XO: A Valuation Model

April 21st, 2009 by · 10 Comments

[This is an anonymous guest post by Homer. The author is a denizen of the the message boards on InvestorVillage and is long XOHO. Do you want a turn at the microphone?]

The Basic Model:

The VALUE of the common equity of a firm is the value of the residual claim on the assets. For most ongoing concerns this is equal to the present value of future cash flows minus net liabilities. Thus, we need to look at three variables:

  1. Net Liabilities (i.e. all claims on the firm minus cash).
  2. Estimated future cash flows (typically some sort of EBITDA measure)
  3. Discount rate for future cash flows. This is another way of saying “multiple”. It is depended on interest rates, overall market conditions, and firm specific uncertainty of cash flows.

Furthermore, empirical research has demonstrated that financial structure also matters (this is not an obvious one, under perfect markets assumptions financial structure should not matter... turns out markets are not perfect). Thus we have a fourth variable that affects the value of the common stock:

  1. Financial structure of the firm (i.e. leverage and liquidity considerations). This assumes market frictions that prevent costless adjustments of the balance sheet. By the way... leverage is not always a bad thing.

Finally, in conditions of distress, one must also consider the optionality value of equity. This is because even when the value of the residual claim of assets is zero (because the net liabilities are greater than the present value of expected future cash flows) the common stock will still have some value because there is a probability that the residual claim might turn positive in the future. The further out of the money the equity is or the shorter the time left for this residual claim to become positive, the lower the optionality value of the stock (Black Scholes). In “normal” companies (i.e. firms where this option is well in the money) the optionality value of equity is negligible.

Now, the PRICE of the common stock is equal to the VALUE of the common stock plus an error term.. This error term - or noise- is related to the market liquidity of the common stock security. The standard deviation of the distribution of the error is positively correlated to lack of liquidity. Thus, a large buyer or seller of an illiquid stock can temporarily move the price quite a bit. This noise should, however, eventually go away.

The problem with this model is that Net Liabilities is the only known variable (and sometimes not even that - as bankruptcy courts have taught us, claims on assets can easily multiply). So everything is estimated. And different variables cannot always be separated in a very clean manner. But the model does give us is an indication of how changes in each of the independent variables affect the dependent variable (i.e. stock price).

Looking at XO one variable at a time:

  1. Net liabilities - The number seems to be 660 million. And growing every quarter. But not by much.
  2. Estimated future cash flows - This is where industry experts come in. Looks like 125 million EBITDA for 09 is quite reasonable (if not conservative). Many think starting in 2010 EBITDA might really take off. My own estimates are 125-140 (2009), 135-160 (2010).
  3. Multiple - Industry multiple seems to be close to 6 (using TWTC of CCOI as references… these are relatively deleveraged players… reasonable visibility into future cash flows). The "Icahn" factor also comes in here. Definitively a discount but probably not as much as many think. Ultimately XO is an acquisition candidate and the value to others is ex-Icahn. The discount factor for Abovenet for example (which for years was unable to provide any financials at all) was never that high.
  4. In my opinion not real liquidity issues - Next year’s preferreds should be easily taken care of via a senior secured deal. Fairly cheaply too (even in these credit markets).

Does the optionality value of equity come into play here? Definitively. A quick calculation tells us the equity is not that much in the money (125 million times a multiple of 5.5). How much is this optionality value? Another quick calculation tells us this value should be significant. An at-the-money three year option on 660 million using an implied volatility of 10% -which is incredibly conservative-is worth 110 million. The current market cap of the common is around 30 million. The numbers are quite sensitive to assumptions but it is safe to say the optionality value of the common should be significant. LVLT’s current valuation-which is a multiple of close to 7-, for example, reflects the fact that a good chunk of the value of its common stock lies its optionality (i.e. the stock is probably out of the money when looking at 2009 EBITDA vs net liabilities, but a multi- year option on the upside potential of LVLT’s assets and business is in fact worth a lot of money). The longer they push out maturities the higher the option value.

So if optionality value of the stock is supposed to be high how come XO trades for less than 20 cents? My best argument is for the error term in the model. I am convinced the lack of institutional investors, combined with a pre-crisis dominance of hedge fund investors caused the demise of the stock. The dumping into an illiquid market has annihilated the stock price (and still is). Should eventually correct itself (the distribution of the error should have a mean of zero).

For those that made it through the end … feel free to ask any questions.

Categories: Financials · Guest Posts

Join the Discussion!

10 Comments So Far

  • Ronin says:

    Very interesting post Homer, thanks. I agree XO has some interesting assets, but don’t buy a few of the premises in the argument to get me interesting in owning the stock.

    For example, I don’t think the $1bn in preferreds XO has can be easily refinanced in 2010 with a senior secured deal. XO will need to keep at least $200mm of cash liquidity given they are FCF- or barely FCF+ currently, so that still leaves ~$800mm of preferred overhanging the common. Using your $125mm of EBITDA figure, that is 6.4x leverage, which is extremely doubtful they could get done, even if they did a bank and bond deal.

    Second, I agree the common is trading on option value, but I’d suggest the better way to value that option is not via a black scholes model using a market vol figure (which assumes a normal distribution of outcomes), but rather a binomial tree of probability weighted outcomes. I think the potential for the common to get hosed by the preferreds here is high, so that’s why I can’t get bullish on the common.

  • Profile photo of Rob Powell Rob Powell says:

    Ronin, I believe that only some $276M of the preferreds must be refinanced in 2010. That said, I don’t think it will be painless either.

    • Homer says:

      The proxy filed last night has one very important piece of information: the “special committee” within the BOD has been dissolved. This was the committee made out of “outsiders” to negotiate with Mr. Icahn the terms of the first refinance. Such committee is absolutely required to deal with insider transactions. The conclusion is pretty straight foreward: the next refinance does not involve Mr. Icahn (and the possibility of the greatly feared second round of dillution has pretty much evaporated). A straight senior note with a third party is now almost certain.

  • dcgaffer says:

    Here is a simpler way:

    $125M EBITDA * 5.5x multiple = $660M

    Add $220M of Net Working Capital (NWC is Current Assets minus Current Liabs – Too often even supposed “experts” say “Working Capital” (which means CA) when they really mean NWC)

    660 + 220 equal “Total Enterprise Value” of $880.

    Subtract the $1.0B in preferred (since preferred is treated as debt and invariably has a liquidation preference) leaves negative common equity value of $120M.

    You came up with an optionality value of $110, but since the delta here is $140 (calculated negative equity of 120 plus current market value of equity of ~$30 million, tells me that the market says the volitality facor is a little higher than your admittedly low 10%.

    Now of course the optionality value should include the error estimates in the back end valuation of present value of the free cash flows (PV-FCF)….

    • Homer says:

      Thanks for your input. Just a couple of observations:

      First, you cannot use the 5.5 ratio unless adjust the comparables using NWC instead of cash. Sort of you cannot compare a kid’s height without shoes to a classroom average calculated with everyone’s height measured with shoes. And since you are adding current liabilities to numerator of the ratio and current assets tend to be significantly in cash, this multiple will inevitably go up. This means the equity is worth more than your calculation indicates (using your own methodology).

      Second, even if we were to assume that equity is out of the money by 150 million (i.e. less that 25% of TEV), a multi year call option on this business should be worth significantly more than 30 million (probably several multiples of that). Really no need to calculate anything to get a feel for that.

  • Homer says:

    Last year’s refi was significant in the sense the company was left with no senior claims and no cash interest payments. It is all accreeting preferreds. Clearly Icahn wants to leverage his own investment with someone else’s cheap money. This is a textbook case for a senior note (less than 2x EBITDA needed, no other senior claims, zero cash interest expenses). Bankers are probably tripping over themselves trying to get this deal… even in today’s credit environment.

    Once everything is done (i.e. including refinance #2) Icahn net cash outlay in refinancing XO will actually be quite small, if not negative. He put in 329 million new money last year but will get 215 back early next year or even sooner (via pref A redemption). His remaining 115 million will probably come from tax benefits.

    FWIW… the common already got hosed. It happened last year.

  • Ronin says:

    I stand corrected, thanks Rob. I re-read the 10K, and there is only $276 of the preferreds maturing, which to Homer’s point, is an easy refi via a senior secured deal (as lvlt showed possible recently)

    But with $800mm of preferreds still ahead of the common (which are accreting in value each quarter) and some cash burn, we’ll need to see a lot of ebitda growth before there is incremental value for the common (ie. the high end of Homer’s $135-160mm range) or higher value mulitlpes than 5.5x. A lot of ifs…

    • Homer says:

      Ronin… the new preferreds are perpetual -i.e. no maturity to worry about- and the accretion rate is a weigthed average of 7.72% per year… on the liquidity front… operating cash flow is actually positive (+71 million in 08)… zero cash interest expense… and capex is expected to tapper off significantly in 2010 (they have spend some 400 million upgrading their network over the last year or so)… they should be FCF positive in 2010… the obvious question is whether this investment will in fact produce the improvement in opperating margins they had projected (they did give investors a presentation with some ROI estimates for the capex binge last year… achieving anything remotely close to that would make most EBITDA estimates -mine included- look awfully inadequate).

      • DLM says:

        In my mind the big question regarding XO is why are the margins so low. Gross margins below 40% are unheard of in the telecom business. EBITDA margins below 10% are equally rare. If XO merely raised EBITDA margins to a very poor 15%, then EBITDA would be close to $220 million and the value of the shares even at 5x would be multiples of the current 19 cents. TWTC’s margins are greater than 30% and LVLT’s are in the mid 20’s so 15% should not be an impossible task for XO. Maybe they still have duplicate costs from the Allegiance deal but at some point you have to rationalize those.

Leave a Comment

You may Log In to post a comment, or fill in the form to post anonymously.


  • Ramblings’ Jobs

    Post a Job - Just $99/30days
  • Event Calendar