In my view, RDCFs are a useful tool to test against the implied consensus built into the share price. Therefore I try to reduce own assumptions to the extent possible by using consensus estimates as far out as possible - and if needed - adding a staging element to it (e.g. tapering growth, etc.).
That would give you an implied market expectation - I.e to justify the current price, the topline to must x% for x years, margins to grow x pp, etc. Based on that, you must evaluate if said expectations are too conservative/optimistic.
Great article! I also prefer solving for "r". Some questions/comments.
- I also like NOPAT for measuring profitability of the company. However, but for the purpose of calculating cash that can be distributable to owners, shouldn't we also subtract average expected interest payments? Because we use effects of debt in growth.
- Depreciation and amortization (D&A) seems substantially bigger than CAPEX, but it needs go into CAPEX first than to be discounted in D&A, am I wrong?
Hi, thanks for your comment and I'll try and clarify some things.
First, I use NOPAT because I want to calculate the intrinsic value of the entire business, for which it does not matter what the capital structure (debt and equity) looks like. Therefore, I don't subtract interest payments. I recommend reading this article by Aswath Damodaran: https://aswathdamodaran.substack.com/p/earnings-cash-flows-and-free-cash
Second, D&A in this example is much higher because Salesforce is a capital-light business. Most of their investments are expensed as R&D instead of capitalized. Don't take this example too seriously, it's only for illustrative purposes. Valuing Salesforce would involve capitalizing R&D expenses and amortizing it over time.
Third, in regards to solving for r instead of g, it seems to me mostly a personal preference. Setting the discount rate to the WACC, for example, and solving for g will lead you to a similar conclusion. I like solving for r, because I can easily compare implied returns to a hurdle rate and find price targets, etc. It's just more intuitive to me!
I absolutely love this approach! Reverse-engineering DCF gives you so much more flexibility compared to the traditional method. It flips the conventional idea of risk—usually tied to the hurdle rate—into a lens for exploring potential returns. For me, this opens up a whole new level of flexibility in the assumptions (and potential outcomes) investors can play with. I recently tried this method on a valuation of Edenred, and ever since, I’ve been a huge fan. It’s such a game-changer!
In my view, RDCFs are a useful tool to test against the implied consensus built into the share price. Therefore I try to reduce own assumptions to the extent possible by using consensus estimates as far out as possible - and if needed - adding a staging element to it (e.g. tapering growth, etc.).
That would give you an implied market expectation - I.e to justify the current price, the topline to must x% for x years, margins to grow x pp, etc. Based on that, you must evaluate if said expectations are too conservative/optimistic.
Great article! I also prefer solving for "r". Some questions/comments.
- I also like NOPAT for measuring profitability of the company. However, but for the purpose of calculating cash that can be distributable to owners, shouldn't we also subtract average expected interest payments? Because we use effects of debt in growth.
- Depreciation and amortization (D&A) seems substantially bigger than CAPEX, but it needs go into CAPEX first than to be discounted in D&A, am I wrong?
- Although it seems for solving for r seems best option to me as you do, the published material on that is very limited. For Reverse DCF - people generally refer solving the equation for g: https://www.investopedia.com/articles/fundamental-analysis/09/reverse-discount-cash-flow.asp
IRR for solving r: https://www.investopedia.com/terms/i/irr.asp
Hi, thanks for your comment and I'll try and clarify some things.
First, I use NOPAT because I want to calculate the intrinsic value of the entire business, for which it does not matter what the capital structure (debt and equity) looks like. Therefore, I don't subtract interest payments. I recommend reading this article by Aswath Damodaran: https://aswathdamodaran.substack.com/p/earnings-cash-flows-and-free-cash
Second, D&A in this example is much higher because Salesforce is a capital-light business. Most of their investments are expensed as R&D instead of capitalized. Don't take this example too seriously, it's only for illustrative purposes. Valuing Salesforce would involve capitalizing R&D expenses and amortizing it over time.
Third, in regards to solving for r instead of g, it seems to me mostly a personal preference. Setting the discount rate to the WACC, for example, and solving for g will lead you to a similar conclusion. I like solving for r, because I can easily compare implied returns to a hurdle rate and find price targets, etc. It's just more intuitive to me!
My point is using a version of FCF to Equity owners would be better approach to value stock price, at least in my opinion.
It's up to preferences, but I would like to know the value of the entire operating business regardless of capital structure. Therefore, I use FCFF.
I absolutely love this approach! Reverse-engineering DCF gives you so much more flexibility compared to the traditional method. It flips the conventional idea of risk—usually tied to the hurdle rate—into a lens for exploring potential returns. For me, this opens up a whole new level of flexibility in the assumptions (and potential outcomes) investors can play with. I recently tried this method on a valuation of Edenred, and ever since, I’ve been a huge fan. It’s such a game-changer!
I completely agree, glad you liked it!
Good work! I'll dive in it asap.