What determines what multiple an industry trades off of?

HireUp212's picture
Rank: Orangutan | 339

I know we all are told that this industry trades off of FV/EBITDA multiples or that one trades off of P/E multiples...but what actually determines what the appropriate valuation multiple(s) for an industry are. Why do certain industries trade primarily off of this or that multiple as opposed to some other one?

Comments (18)

Sep 12, 2010

It can be the result of various factors, but generally speaking it is a function of an embedded expectation in future growth and/or relatively strong/weak financial margins.

Sep 12, 2010

I don't follow - why would growth and margins determine whether I should look at EBITDA multiples as opposed to P/E multiples or vice versa? I'd expect both to be higher, all else constant, for higher-growth and more profitable companies.

Sep 12, 2010

EV/EBITDA multiples gives you a sense of the economic spread being created by the assets of a company whereas the P/E multiple gives you a sense for the value being generated both from the underlying economics of the business (revenues, COGS, operating margins) but also the effect of leverage. If a business is highly profitable, it can carry a larger debt load which translates into greater returns to equity holders - therefore, the price of the stock goes up because investors have a demand for higher returns at any given level of risk. So, ratios are a function both of the underlying dynamics of the business and also market perceptions of the value of the business. The reason we value businesses using EBIT or EBITDA multiples is to look at the value of the company divorced from the capital structure, which is prone to change.

    • 1
Sep 12, 2010

i believe it's whatever consensus in the analyst community uses and looks at. i imagine it's just convention

Sep 12, 2010

I thought you were questioning what drives the individual multiple range within one industry relative to other industries. Apparently I was mistaken.

If you are asking why certain industries use certain multiple methodologies, just think about the actual statistic that is driving that multiple.

Why would a company that hasn't produced consistent earnings be valued based on P/E multiples, being that it may be trading on negative earnings? Instead, the more accurate valuation metric may well be revenue. Or, why would a company that operates using significant amounts of tangible and depreciating assets be valued based on earnings, being that such earnings are so significantly affected by non-cash depreciation expenses? EBITDA would probably be the more applicable metric driving value.

Sep 12, 2010

The market determines it. Its not a rule of thumb set by a governing body that everyone abides by. But rather it an observation of a prevailing trend.

EV / EBITDA is generally impacted by a few things. The most obvious one being future growth expectations. How would the future growth expectations of a pure play rotary telephone manufacturer differ from that of a smart phone manufacturer? The rotary phone manufacturer would trade at maybe 2x EBITDA while the smart phone manufacturer trades at 16x EBITDA. Im making these numbers up, btw. Similarly, certain industries have distinct characteristics which don't change very often so one can observe that the industry generally trades within a certain band.

So take a large beer brewer for example. Miller, Anheiser, etc... sure there are industry trends like a shift towards lighter/low-carb beers or the micro-brewery wave that came up some years back, but generally speaking, the overall growth the industry is pretty predictable. Which leads us to another factor driving valuation multiples... predictability of performance or risk. Your shown two smart phone companies. One has a management team mixed from various electronics background.. a Sharp television executive, an LG household appliances executive, etc... the other is an all-star team of the who's who of cell phones. The guy who lead the development of the iPhone at apple, the CFO from RIMM, the head of business development from some other hugely successful wireless company. All else being equal, they have the same exact projections, business model, etc... which would you pay more/less for? You'd likely be willing to pay say 8x EBITDA for the NonBadAss Co vs. 11x EBITDA for WirelssBadAss Co... because WirelessBadAss Co. is more likely to deliver on their business model. They have the know-how, expertise, experience and track record. Its less risky to invest in a bunch of hugely successful industry veterans.

Going back to the beer example. There overall industry's growth horizon is pretty predictable. Within the industry, there are trends... so you've got the low-carb movement. You've got the flavored drinks that came out 10-15 years back that were a hugely successful (e.g. Mike's Hard Lemonade, Smirnoff Ice, etc..), you've got the energy beers that started coming out before many local legislators shot them down. So within a given industry, each company is somewhere within that band... maybe they are the industry leader in flavored beverages, maybe they're the leader in low-carb beers... maybe they are the laggard, etc...

    • 1
Jun 15, 2014

All of these answers are either lazy or off base. Industries trade off different multiples because there are different measures that best capture the actual productivity and earnings of the companies within them. For example, look at consumer retail vs oil and gas. Completely different industries. In consumer retail I imagine there's a pretty consistent accounts receivable or payable balance based on inventory levels, which will still be factored into net income because the transactions are completed. There is also a relatively small DD&A balance which will not create a massively different cash flow number. Because the transactions driving the receivables are important, I would guess that industry is going to be driven off a P/E, or something that incorporates those aspects. Oil and gas on the other hand, has massive depreciation and depletion balances, very little working capital swings, material maintenance capex, etc. That all means that an earnings number is going to mean essentially nothing for valuation. Because of that, oil and gas stocks should be viewed in the context of cash flow multiples. Just a simple example, but hope this shines some color on this for anyone else that may come across this 4 yr old thread.

If someone reads this and thinks I'm off base, chime in but don't tell me I'm wrong with a message like what is above, completely lacking substantive counterpoints.

"Now that I think about it...I never actually know when you guys are here in the office. Kind of like furniture...I just assume."

    • 3
Jun 16, 2014

I'm glad that the guy who was himself looking for advice on how to get a job in finance has in the course of a year developed such a deep and profound expertise in valuation and the markets as to call out the same people who's advice he was soliciting as being lazy and intellectually bankrupt.

    • 1
Jun 15, 2014

I've looked into and thought a lot about this topic and I like to take an academic approach here. Higher free cash flows (and growth) and lower risk are what equity and debt holders look for in a firm broadly speaking (e.g. DCF's discount FCF's by the firm's WACC which is a measure of risk). However, instead of doing a full blown DCF and trying to make assumptions for all its components for the next 5 years, what if we just looked at what what multiple the market valuation (EV) was of next year's FCF? That would simplify things, but we would need to make sure that all our comps had similar risk and growth profiles. To see why look at the definition of FCF:

FCF=EBIT(1-t)+D&A-Capex-Increases in NWC

In a DCF, we discount these by the WACC minus the growth rate so if we want to use a FCF/EV multiple to compare firms, they should have similar WACC and growth profiles.

Of course FCF itself is highly variable and can be tough to predict even for one year (who really knows what's going on with NWC on a year-to-year basis?) so investors often simplify things further. If we look at an EV/EBIT multiple across firms for example then we are implicitly assuming that EV/D&A, EV/Capex. and EV/(Change in NWC) are the same for each comp. An EV/Sales multiple would go further, by assuming that each firm has similar EBIT margins.

What does all this mean regarding what multiple an industry trades off of? If you can't project a line item in the future with reasonable accuracy and/or if you're fairly confident that your comps are financially similar enough to justify a multiple's implicit assumptions then that may warrant doing multiples analysis. Which multiple you use depends on which assumptions you're willing to make. Maybe that EBIT margin assumption is too aggressive. I might stick with an EBIT multiple instead of a sales multiple in that case. Note that a lot of times people subjectively account for these assumptions (i.e. one firm has a better growth profile than its peers therefore it should trade at a higher multiple).

One specific takeaway I've found from thinking about this is that EBIT multiples are generally much better than EBITDA multiples. An EBIT multiple assumes that each comp's EV/D&A and EV/Capex are equal (a large assumption I agree). But, D&A is a real expense (you have to replace those fixed cost assets at some point) even if its non-cash and is kind of a smoothed-out version of Capex. Because of that, I think the real assumption you make with an EBIT multiple is just that each firm's D&A and Capex will roughly equal each other from year to year which is much easier to swallow in my eyes. An EBITDA multiple on the other hand, just assumes that EV/Capex will be similar across the board which is almost never the case.

Jun 16, 2014

Couldn't help but notice that there are a lot of odd answers in this thread. That or our academic / professional training really is that varied...

Think this response from above mostly hit the point - "The market determines it. Its not a rule of thumb set by a governing body that everyone abides by. But rather it an observation of a prevailing trend."
A bit more nuanced in that the decision is generally formed on the basis of the most representative financial metric for the industry

I think we need to take a step back and just answer OP's question - what determines how companies are valued in an industry? We can spew nits and compare boner sizes on adjustments / differences between various financial metrics but at its core its really how ER / investors wants to look at the company and what they believe is the most relevant metric. Industry like semis for example - almost everyone is valued off P/E as investors just care about earnings growth and things like capital structure decisions which flow down to EPS are part of running the company (like when the market reacts positively to an Accretive debt financed deal like LSI / Avago). Profitable / established software companies are valued off FCF since the structure of most contracts results in deferred revenue that isn't captured in traditional earnings and EBITDA. Profitable internet companies are slaves to EBITDA multiples (over P/E since investors care more about cash flow) with a sprinkling of DCFs and high growth ones which no profitability trade on revenue, the only metric they have.

Have no experience to speak for the non tech industries, but I would imagine thematically similar.

Jun 16, 2014

P/E multiples are a rearrangement of the components of the Gordon growth model. A P/E multiple equates to 1/(k-g).

EV/EBITDA applies the same concept to produce enterprise value.

The connection to the GGM is often forgotten and you see multiples used for businesses where GGM assumptions aren't met eg people apply multiples where a business is in a period high growth, which will be followed by a longer period of more mature growth. Conceptually incorrect, but pretty common.

Jun 16, 2014

A lot of long-winded (albeit informative) answers in this thread that don't answer the OP's question lol.

As someone above alluded to, I think the very simple answer is that which multiples are used for a given company depends on what best captures the fundamental performance/health/growth prospects of the company, with the assumption being that this fundamental information somehow translates to market performance. This depends on what kind of business a company is involved in and the specific nature of the company.

For example, you wouldn't use a cash flow multiple to assess a company in which those particular cash flows in question are often negative, near-zero, or haphazard, maybe because it's early stage for instance.

Getting philosophical here, think of the true fundamental status of a company (which is ultimately unknowable in the truest sense except maybe by senior management) as Plato's Forms, the ideals that are imperfectly reflected in our perception of reality, which in this case would be public filings, market performance, public opinion, conferences, etc. Not all lenses or reflections are equally imperfect though, and your goal is to figure out what most closely captures that fundamental ideal reality that is not directly visible.

Jun 16, 2014

Aren't we all supposed to be born with knowledge of the Platonic forms (eg a right angled triangle and Pythagoras's theorem deduced from that knowledge)?

Jun 16, 2014
SSits:

Aren't we all supposed to be born with knowledge of the Platonic forms (eg a right angled triangle and Pythagoras's theorem deduced from that knowledge)?

That's true, but you have to work to "remember" it. Personally, I was "recollecting" my innate knowledge of P/E multiples when I was 3 years old and deducing from that knowledge which companies were rich or cheap.

Jun 16, 2014

@"Marcus_Halberstram" is on point, the market determines each company's (and in turn, each industry's) multiple based on expectations. if they think old tech will do poorly (HPQ, INTC, IBM, etc), they will trade them down and the multiple will contract, vice versa for optimistic plays.

the interesting thing about this is a lower multiple company/industry is usually a LOWER risk play than a higher multiple one. of course there are exceptions (relative level of valuations, credit quality, balance sheet strength, dividend characteristics, etc.) but when we allocate capital to companies, we always prefer lower valuation over higher valuation.

one exception where I'd agree with a higher valuation is the "grandma" stocks. I call them this because it seems every grandma in the south owns them: stocks like KO, MCD, PG, GE, PEP, HSY, JNJ, etc all trade at a PE higher than the market and usually do, but they still create value and have been great companies to own for decades and probably for decades to come.

Jun 16, 2014

In light of people's complaints about all the lengthy responses here's the abbreviated version of how I see things. The most reliable multiples are also the ones that require the most assumptions and projections while the least reliable are the ones that require the most assumptions. Take the following common types of (firm not equity) multiples:

Volume (e.g. user base for tech companies)
Sales
EBITDA
EBIT
EBIT less capex
FCF

FCF is very reliable as a multiple provided you can project it accurately. Volume on the other hand doesn't take into account things like profitablility or fixed costs, but it is much easier to project. That's why you see harder to value growth spaces like tech using volume and sales multiples (ok also because they're often not cash flow positive yet) and more mature industries that have less diversity in people's opinions on them using line items from further down the income statement and/or incorporating cash flow items.

Also, it's definitely true that the investors of each industry dictate what the constituents trade off of, but that's more of a trading mentality than an investing one. The whole point of all of this is to see if you can find something that you think the market isn't thinking about correctly. Maybe that comes from a divergent view on earnings for example, but it's just as plausible to say that a given company shouldn't be valued on a P/E basis in the first place.

Jun 17, 2014

Growth
Margins
Risk
Cash Conversion
Quality of Earnings

Jun 17, 2014
Comment