Email exchanges in March 2019 among:
Carlo Alberto Magni
As I already explained to Joe (who unilaterally chose to share only part of what I wrote to him)
FCF is designed to be the same no matter the capital structure. That is the point of it. Force people to not get confused and forget MM.
Using FCFE discounted at the cost of equity gives you the same results, if you do it right. If you know what you are doing, I do not advocate one or the other method. If you are going to do FCFE please avoid:
- Feeling that debt is cheaper than equity because Ke > Kd
- Forgetting to account for net proceeds from issuance of debt – very easy in a growing perpetuity firm model with constant leverage, something you are likely to use for residual value in a DCF valuation model
- Forgetting the MM theorem in any other way
Mattia Landoni <firstname.lastname@example.org>
FCF is easier to sell than CCF to corporate audience. People love tp hear “free cash flows”. I guess it might be the reason it is so popular and Stern was the big sponsor in their book The Quest of Value.
FCF is the project cash flows..without concerning too much on TS. Many finance decisions at corporate life, they are more focused on whether ROA is higher than Kd when the financing decision comes up. TS becomes limited now since the tax authority limit debt to equity for max 4x. So we cannot exploit lnterest TS too much.
I see FCF as unnatural for the following reasons:
1) It is not the cash flow which is distributed to capital providers. It is only a part of the distributed cash flow (the one which does not depend on the capital structure)
2) the textboook WACC method assumes that a firm rebalances its leverage ratio. This is a strong assumption, most firms do not rebalance the leverage ratio (even more so if one is evaluating a single project rather than a firm).
Carlo Alberto Magni
Rauf Ibragimov <email@example.com> wrote:
The next moment a firm engages in the external financing transactions or retains cash, the FCF (as it is basically defined) becomes an artificial construct. CCF (as defined by Ruback) incorporates the tax advantage of debt, so it is an improvement (though, limited, as it builds on the FCF) in measuring cash available to satisfy existing claims. An advantage of the CCF approach in valuation could be no circularity and absent need to periodically recalculate the discount rate, however, forecasting CCF requires an explicit financing plan. The latter could be advantage as well, since implicit in the FCF-WACC approach is a generally unrealistic assumption of a constant leverage (market values) financing policy
As with the daily goods, consumer preferences follow the size of the marketing budgets for competing offers
Mattia Landoni <firstname.lastname@example.org> ha scritto:
I already shared the second and last part of what I wrote to Joe.
Further to the point:
1) I never saw myself (or my colleagues) as victims of marketing. I have studied finance a long time and I have earned the right to claim that I am an independent thinker on the subject. Also, as I stated, my favorite approach is state prices.
2) FCF is needed to do either WACC or APV, so if one doesn’t like the constant leverage assumption of WACC, that doesn’t eliminate the need to define FCF. FCF is the cash generated by the firm’s operations that is available to distribute to investors. It is not a part of the cash flow distributed to investors; it could be more OR less than what is distributed to investors in any given period, and typically more in the long run.
3) I teach three valuation methods: FCF-WACC, FCF-APV, and FCFE (which I assume is what we are calling “CCF” here. I could teach a fourth if it existed and if I learned of an example in which it is clearly superior.) More often than not I observe that my undergraduate student become confused when they do FCFE with time-varying debt.
4a) Both FCF discounted at WACC, and FCFE discounted at Ke, have an assumption of constant leverage. If you don’t agree, you appear to be forgetting the MM theorem. That is why I like FCF with APV – no confusion. You get separate estimates of the value created by financing and the value created by operations.
4b) Nonetheless, the thing that started this whole discussion is a table in Berk-DeMarzo showing that you can do FCF-WACC with time-varying leverage; it’s just hard work. The same hard work is necessary if you want to use FCFE discounted at Ke.
5) Finally, the constant leverage assumption is necessary and reasonable in certain situations. If I write down a DCF to value a firm in years 1-10, and I need a residual value for years 10-1000, I assume constant leverage. As Keynes would say… “what do you do, sir?”
Carlo Alberto MAGNI <email@example.com> wrote:
Hi Mattia, thanks for your comments.
The FCF is one “part” of the CCF, the TS being the other part:
Obviously, as you imply, either part of CCF may be positive or negative.
I agree that APV creates no confusion: both parts are discounted at the respective risk-adjusted rates of returns, and there is consistency in the both ratios in the following sense: The unlevered cost of assets (denominator) reflects the risk of FCF (numerator), and the discount rate for TS (denominator) reflects the risk of TS.
Precisely for this reason I do not like FCF-WACC method: it does not preserve consistency between numerator and denominator and turns an unlevered cash flow into a levered value. I see the WACC as a plug which is necessary to make the leap from unlevered perspective to levered perspective. While mathematically correct, it is conceptually unsatisfying and difficult to digest.
In contrast, CCF is discounted at a rate which is significant, because it is the mean of the discount rate for FCF and the discount rate for TS. This is rather natural, given that CCF may be viewed as a portfolio of FCF and TS. Further, since the discount rate for CCF is also equal to the average of cost of equity and cost of debt, we have two different perspectives for conceptualizing it: investment perspective and financing perspective, respectively. This reinforces its significance.
“Mattia Landoni” <firstname.lastname@example.org>:
Thanks. I thought CCF was another word for CFE or FCFE. I had never heard about CCF. I see now that we were talking about different things.
If you define CCF as FCF + tax shield, and the appropriate discount rate as a weighted average of the unlevered cost of capital Ku and whatever appropriate rate for the tax shield Kts, then this sounds very similar to APV – I would call it “APV without keeping the pieces separated”. I suppose I have nothing against it, but before I teach it to my students as a fourth method, I need to be convinced that it has distinct advantages over APV.
May be difficult to apply,
Used for matching,
Effective as a check on consistency,
Practically difficult to estimate
Totally ignored and neglected,
Superior to the FCF,
Usefulness is underestimated
May be confusing for accountants
Indeed, one may view the CCF method as a reframing of the APV method, where the two components, FCF and TS, are merged together into one single cash flow (the CCF) which represents is the cash flow which is distributed to investors. For this reason, It is also called “compressed APV”.
VL = CCF/x = Vu+VTS = FCF/ku+TS/k^T
x=(ku Vu+ kT *VTS)/(Vu+VTS)
VL = CCF/x = E+D = CFE/ke+ CFD/kd
x=(ke E+ kd D)/(E+D)
In finance textbooks the expression “CCF method” is mainly used with alongside the assumption kT=ku. This implies
x=ku=(ke E+ kd D)/(E+D)
so that V=CCF/ku.
The latter version has been introduced by Ruback.
Hmm. The disadvantages of CCF do not readily come to mind. However, in the spirit of full disclosure, we must confess that we not unbiased assessors since we have been unsuccessfully peddling the CCF over two decades!!!
I prefer CCF as intuitively much simpler than APV. I explain CCF in my lectures, not APV.
Carlo Alberto Magni
Would you have a detailed financial example that would demonstrate clearly the superiority of CCF? I fully support Mattia’s skepticism. He is not happy with my hand waving and nontechnical explanations, understandably.
Thanks for your support, Joe 🙂
I am attaching my slides in which I explain 3 valuation methods to my students (APV, WACC, FTE). They are full of examples that should help as a starting point. The slides start off with an example that is easy with all 3 methods and end with examples where one or the other method is clearly preferred. Perhaps one of these examples can be made even easier or more insightful using the CCF approach. Or perhaps these examples will provide inspiration to come up with a different example.
In the interest of science and education I am taking the unprecedented step of sharing the whole slide deck, including my private notes to self (the gray pages). Please understand that these are raw and unedited.