There seem to be a lot of different ways to measure free cash flow, and I’m curious how the P123 community thinks about the various issues.
The most basic measure is OperCashFl - CapEx.
P123’s formula is OperCashFl - CapEx - DivPaid.
An alternative formula that appeals to me is OperCashFl + CashFrInvest. Basically this formula makes no distinction between CapEx and Acquis. There’s certainly a strong case that can be made, especially when it comes to tech companies, that the distinction between capital expenditures and acquisitions is nil. And why should divestitures not diminish capital expenditures?
Then we get into the financing activities part of the cash flow statement. What if your capital expenditures are financed 100% by new debt? In that case the money you’re spending would appear in the financing activities rather than under CapEx (or am I wrong about that?). And is it right to include dividends paid in a measure of free cash flow?
And then there’s unlevered free cash flow. This is a basic item in DCF analysis. While free cash flow can be compared to equity, unlevered free cash flow should be used for the firm. So I use unlevered free cash flow to enterprise value as a value ratio and unlevered free cash flow to assets as a quality ratio. To get unlevered free cash flow, you add back most of the interest expense (deducting a little for taxes). Now you wouldn’t want to include dividends paid in unlevered free cash flow, right? But what about acquisitions and divestitures?
Yuval, I can’t comment broadly on variations in free cash flow measures, but I “messed around” some with adjusting for R&D and ultimately it seemed unrewarding. In my mind R&D (or at least a fraction of R&D) seems to have similarity to capex (an investment in future earning potential), but I didn’t find adjusting FCF or unlevered FCF for that to be fruitful. Maybe I was looking at it improperly, but it just didn’t seem to make much difference.
If you look at the cash flow from operations, R&D is always reported as part of the operating expenses. So if you deduct it from free cash flow, you’re basically deducting it twice. Capital expenditures, however, are by definition not part of the operating expenses, so it’s proper to deduct them when you’re calculating free cash flow.
It could be argued that acquisitions are usually one-time expenditures, but I’ve found that not to be the case. If you screen for companies that reported acquisitions over the last year, well over half (as high as 80%, depending on what universe you look at) also reported acquisitions either the year before or the year before that. So I’m now leaning strongly in favor of deducting acquisitions along with capital expenditures, and to get to net acquisitions, adding back divestitures. As for cash flow from other investing activities, I need to do some more research. If these are already included in the net income part of the cash flow from operations, I don’t want to count them twice.
I am not an accountant, so puzzling this stuff out takes me a long time. But I’m always cautious when there’s a lot of adding back or removing this or that expense. For instance, I’ve read (I forget where, maybe something Buffet wrote) that EBITDA isn’t a good measure because the DA are real costs to a business, so EBIT is better, but people use it for what seems like good reasons and it tests well. OperCashFl - CapEx makes sense to me because if I owned the whole business I wouldn’t be paying dividends to anyone else. It seems to me that adding back interest expense would make highly levered firms look better.
I might be misremembering my accounting classes, but I could have sworn that they would both be there. That is, they’re not net of each other, so if you adopt more debt to spend on CapEx then you’d add to both the debt lines and the CapEx line.
The accounting department will simply add to the debt if you take out more debt, and then if you use those funds for capex, it will simply add to the capex. As I recall, in journal accounting, nothing is ever deducted from the accounts (aside from error corrections).
The idea of including cash dividends in our free cash flow calculation is that traditional business analysis treats dividends as voluntary. They are, in a sense. Management can slash the dividend whenever they want, and we investors all probably had a visceral emotional reaction to the first part of this sentence. That’s the problem.
Management can slash the dividend whenever they want, but they will get their butts handed to them by the board if there’s no good reason and the stock price will plummet if there is a good reason. It’s an extremely bearish signal. Once they start paying dividends, they’re committed just by market forces.
That’s why we don’t treat dividends as voluntary. (I also believe that the line that we use also includes preferred dividends, which are explicitly involuntary.)
Think of our free cash flow this way: Rather than figuring out what the maximum that a company could pay in dividends, we’re looking at how much more they could pay than they are paying now. If that doesn’t meet your needs, you can just add the dividends back.
Loosely, I define free cash flow as the amount of cash leftover from operations which a company is not compelled to reinvest in order to maintain the “status quo” (e.g., in terms of stock price, future revenues, cost structures, etc…). Using this definition, estimating maintenance capex is obviously an essential step. Maintenance Capex is the amount of spending which is mandatory. The remainder of CapEx then is discretionary (vis-a-vis “growth” capital).
Stern’s Value Managements Enterprise Value Add methodology makes explicit adjustments to a single period cash flow which is meant to be descriptive of the firm’s sustainable excess earnings. See: http://www.investopedia.com/university/eva/
Warren Buffett himself said that the best estimate for maintenance capex is most often stated depreciation, amortization, and depletion. Generally, he is correct. However, declines which may result from under-investment are not linear and/or immediate as implied by accounting conventions used to calculate DD&A.
If you find a better mechanistic way to estimate Maintenance CapEx, you may have found a probabilistic edge over other market participants. If you think of something, please do share in private and I will be sure to reciprocate.
I guess we need to think about how to define free cash flow before figuring out how to calculate it. One way to think about free cash flow is that it’s the excess cash available to a firm that they can use to either give back to shareholders or to finance future growth. If we include dividends in free cash flow, then it controverts that definition. And there’s no good reason not to include cash flow from investments when figuring this out, especially since for many tech and health care firms, their acquisitions basically ARE their capital expenditures, and since for many financial firms, the stuff in the InvstOther line is mad money.
Another way to look at free cash flow is for DCF analysis, which I think is really the way David defined it above: “the amount of cash leftover from operations which a company is not compelled to reinvest in order to maintain the status quo.” In this case you would limit yourself to operating cash flow and maintenance capex, adding back after-tax interest since you’re dividing by enterprise value and you don’t want to count debt twice.
So for my ranking systems, when I look at free cash flow growth, free cash flow margin, free cash flow to assets, and free cash flow yield, I’m going to use OperCashFl + CashFrInvest, and when I look at unlevered free cash flow to enterprise value, I’m going to use OperCashFl - CapEx + 0.8*IntExp. That seems to be what works the best for me . . . But I may be mistaken . . .
Yuval, I deleted a post I made above, but I think should’ve left it. Even though I couldn’t see it make much difference, conceptually adjusting for R&D seems sound if you think R&D is more like investment than expense, and if you think perhaps all R&D shouldn’t be expensed in the year it occurs. Maybe you can make more of it than I could.
The basic idea is to capitalize R&D instead of expense it. So if you treat R&D as an investment instead of an expense, and pull out R&D amortization over time instead of expensing it all at once, it could make a difference for some companies. It requires building an R&D asset and pulling out R&D amortization from that. I like the concept. Some companies have huge R&D expense that isn’t all tied to current year earnings. I tried putting R&D expense back in OpCF and making an R&D asset that depreciates straight line over 6 yrs and amortizing that amount, but couldn’t tell that it made much return predictability difference vs. vanilla calc. But very possible I’m just doing something wrong or not thinking about it properly.
Yes. But more because of laziness in changing things than because I’m convinced that they’re correct.
As I wrote last year, “There is no clear method for determining future cash flows. I must have come across at least a dozen of them. (Mauboussin’s is explained in a book he co-wrote with Alfred Rappaport called Expectations Investing.) Most of them involve inputting a lot of guesses. There isn’t even a standard way to measure present free cash flows. Many analysts use different formulae from those provided by the CFA Institute. A big question is how to calculate maintenance capital expenditures (as distinguished from those that contribute to growth); if one deducts all capital expenditures from operating cash flow, then certain industries, such as utilities, have overall negative cash flows. And then there’s the problem exemplified by Hertz (HTZ), in which a company with a market cap of $1.25 billion has maintenance capital expenditures of $10 to $12 billion a year (the amount it spends on new cars), resulting in massively negative free cash flows.” See How to Be a Great Investor, Part Two: Understand Value - Portfolio123 Blog Regarding Hertz, our data providers categorized its new car expenditures as cap ex but its income from selling old cars in an entirely different category but still under cash flow from investing. When it comes to free cash flow, ideally each company should be examined individually. This is one of the signal flaws of the algorithmic approach that I take to investing, and I consider it a serious one.
One possible solution is to simply rely on the analyst estimates. I haven’t explored that very much.