"Firstly, as you probably know by now, I am not prone to thinking in terms of EBITDA. I particularly have a problem with it when there are substantial R&D costs being capitalised. These capitalised costs only have a hope of being reflected at the bottom line via the "A". ARB has been hitting its R&D lever increasingly hard, and its capitalised component has been increasingly exceeding its amortisation expense. As such, EBITDA is particulalry non-conservative, in this regard."
@MarsC,
Sure, EV/EBITDA as a valuation measure, is a crude one, not much better in terms of its refinement as "P/E".
Indeed, as you know, I've on occasion acknowledged that not all EBITDAs are made equal (for the very reasons you've cited, and more) and that they need to be given a good sanitising in order to be able to be legitimately compared from one business to another.
And, clearly, an EV/EBITDA of ARB is a totally different quality EV/EBITDA to a more capital-intensive, business with highly variable financial performance, such as, say BSL, or BHP.
Like you, I am also a Free Cash Flow Yield investor.
Trouble is, deriving Free Cash Flow requires the ability to discern what stay-in-business capex is, and well as having an understanding of the investment required in working capital (which, in my experience, is often a material quantum, yet is often not considered by investors[*]).
These required insights don't come without some analytical effort (and a dose of "good engineering assumption", to boot).
So, yes, Free Cash Flow is the "pure" valuation approach.
But the cost of that purity is a degree of effort and research.
By contrast P/E and EV/EBITDA can be calculated solely on numbers that are reported.
Which is why the default valuation "ready reckoners" tend to be a combination of P/E and EV/EBITDA (the former because it captures non-operating items such as tax and interest expense, as well as accounting treatment in the case of D and A, and the latter because it captures major differences in capital structures)
As for comparing with cash earning yields with low-risk alternatives, yes that is theoretically what the market tries to do to.
But the problem with that is that there are a number of indeterminate moving parts, not least of which is guessing the right level of the thing that is being used as a reference point (i.e., low-risk, or risk-free, interest rates), which is subject to change over time.
Not just that, but when we compare the cash earnings yield for ARB with, say, the 10-year government bond yield, we are effectively comparing something with a
growing income stream (hence, increasing cash earnings yield) with something that merely has a
fixed income stream.
So, sure ARB's 3.0% to 3.5% Free Cash Flow yield today might look skinny compared to the prevailing rates offered by low-risk alternatives. But when viewed in terms of a prospective five-year investment time horizon, the picture is likely to look somewhat different,
ceteris paribus.
[*] As case in point, ARB has invested more than $100m in Stock and Receivables over the past decade. For context, total investment in PP&E over that period amounted to $200m, so the working capital demands for ARB are a significant factor in deriving FCF. For greater context, the working capital investment ARB has to make is equivalent to a not-immaterial 25% of Operating Cash Flow.
(20min delay)