Key metric per industry.

[14value]

While reading the "ideas with catalysts" thread, oddballstocks brought up the important point of having the right metrics to value specific sectors.  I have a list of ratios I found in a white paper for some sectors and would appreciate any comments.  I am trying to better my valuation process.

Thank You,

 

Autos and housing:  rel.yield, p/b, p/fcf, ev/ebitda

basic materials:  rel.yield, p/b, p/fcf, ev/ebitda

consumer staples:  p/e, rel.yield, p/fcf

energy:  rel.yield, p/b, p/fcf, ev/ebitda

financial services:  p/e, peg, p/b, p/fcf

healthcare:  p/e, peg, rel.yield, p/fcf, ev/ebitda

industrial goods and services:  p/e, rel.yield, p/b, p/fcf, ev/ebitda

leisure:  p/s, ev/ebitda

retailing:  p/e, p/fcf

technology:  p/e, peg, rel.yield, ev/ebitda

transportation:  p/e, rel.yield, p/b, p/fcf, ev/ebitda

telco:  p/e, peg, rel.yield,

utilities:  p/e, p/s, p/b, ev/ebitda

[frith2012]

Someone we all know made the comment along the way to value things in a per unit basis.  For example, if you're looking at an oil company, how much are you paying per barrell of oil.  In that context, here are a couple ideas:

 

retail: sales/sf

Oil and gas e&p companies: price/barrell or boe

real estate companies: price per acre (you can apply this to homebuilders, real estate developers, retailers with land holdings, etc...)

 

[mikazo]

"Operating ratio" is popular in the railroad industry:

 

http://www.investopedia.com/terms/o/operatingratio.asp

[LC]

combined ratio in the p&c industry

[netnet]

This is a good topic, which makes me think that each stock on the investment board should begin with what is the key ratio for this company!

 

By the way, ROIC and RD expenses/revenue for tech companies.  Owner earning/EV for all companies.  for oil companies exploration cost per incremental barrel of oil.

[Phaceliacapital]

SaaS companies: Revenue???

 

Kidding ofc

[ItsAValueTrap]

energy:  rel.yield, p/b, p/fcf, ev/ebitda

They have depleting assets.  I would use discounted cash flow (DCF), or market/liquidation value of all their assets.  DCF and market value are linked... in the DCF model, the fudge factor is the discount rate that you use (and commodity pricing assumptions).  The discount rate varies because nobody knows what the right discount rate is.

 

Unfortunately that metric is difficult for investors to verify.  Yes, many energy companies provide a PV-10 value.  However, there are ways to fudge that number due to uncertainties in resource extraction.  Look at all the people who believed ATPG's management...

 

2- Basically, here's how I would value an energy company:

a- Value of their assets.  Buyers of these assets use DCF.  They do not use $/barrel.

b- Usually, these companies are incredibly difficult to value.  When they sell assets, they will open up a data room for buyers to do due diligence.  Smart buyers will do that level of due diligence.  Few stock market investors do that level of DD (and few even have the access to data and expertise to do it).

c- Does the company deserve a premium or discount for the quality of management?  Quality of management is tied to the company's cost structure.  Being the low-cost producer is the only advantage you can get in the industry.  Those who can explore cheaply deserve to trade at a higher value.

 

[ItsAValueTrap]

Metals:

 

Here's what I learned from reading mine engineering textbooks.

1- For any given deposit, engineers will have wildly differing opinions as to the project's economics.  And this is assuming that you are dealing with honest people.  In the stock market, you are mostly dealing with promoters and other scum.

 

2- Valuing land is brutally difficult.  Textbooks may give you 3 different ways of valuing land.  Nobody really knows how to value land precisely.

 

3- For operating mines, you use DCF.  In a DCF model, it is standard to plug in the 3-year trailing average for the commodity price.  An exception can be made for commodities with liquid futures markets.

 

Of course you can plug in your own commodity price.  But if everybody uses the 3-year trailing average as a baseline, it makes projects easy to compare.

 

In practice, most technical reports are inflated so don't take them too seriously.  One of the tricks is to not use the 3-year trailing average so you can plug in a higher commodity price (*this doesn't happen in the gold and iron ore markets that often if the company is trying to pander to analysts, who are permabears on their commodity).

 

---

If a very sophisticated buyer (like Teck in the old days) does their DD and purchases a stake in a mining company or one of its assets, then you can infer the market value.

 

In practice, I think that this is the best easy method to value some of the mining companies out there.  But only if the buyer did their due diligence (e.g. having an experienced engineering staff is a prerequisite).

[Guest deepValue]

Get to know some hedge funders. They like things like value per key (for lodging REITs), value per subscriber, value per barrel, etc. These sorts of things are useful when comparing peers.

 

As for key valuation metrics, it usually takes something a little more custom-fit than slapping an EV/EBITDA multiple on a stock. Figure out whatever is important in the company and value that asset the way a businessman would. If you take that mentality, instead of "this is a bank, I should pay a given multiple of book value," then you'll tend to get better insights than if you analyze companies the same way everyone else is doing it.

[oddballstocks]

Get to know some hedge funders. They like things like value per key (for lodging REITs), value per subscriber, value per barrel, etc. These sorts of things are useful when comparing peers.

 

As for key valuation metrics, it usually takes something a little more custom-fit than slapping an EV/EBITDA multiple on a stock. Figure out whatever is important in the company and value that asset the way a businessman would. If you take that mentality, instead of "this is a bank, I should pay a given multiple of book value," then you'll tend to get better insights than if you analyze companies the same way everyone else is doing it.

 

I should note, the metric I've seen for banks is 11x post cost take out earnings.  This was in a letter from a hedge fund that looked at all of the bank mergers in the past decade and concluded that almost all of them were almost exactly at that multiple.  So that's the key metric for a bank's ultimate value.  The fudge factor here is how much savings a banker thinks they can achieve.  I like P/TBV and a multiple on PTPP, using those together is a short hand for getting to the 11x multiple.