I have written a few posts recently offering valuations of companies I am either interested or invested in. As a result, I have had a few questions regarding how exactly I implement discounted cash flow (DCF) calculations into my valuations. This week I have therefore decided to write a little about how and why I incorporate a DCF when looking at companies.
The basic principle underpinning the discounted cash flow method is in projecting future cash flows and discounting them back to the present using an appropriate rate. To simplify, “what do I have to pay today to earn a predicted rate in future”? The DCF method has been spoken about at some length by legendary investors Warren Buffett and Charlie Munger. Obviously, and with good reason, people frequently take the opportunity to ask them both questions, and they have spoken regarding DCF models in some detail:
Buffett: “All investing is laying out cash now to get some more back in the future. If you need to use a computer or calculator to figure it out, you shouldn’t [buy the investment].”
Munger: “Warren talks about these discounted cash flows. I’ve never seen him do one.”
That being said, Buffett does incorporate some form of a DCF into his valuation models. It all comes down to the discount rate. Where most investors will use a complex system to determine their discount rate (and adjust it according to each company), Buffett uses the yardstick of the 30 year Treasury rate across all valuations. It helps, of course, that he has over 50 years experience and a financial brain far more advanced than us mere mortals. No wonder he doesn’t use a computer. For an excellent discussion on determining discount rates, I recommend visiting Propertymetrics.com.
The Discounted Cash Flow Method
First let me say this. Using a DCF is one of many variables I use when determining the value of a potential investment. In fact, it is one of the final pieces of the puzzle for me. I have been remiss in previous posts in suggesting one intrinsic value, and not assigning a range of possible valuations to a company using a DCF method. If anyone tells you the intrinsic value of a company is X, run away very fast and very far.
Secondly, you may already realise a DCF method is only applicable to certain companies. The method is based upon free cash flow, therefore companies which do not generate FCF cannot be valued using this method. Companies with unstable or unpredictable earnings (i.e. junior oil or mining stocks) will likewise fall foul of a DCF. It is most effective when used with established, profitable companies with relatively predictable cash flows and stability. One reason I enjoy using a DCF is because it keeps me out of anything speculative. Been there, done that, lost money.
Thirdly, a DCF uses a number of variables when arriving at a valuation, and is therefore very sensitive to manipulation. You can attain wildly different valuations by adjusting your inputs, hence why you are better off establishing a potential range, rather than one fixed figure. The method is also forward looking. Projections of future cash flows are required, based on historical figures. Whilst it is possible to some degree to do this, any prediction is just that.
Buffett again: “The concept of “a bird in the hand” came from Aesop in about 600 BC. The question is, how many birds are in the bush? What is the discount rate? How confident are you that you’ll get [the bird]? Et cetera.”
How I Use a DCF
I believe the best way to explain is using an example. I have therefore chosen Greggs (GRG) as my valuation test subject. Firstly, I use Morningstar to obtain ten years’ worth of company figures (for some reason the US site is more comprehensive than the UK). Greggs has grown revenue each year for the past ten years, however net income has been inconsistent, as has free cash flow.
Using the ten year average FCF gives us a figure of £20.8m. I am happy to use this average as FCF has, as mentioned, been inconsistent. A smaller, growing company with growing cash flow may require projections of anticipated FCF growth, determined on an ad-hoc basis.
FCF Growth Rate
I then need to establish a FCF growth rate. For Greggs, this is tricky. On one hand, FCF has grown at an annualised 18% for the past ten years. For five years, it has grown at 23.58%. However, for the past nine years it has only grown at 10.5%. The difference of five years, or even one year is substantial. What about the growth in net income? Over ten years, this has grown at a pitiful 1.24%. We’ll bear these figures in mind for later.
FCF Growth Period
We thirdly need to determine how long this growth can continue. Due to the sporadic nature of Greggs’ FCF I’d rather use a shorter time frame (5 years) than longer. Again, for a smaller, faster growing company I would feel more confident using a longer time frame (10 years). Beyond this term I use a figure of 3% annual growth in perpetuity, regardless of company. This approximately matches average rates of inflation.
Lastly comes the discount rate. The rate of return I would be happy to achieve were I to take a position in a company. There are many formulaic ways to attain this, but I confess I try to keep it as simple as possible. We already know Buffett incorporates the 30 year Treasury rate (currently a shade over 3%) and wants an adequate margin of safety over this. Some people adjust their discount rate by the assumed risk attributed to the company (i.e. a higher rate for a more risky company). This to me is nonsense. Why should a riskier company automatically promise higher returns?
I generally begin with a rate of 10%. 10% annualised would be satisfactory, if not spectacular. I prefer to be conservative with my calculations as to attempt to incorporate my own margin of safety, but not so conservative I end up with an unsatisfactory return.
I currently use the Warren Buffett Intrinsic Value Calculator on the Buffetsbooks.com website which takes all these inputs and calculates the intrinsic value automatically. It is possible to put these formulas into Excel, and many sites will provide these formulas, but I’m lazy. So say we use:
- A FCF average of £20.8m
- FCF growth rate of 17.38% (mean average of 5, 9 and 10 year average FCF growth)
- A 17.38% FCF growth rate for the next five years
- 3% growth in perpetuity after this
- A 10% discount rate
This returns an intrinsic value of £5.72, against a current price of £10.05. Obviously not an ideal investment from this perspective.
Last years’ free cash flow was £49m. If we use this, the intrinsic value goes up to £13.49. A massive difference, which may change our opinion on this potential investment. This is where the calculations become more art than science. My personal view, based on looking at ten years’ figures for Greggs is that they are unlikely to consistently return £49m FCF going forward, and certainly wouldn’t grow it at 17.38% a year. Even at 10.5% growth (five year average), intrinsic value is determined to be £9.79. When we also factor in the aforementioned ten year net income growth of 1.24% this only seeks to reinforce my belief that the lower end of the valuation may prove more accurate going forward.
Lastly it is worth repeating that this DCF method is very much a blunt tool to try to help ascertain the suitability of an investment. I would usually be mostly decided on a company before even doing a DCF. It is also worth mentioning that it is quite easy to manipulate the figures to fit your assumed valuation. It may be that you love a company, or believe it has a bright future, and often times it can be easy to just tweak everything upward a little bit to get the figure you want. Like most aspects of investing, the psychology underpinning your methodology is of paramount importance.
If you have any views on this article, or would like to discuss further, find me on Twitter @britishinvestor.