I was just looking at this company, so this deep dive is very timely. I was thinking about things like the strange situation with "compounders", or whatever they are called, making legal knockoffs of the weight drugs. Will oral weight drugs happen and take off? Will Novo be the one to make them. This way of thinking about R&D and capex as growth engines and not just an expense is interesting. As a slight counter: as a company grows wealthy, it has more money to throw at new ideas. Is it more successful at picking the new ideas? Will the research pan out, just because more resources are going towards it. I would point towards Meta and the buckets it has thrown at virtual reality, which still seems to have as much practical application as crypto at this point. I don't know how to make a paragraph break in a comment. Back to pharmaceutical companies, the industry has a known issue of, we don't know how to figure out the next great products. Maybe AI augmented research will get us there. Suffice it to say, seeing Novo Nordisk as a significant percentage of its country's gdp is astounding. Your article was food for thought for this company and others. Awesome.
What I usually do is I accept FCF with the SBC included in it; however I look at the change in total shares outstanding over the last 48m or so and extrapolate that into my DCF and other valuation metrics assuming the same trend of dilution. Would you say that’s incorrect ?
There’s several companies that use SBC but also do buybacks.
Loved the way you split up the capex; this was exactly the kind of read on FCF I was looking for; defining the earners returns !
Hey, I came across your comment and wanted to give my answer.
A DCF considers all future cash flows and discounts them to today to reach a certain valuation.
Shares are bought back using free cash flow. Therefore, if you're building a DCF and assuming share buybacks, you're essentially double counting.
This is a common error even Michael Mauboussin has discussed in some of his work. So I would never extrapolate the change in shares outstanding in a DCF model; it simply overstated a stock's value.
I like the thinking here; treating both R& D and Capex in a more nuanced and accurate style. I’m sure it hasn’t escaped your attention that the CEO has been replaced and the company is losing the GLP war against Lilly. I’d argue that in Pharma, some intangibles like management quality and company culture are critical to success.
Thanks -I will. Even before the GLP-1 story, Novo was an interesting company; highly successful despite (because?) its focus on diabetes, selling ever more sophisticated versions of a very old drug (Insulin).
Great article - thank you for sharing your work! I was curious about your view on ROIIC — if you were to capitalize the R&D spend and put it on the balance sheet (which I think you are eluding to), over how many years would you depreciate it? The economics, and thus ROIIC, would be highly sensitive to that assumption.
I’m not sure whether this is visible in the reported numbers, but if their expansion projects have experienced budget overruns or dealys — or if some of the investments were made during a period when Novo was far ahead (with lofty sales and margin expectations) — potential write-downs could be on the horizon. In a sense, making some of the growth capex investments less attractive/not as good capital allocation decisions. How would you incorporate things like this?
I wrote a follow-up article (not sure if I linked it in that post) in which I walk you through my math.
It's not exact math anyhow, just helpful to think about the economics from a different angle that I believe is closer to "truth" than the reported figures.
Regardless, growth capex or maintenance this would be included in a DCF so the value per share you will arrive at will be much lower than "hypothetical FCF". You can capitalise R&D, estimate useful life and then ammortise it, which will be a way better metric to uplift earnings (and FCF) but any cash investment into Capex or R&D needs to go through the cashflow statement.
For drugs there is no steady state FCF. You could try and model each drug's FCF and then value Novo on a SOTP basis. The steady state value that you would derive by assuming zero growth and using "maintenance R&D and Capex" would vastly overstate the value for a biotech. Usually drugs have a 20-year patent life + potential market exclusivity. Even to maintain a drug such as wegovy or ozempic would require R&D > assumed maintenance R&D.
Sure, I looked at it through a "multiple lens." If you want to adopt the DCF approach, realizing that a good chunk of R&D is growth-related, you could model faster growth once those investments are slowed down.
I also agree that pharma is a R&D heavy business. But the most recent CapEx and R&D levels of Novo greatly exceed historical levels (as a % of revenue). Being aware of this and possibly making adjustments is important imo.
Although, I have one concern. By adding all growth R&D directly to FCF, you implicitly assume that all R&D investments will be successful and generate future FCF.
In my opinion, this is is far to optimistic, especially in the pharmaceutical industry, where not all R&D projects succeed. Some R&D investments fail, not every medicine will pass clinical trials and make it to the market. In fact, very few do. Yet this approach assumes a 100% success rate, which gives an overly optimistic view of the company’s FCF.
I think, a more realistic way to treat R&D expenses as investments is to capitalize them on the balance sheet.
I was just looking at this company, so this deep dive is very timely. I was thinking about things like the strange situation with "compounders", or whatever they are called, making legal knockoffs of the weight drugs. Will oral weight drugs happen and take off? Will Novo be the one to make them. This way of thinking about R&D and capex as growth engines and not just an expense is interesting. As a slight counter: as a company grows wealthy, it has more money to throw at new ideas. Is it more successful at picking the new ideas? Will the research pan out, just because more resources are going towards it. I would point towards Meta and the buckets it has thrown at virtual reality, which still seems to have as much practical application as crypto at this point. I don't know how to make a paragraph break in a comment. Back to pharmaceutical companies, the industry has a known issue of, we don't know how to figure out the next great products. Maybe AI augmented research will get us there. Suffice it to say, seeing Novo Nordisk as a significant percentage of its country's gdp is astounding. Your article was food for thought for this company and others. Awesome.
Really enjoyed your article Rene !
On FCF:
What I usually do is I accept FCF with the SBC included in it; however I look at the change in total shares outstanding over the last 48m or so and extrapolate that into my DCF and other valuation metrics assuming the same trend of dilution. Would you say that’s incorrect ?
There’s several companies that use SBC but also do buybacks.
Loved the way you split up the capex; this was exactly the kind of read on FCF I was looking for; defining the earners returns !
Hey, I came across your comment and wanted to give my answer.
A DCF considers all future cash flows and discounts them to today to reach a certain valuation.
Shares are bought back using free cash flow. Therefore, if you're building a DCF and assuming share buybacks, you're essentially double counting.
This is a common error even Michael Mauboussin has discussed in some of his work. So I would never extrapolate the change in shares outstanding in a DCF model; it simply overstated a stock's value.
Hope that helps!
Great job!
Thank you, Nisim. Have a great weekend.
Excellent article ! A refreshing and interesting perspective on company valuation - a great new tool
You're quick JG! I appreciate that, thanks very much and have a nice weekend ahead.
This is pure Wisdom. Thanks!
I like the thinking here; treating both R& D and Capex in a more nuanced and accurate style. I’m sure it hasn’t escaped your attention that the CEO has been replaced and the company is losing the GLP war against Lilly. I’d argue that in Pharma, some intangibles like management quality and company culture are critical to success.
Thanks -I will. Even before the GLP-1 story, Novo was an interesting company; highly successful despite (because?) its focus on diabetes, selling ever more sophisticated versions of a very old drug (Insulin).
Sure. I‘ve released two new pieces on Novo recently. Go and check them out.
In terms of Novo losing … I don’t know. Things seem to be turning again as of last week.
Great article - thank you for sharing your work! I was curious about your view on ROIIC — if you were to capitalize the R&D spend and put it on the balance sheet (which I think you are eluding to), over how many years would you depreciate it? The economics, and thus ROIIC, would be highly sensitive to that assumption.
I’m not sure whether this is visible in the reported numbers, but if their expansion projects have experienced budget overruns or dealys — or if some of the investments were made during a period when Novo was far ahead (with lofty sales and margin expectations) — potential write-downs could be on the horizon. In a sense, making some of the growth capex investments less attractive/not as good capital allocation decisions. How would you incorporate things like this?
I wrote a follow-up article (not sure if I linked it in that post) in which I walk you through my math.
It's not exact math anyhow, just helpful to think about the economics from a different angle that I believe is closer to "truth" than the reported figures.
https://www.compoundwithrene.com/p/forget-roic-this-metric-tells-you?utm_source=publication-search
Regardless, growth capex or maintenance this would be included in a DCF so the value per share you will arrive at will be much lower than "hypothetical FCF". You can capitalise R&D, estimate useful life and then ammortise it, which will be a way better metric to uplift earnings (and FCF) but any cash investment into Capex or R&D needs to go through the cashflow statement.
For drugs there is no steady state FCF. You could try and model each drug's FCF and then value Novo on a SOTP basis. The steady state value that you would derive by assuming zero growth and using "maintenance R&D and Capex" would vastly overstate the value for a biotech. Usually drugs have a 20-year patent life + potential market exclusivity. Even to maintain a drug such as wegovy or ozempic would require R&D > assumed maintenance R&D.
Sure, I looked at it through a "multiple lens." If you want to adopt the DCF approach, realizing that a good chunk of R&D is growth-related, you could model faster growth once those investments are slowed down.
I also agree that pharma is a R&D heavy business. But the most recent CapEx and R&D levels of Novo greatly exceed historical levels (as a % of revenue). Being aware of this and possibly making adjustments is important imo.
Interesting article. Shouldn’t Novo Nordisk 11.5x price/true-FCF be compared to peers before stating that it’s low? Too low compared to what?
I don't place much (if any) emphasis on relative valuation and try to focus on absolute value only.
Very interesting approach. Thanks for sharing!
Although, I have one concern. By adding all growth R&D directly to FCF, you implicitly assume that all R&D investments will be successful and generate future FCF.
In my opinion, this is is far to optimistic, especially in the pharmaceutical industry, where not all R&D projects succeed. Some R&D investments fail, not every medicine will pass clinical trials and make it to the market. In fact, very few do. Yet this approach assumes a 100% success rate, which gives an overly optimistic view of the company’s FCF.
I think, a more realistic way to treat R&D expenses as investments is to capitalize them on the balance sheet.
The Buffett-Munger Profitability Investing Truism Dharma 124:
Genuine FCF TTM
Based on Bruce Greenwald's Growth Cap Expenditure, I arrive at the followings:
Growth Cap Expenditure TTM
= Average 7 Years { (PPE + ROU + Depreciation of PPE + Depreciation of ROU) ÷ Revenue } × Revenue's Increment TTM
or
Growth Cap Expenditure TTM
= { ∑ 7Year (PPE + ROU + Depreciation of PPE + Depreciation of ROU) ÷ ∑ 7Year Revenue } ÷ 7 × Revenue's Increment TTM
Maintenance Cap Expenditure TTM
= Investing Cash Flow's Capital Expenditures in PPE & ROU
Genuine FCF = Owner's Earnings
= Owner Earnings TTM
= Net Cash Generated by Operating Activities TTM - Maintenance Cap Expenditure TTM
CROIC TTM
= Owner Earnings TTM ÷ Invested Capital
Quality of Earnings TTM
= CROIC TTM ÷ ROIC TTM
Note:
Adjusted Net Profit for numerator of ROIC is chosen because of
FASB was right: Earnings beat cash flows when predicting future cash flows
Ray Ball† Valeri Nikolaev‡
December 2020
****