on purely the expected outcome of the catalyst!?

• In the world of biotech trading, you will often hear about trading "strategies" like covered calls, iron condors, and butterflies that hedge the risk of a given trade. While this may be true, they're not an exact science. Trading strategies are very sensitive to specific conditions, such as the stock price reaching $X target, which make them not always so predictable. Plus, these trading strategies are all based off of proper timing: see the point above for why that's difficult.

Invest in "vibes"

Last year, when I turned 18, I immediately opened an investment account. I've been begging my dad since the third grade to open up an account for me, so this was a dream come true to finally have.

Now, what did I immediately do to lose $50 of my $400 portfolio? I made an idiotic bet in April 2022 on the Twitter buyout, *as the stock price touched the buyout price*! This predictably ended in an embarrassing, poorly timed trade after Elon threatened to pull out of the deal (probably hindsight bias on my part).

My fortune took a turn for the better when I put a chunk of my portfolio into a biotech company called Gingko Bioworks. I had heard quite a bit of hype around it from Twitter and had a fancy ticker, so I shrugged and bought some shares for myself. To my surprise, I was able to nearly recover my Twitter loss from the performance of this stock, all based off of my trust in its brand! 

This is a terrible investment strategy, and it should go without saying. I had dumb luck in both directions without doing any real due diligence. It turns out, by the way, that Gingko makes for a terrible investment... they seem to be guilty of some sketchy accounting practices, which tanked the share price as soon as I sold my stake. Don't pass over the chance to invest in a company that has a very tangible sign of success– i.e. the approval of their drug– to invest in another based off of any superfluous details like a cool ticker or a funny CEO.

There is probably more to say here about more subtle bias in investing and stock picking, but tbqh these are difficult to spot. As long as you are as conservative as you can be with your DCF and other analyses, that's likely the best you can do.

Try to (exclusively) use an LLM to do the hard work for you

Trust me, I tried... we are not quite there yet. During that month I was gathering the data to try to play catalysts, I was simultaneously trying to parse dense 10-Ks, earnings calls, and video transcripts to extract valuable information about my companies of interest.

• A major problem I kept running into was that the context window of the best LLMs, like GPT-4 and even Claude 100K, are still too small. I learned this the hard way: I wasted so many OpenAI credits trying to chunk 10-Ks into readable bits without much success; splitting text into smaller segments to fit into the LLM context window yields the issue of the LLM being unable to holistically analyze a given text. What you end up getting when you do this is a boilerplate paragraph about what the text is generally about (yes, I know it's a 10-K already, you don't need to explain that to me!) and less insights.
• Another frustration I had when trying to use Claude to analyze company documents is that it's terribly inconsistent in its output, even when you give example of what the output should look like. I would ask to get output of only important figures mentioned in the text, and instead I would just get a summary of the text, with or without the numbers I wanted.
• LLMs also hesitate to give conclusive answers about things. Going in to this project I hoped I would be able to have GPT-4 autonomously come to its own conclusions about a given company. I'd ask ChatGPT if a company looked like it was performing well based on some given information, and it would mention important insights and then shy away from saying anything with any risk of being objectively wrong. "While A, B, C are true, is important to acknowledge that X, Y, Z also...." yada yada.
• I am still enormously excited for the potential of LLMs for this kind of thing– it's just a little bit out of reach for now, sadly. OpenAI's addition of function calling to the API looks promising enough that I might take another stab at this. I highly expect that the next iteration of LLMs, perhaps Google's Gemini– will crush this kind of reasoning task.

Look at macro *too* much

Biotech is (mostly) a market neutral industry, meaning that its performance for the most part is uncorrelated with the market as a whole. Thanks to the highly individualistic nature of the business, macroeconomic conditions will not be tremendously relevant for timing your investments.

Macro tends to play a bigger role in the VC world for biotechs, as interest rates highly affect how eager investors working in the most speculative aspect (pre seed, early stage, pre IPO companies) of the most speculative of industries (ahem) are to spend their money.

Expect to be making or losing money quickly

Pharma is a slow paced industry. From preclinical data to FDA approval could take the greater part of a decade for most given drugs. 

I personally find the biotech sector interesting to study for investing because I am naturally curious about the underlying products that have the potential to save and improve the quality of so many lives, i.e. drugs. Investing in biotech companies, directly, also funds medical science and research that advances the world for the better– and you have the chance to profit off of it! It's not always going to be the fascinating thing to invest in in terms of quick returns, and requires a lot of patience to see the beauty of it unfold.

An open secret...

I've spoken to a few people inside biotech companies and biotech related investment roles, and they agree that it can be a bit of an insider's game at times. Just like any other sector in finance, there are ways to learn about things ahead of everyone else.

People who work at well resourced places are able to use alternative data sources, or as they like to call them, "key performance indicators", to make better guesses about how the fortunes of certain biotechs will play out in the near future. For example, two people told me they are able to see how many written prescriptions a given drug of a company has over a given period, allowing them to estimate the sales performance of said company. Investment houses also have proprietary valuation tools that streamline the process of due diligence for them.

Another thing: as an institutional investor, you are able to speak privately with the management of the company you're interested in to assess the company's prospects. 

I don't expect these things to be too insightful or else they would be obvious insider trading, but it is troubling how much leverage institutions have over retail investors in this day and age. The rub? Most hedge funds, even with all the information they are firehosed with, don't perform any better than the S&P 500. So three cheers for that!