Simple but important question. How do you guys value prospective investments?
What I look for is, I want my investment returned in 10 years or less. (avg return = 10%+). So I just project FCF for up to 10 years, and add it to cash + LT ivnestments and subtract debt, if this equals price…it’s undervalued to me…
I never really understood EBITDA/Earnings multiples so never bothered learning those. Which also results in me not being able to understand when other people use earnings multiples…
I use both cash and earnings but not EBITDA (pretty stupid measure to me unless you’re doing PE or something when you want to find out how much interest you can slap on the thing)…for a lot of mature companies, the FCF = Earnings as depreciation equals maintainance capex…
I will add that I use non gaap earnings in certain situations when it makes sense (i.e. TEVA) when there are lots of one time charges (cash charges and non cash) that are not going to repeated for that same stream of revenues…
The easy answer: it depends on the investment!!! Different investments require different methods and different ways of normalizing. The methods I use most frequently are P/E and EV/EBITDA; DCF is best but only when you have clarity and trust on the company’s asset and project values and a sense of what the correct beta/discount rate is. (Of course, even for P/E, you should be thoughtful about capital structure considerations but a lot of analysts don’t look at this stuff, which is their mistake)
Also, I don’t understand what you mean when you “project FCF for up to 10 years, and add it to cash etc etc” and it equals price. I assume you’re talking about running a DCF and backing out net cash, but your description is lacking several key variables which makes me concerned about whether you are doing your valuation correctly. First, are you using levered or unlevered free cash flows? Secondly, what discount rate are you using, and why should this relate to your desired IRR? If anything you should be matching the discount rate to the cost of capital for the firm/project, not your desired IRR. Once you do this calculation, you can then compare this implied equity value (on a per-share basis) and calculate your expected returns to see how that compares to your desired IRR. This is how you should be thinking about whether or not the equity is a good investment.
If this is a topic that interests you, then you must own a copy of McKinsey’s “Valuation.”
Lastly, not to sound daft, but is this stuff covered on CFA Level III? I see so many examples of people building valuation models using the totally incorrect cash flows, betas, discount rates and so forth, even from “experienced” sell-side and buy-side analysts in their models.
One prime reason someone may want to use EV/EBITDA is to normalize for companies with different capital structures. By looking are earnings before interest, you’re essentially looking at potential cash flows to the firm and not just to equity holders. Also, there are a lot of companies whose EPS are so microscopically low because of leverage or high share count, but actually have respectable operating income before interest expense, so EV/EBITDA still provides a good proxy of valuation of the firm.
Also, FCF may not be a good proxy for earnings unless the firm has no leverage, or unless you’ve unlevered and then re-levered the beta in calculating your interest rate so that you are using an industry average capital structure for the potential firm you’re trying to value. You also forgot to mention growth capex and change in working capital in your FCF “formula” above.
I’m not lookiing for help or suggestions on valuation, my only question was what methods other posters use for valuation and what logic is there behind using EBITDA. That was my only question…“How do you guys value prospective investments?”
I thought I answered this question in two of my posts above. I also pointed out some criticism regarding what I felt was your incomplete / incorrect use of an FCF-based valuation (why you seem defensive about this, I don’t know). So, I get that this isn’t exactly what you were looking for (actually, it seems like I offered more than what you asked, which I think most people would appreciate – especially folks on an internet forum with an educational bias).
Therefore, please help me understand how you thought my posts didn’t address your question or somehow were otherwise not useful or not accurate. Not trying to sound snippy here, but I really don’t get what you say I missed about your post. I think I understood it as well as anybody could have.
I find that service industry stocks and other asset-light businesses tend to be best value on FCF and price or EV to sales multiples. There are little or no hard assets, so you are valuing the cash flow stream. Since either net earnings or EBIT can be viewed as a reasonable proxy for cash flow for most companies, you can basically forecast either one based on X factors and then apply a reasonable multiple to it. The rule of thumb is that a price to sales multiple of 1 should correlate roughly to a 10% net income yield on sales. If you have $100mm of earnings, the stock should trade for a billion. If it’s $900mm of market cap, that does not make the stock cheap. If it’s $300mm of market cap but everyone is butt hurt over some temporary issue, that is probably a very cheap stock depending on your horizon (presumably earnings would be in the toilet, so you would be trying to understand how they would normalize over time and if they could get back to 100mm depending on the cycle and structural factors).
For asset heavy companies, I typically look at the value of the assets and the predicted normalized return on those assets or the equity in the compnay (ROA or ROE). My personal feeling is that a book value of 1x is justified by an ROE of 15%, but it really depends on the nature of the business and capex requirements. 15% gives the company some decent cushion to reinvest and still earn their cost of capital. Obviously I would not want to pay 1x book for something with a 15% ROE, and it’s worth pointing out that 15% is kind of my own arbitrary metric (because I don’t bother to calculate WACC for most things and just apply a 10% cost of capital across the board for most companies).
Generally speaking though, the best metric is FCF and what either a strategic or a financial buyer would pay for that stream of FCF.
On the note of strategic, a good operating business will often go out at a healthy multiple of gross profit, not even including any synergies that can be derived by wiping out the middle of the P&L (although synergies are often factored into the price). I’ve had a lot of successful recommendations in underperforming stocks with low EV / gross profit “coverage” (multiple) that either got fixed or sold – though of course that has to be a quality asset to be considered a value. Nobody looks at gross profit because there are a lot of other real costs that come out, but the fact is a lot of costs are redundant to a strategic and they’re often looking at “ball park what does this business do for us” type of metrics during the acquisition process.
I just found one today that is trading for 1x gross profit with an improving backlog and high inside ownership – they could pick up the phone tomorrow and sell the company for AT LEAST 2 or 3x what it’s trading for today. Business either gets fixed or sold, no way the stock bleeds forever with insiders that own over 40% of the company.
bromion, very good post overall and it sounds like you and i are on mostly the same page, despite the fact that i think the 10% wacc is a bit arbitrary (though my guess is that this isn’t too far off from the wacc for an average publicly traded company)…anyway, i can think of one exception to your gross profit metric where the business may not get fixed or sold. i’ve seen companies with entrenched management or boards with supervoting rights or significant ownership trade at very low multiples, since investors don’t see an event where the company gets taken out or liquidated. now i can’t think of any off-hand that were trading at SUCH a low multiple, but there are definitely firms out there that routinely trade at very low multiples (tootsie roll for example), but where the main catalyst to unlock shareholder value would be a transfer of ownership. in these cases, management is just content on gaining some steady returns on its cash and doesn’t want to give up their empire or go through the trouble to fix it. the only reason a firm such as TR doesn’t trade lower is because there is an effective price floor – there are speculators that the owning family is getting progressively old and will eventually decide to sell it at some point, even if there is no imminent event (such as death in the family, god forbid) that would trigger the exit catalyst.
lastly, thanks for your e-mail this morning. i’d be eager to continue our conversation more regularly. is that the best e-mail address to reach you at or is there another? i’m still at firstname.lastname@example.org.
I don’t understand, bromion. Unless you are expecting 0% ROE or imminent liquidation or have bank assets that are questionably valued, shouldn’t you always be looking for at least 1x BV? Companies whose assets can be purchased at less than 1x BV are prime targets for either corporate raiders or mergers.
By the way, excellent discussion question, Palantir. Thanks for asking it.
Bromion, in terms of why you’d need 15% ROE to justify 1xBV, is the 15% ROE to justify 1x BV just your way of ensuring a margin of safety? 10% WACC assumption and 15% ROE implies a 5% spread over and above market risk premium, so you have something like 5% margin of safety.
Regarding Palantir’s question, I think comparing the growth in BVPS vs ROE might be useful to see what’s happening to the P/BV multiple. ROE is more about the fundamental performance of the business, whereas BVPS incorporates the market sentiment portion of things.
Yes, the numbers are a bit arbitrary (10% WACC is unquestionably arbitrary but my guess is that is roughly about right for most companies). The reason I do it that way is because I have tear sheets on companies with up to 20 years of publicly available data – basically quickly looking for anything that jumps out on the long or short side and digging in deeper from there. I don’t need to calculate the WACC for anything unless it becomes highly relevant (which it does sometimes – you would be surprised how many companies lack an understanding of their cost of capital (including the CFO) and make stupid acquisitions that can never pencil mathematically). Time management is one of the most important factors in successful investment returns and is widely underrated by professionals in the industry IMO (although if that’s true, why do I spend so much time on AF? Fail).
15% ROE – also somewhat arbitrary and it does incorporate some margin of safety (although margin of safety is a term that gets thrown around all the time and there are many types of margin of safety, just wanted to clarify my comment with that disclaimer – the best margin of safety is buying into growing annuity cash flow streams with barriers to entry at reasonable current prices). Basically to me it seems that capitalism is very good at grinding out the lowest per unit economics it can. The world is very competitive. Some asset that can earn a 15% ROE over time is doing well and should be worth at least it’s book value in the majority of cases – someone is paying that asset to add value in some way. If they earn 5% is that worth book? Maybe in some case (although much harder to see from a very high level), but generally, a firm’s value proposition is either strengthening or weakening over time (a lot of assets exist simply because nothing better has come along to kill them off yet), so if the value proposition is pretty weak already, I assume in most cases it will not get stronger over time, so why would I pay book for that (which doesn’t even factor in ongoing and future capex needs)?
I do look at BVPS growth over time as well (“this company creates value”), especially TBV in the case of acqusitive companies (most acquisitions don’t add that much value apparently from what I’ve seen, and synergies tend to be illusive). The ROE metric is more of a normalized output on the assets (net obviously) – ROA is also useful for the same reason. To me the two metrics serve slightly different purposes but are both useful (and have overlap).
Numi is correct in its EBITDA assessment and yes it does “neturalize” the capital structure…its a PE thing…i normally use the ROA to get a sense of return on captial…either way you put it and anyway you want to go about it, if you only earn 1% on your assets debt or no debt that business is kinda crappy right?
Teva has 3 main 1 time charges related to acquisitions 1) Inventory step up 2) restructuring 3) expensing of research in development/purchase intangibles…Buffet (WEB?) likes Teva? i am not aware and I stalk warren more than i stalk Kim…
BVPS can be gamed by buying back own stock on the cheap (see Lowes Corp’s amazing rise in BVPS). Also, for financial/insurance beware of using BVPS without taking into account AOCI…
Yes, drop me a line any time – email@example.com is the best way to reach me (anyone else should feel free to drop a line as well).
I completely agree with your point – I have never looked at TR, but there is no question that some companies are public / exist to pay their board & management egregious compensation. I looked at one earlier this year but the name eludes me (starts with a T) – there was a block selling at liquidation value (defined as current assets minus total liabilities in this case) and I called the company to see if they had a clue or would sell. The change in control provisions were completely exploitative to shareholders (like $8 or 10mm on a $50mm or whatever market cap company) and the top 3 people at company were making in excess of $2.5mm a year (holy shit batman). It was a tiny company that sold needles, uranium butt pellets to treat rectal cancer (there is a technical name but I forgot – butt pellets is a good description though) and other consumable medical widgets to a niche market (basically doesn’t even need to exist as a stand alone entity – should be sold to a better owner). I asked the CEO point blank why the company was public / hadn’t been sold and she said, “We’re just waiting for a phone call but we’ve never gotten one” – yeah bullshit, public companies get calls all the time, especially at liquidation value. Company exists to bilk shareholders so stock will not reflect fair value any time soon, pass.
If I recall correctly that’s on CFA L2. The thing about the curriculum, in my opinion, is that there is so many “shaky” valuation stuff that most candidates might miss the more relevant stuff and all the small print that comes with it. Plus, CFA writing is sometimes pretty dry. Sometimes they don’t focus enough on the why and, when they do, we may not think too hard about it because we gotta learn that LOS fast since there are still 300 others to go. That may hurt long term recall. I mean, you need good understanding to pass the exam, but it is different to get a good understanding than to calmly test every little bit of info they have to see what can work in your own valuations.
L3 is much more focused on asset classes and how they interact. There is very little, if any, quantitative valuation stuff for specific securities - there’s a lot of stuff that you can use indirectly for it though. I think whatever appears in practice questions is mostly asking about some other thing and just assuming you remember your L2 stuff. Still, I think L3 has the most useful content by far. It is worth a read even if you don’t intend to take the exam, unless you have no interest in portfolio management. L3 has the same issues though - some of the more interesting info shows in a single line in the last paragraph of a topic that doesn’t seem that relevant to begin with.
I think Palantir method is legit though. It reminds me of Bruce Greenwald’s valuation methods, which seem pretty sensible when you read his arguments. Basically, he does not try to forecast what many may view as not possible to forecast with any degree of decent accuracy. If I recall correctly, Greenwald just says screw it and multiplies current cash flow by 10, and that approach has actually beat the market in the past. There’s more to it, but not that much more. His book is “Value Investing: from Graham to Buffet and beyond” or some title very close to that.
Yes, it does let a lot of info out of the equation. The point is to decide if that extra info make for better or worse overall valuations. Too much info (especially on thesis for growth) may lead to overconfidence. I don’t really have an opinion here, except that this may be a valid approach.
Edit: Wow. There’s like a dozen comments that appeared below that one. Gonna read the thread now.
Edit 2: OK. This thread is amazing! I’m sorry I went off-topic with the CFA stuff. On Greenwald’s ideas, he only applies that to stocks that are compatible with his value criteria, so it’s not as crazy as I may made it sound.
I laughed out loud when I read this. Wow, $50mm is seriously micro-cap and you wonder why anybody who has done some iota of homework has invested in this company. Was this company’s malfeasance that obvious, and if so, were there other posts on the internet about how crappy it is? Usually, these types of things don’t go unnoticed. In any case, let me know if you remember the name of the company. That is really hilarious.