RTCrawford’s Weblog

My friend Stuart from the Yahoo message boards indicated recent consideration of Accuray, Inc. (NYSE: ARAY) as a stock investment. Stuart is a journalist by training, with a focus on healthcare, and his investing approach relies on in-depth research into the market, the economy, and the story behind those companies he is considering. My approach is that of a deep-discount value investor. Comparing the approaches struck me as an interesting exercise, which includes my quantitative/fundamental assessment of Accuray.

We should begin with Stuart’s initial posting.

Sorry if this post has nothing directly to do with AEO but I believe it is an instructive example of piecing together information to make an investment decision.

Accuray (ARAY) developed the Cyberknife, an advanced technology for treating solid tumors, primarily brain and lung tumors.

News of Sen. Kennedy’s diagnosis of a brain tumor caused ARAY to jump from about 9.15 late Tuesday afternoon to above 10 at yesterday’s high — about a 10 percent move in less than a day.

I looked at the ARAY message board shortly after the Kennedy news was announced Tuesday and saw that some posters were already buying the stock, putting two and two together.
In addition, the technical picture was positive as the stock had put in a low around 7 and was recently finding support around 9.

ARAY has a notoriously bad management but the technology is superb and more oncs are adopting it. My belief is that ARAY will eventually be bought out (possibly by GE which owns a million shares of ARAY). I could see a buyout price at a nice premium.

With minor edits, here is how I responded (minus the graphics) — I’ll end this with Stuart’s response:

ARAY is an interesting choice as a point of comparison with value investing. On the one hand, it has been producing SEC-posted financials for just two years, which makes using past data to identify cash flow trends impossible (you need at least five years if assuming a normal distribution, a la Pareto and Shewhart).

Because it is start-up (i.e., recent IPO), other valuation metrics are difficult. Shares outstanding jumped from 15 million to 30 million, year over year (cutting earnings per share in half if earnings are held steady), but a secondary IPO is not uncommon at this early stage in a company’s life cycle.

Similarly, the company’s overall cost structure is uncertain (the division of debt to equity for financing). Free cash flow coverage of current liabilities was 19.3% and 3.6% in 2006 and 2007, respectively. And cash return on invested capital dropped from 15.49% to 2.09%, year over year, as well (-7.4 percent in the latest quarter).

Both would be enough for me to move on to other opportunities, but there are a number of distinct positives about the stock and its fundamentals.

For example, its replication book value is $4.64/share versus it’s current price of $9.96 (for a market-to-adjusted-book-value of 2.14 — not great but not too bad, either). A value investor would look for something below 1.6 with a company whose prospects are uncertain and be willing to go as high as 2.5 if the company were proven and excellent.

As for franchise value, it is necessary to calculate the required return. Yahoo indicates the beta is 2.84.

This should be multiplied by the standard market risk premium of 8.6 percent and added to the risk-free-cost of capital (10-year bond yield) of 3.88 percent. This puts the required return at 28.30 percent.

If you take Buffett at his word and discount beta entirely, the required return drops to 12.48 percent (Ollie, you were right, by the way).

At 12.48 percent as the opportunity cost, the weighted average cost of capital comes in at 12.48 (which is just below the CROIC in 2006 and significantly above it 2007 and the most recent quarter) and renders a Earnings-Power Value (EPV) of $50.99. EPV is not the value of the stock but, instead, is a point of comparison with the replication value. So, the company has a significant franchise benefit, which, if managed properly and if possess adequate working capital, becomes a distinct positive.

In other words, it is a mixed bag. With negative operating income, it will need to turn profitable in the next two years given its burn rate, or, alternatively, take on debt, or issue more shares.

As for intrinsic value, if you take 75 percent of the weighted average cost of capital as your projected free cash flow growth rate for the next three years, …

reduce that by 10 percent for years 4 through 6, reduce it a further 10 percent for years 7 through 10, …

assume 5 percent for the second decade, …

use a discount rate of 15 percent, and use shareholder’s equity as the terminal value …

, you get $5.77.

If using Williams/Pabrai and Ponzio multipliers, you get intrinsic value of $2.01 and $6.01, respectively.

Because the break-up value of $4.18 is higher than Williams/Pabrai, we would use the range between break-up value and Ponzio (which is marginally higher than my DCF estimate of $5.77 to identify a range of intrinsic value of between $4.18 and $6.01. A value investor would require a further margin of safety due to the short length of time for which we have audited data. If the company were exceedingly strong and possessed proven market demand, we could justify a 25 percent margin (on top of the 15 percent discount rate). In this case, as a deep-discount value investor, I would use a 50 percent margin of safety, however. This urge considering purchase at stock price below $3.00 per share.

In other words, ARAY is, either, a trade on the Kennedy diagnosis or a wager that the company will grow cash flows at a compounded rate of somewhere between 26 percent and 29 percent for the next decade (in order for growth to serve as your margin of safety). If 20 percent is the maximum a company can comfortably grow with average to marginally above average management, you’ll need to possess greater faith in the acumen of management to make good on your wager if viewing ARAY as a longish-term investment. I gather, however, that you see it as more of a short-term trade (?).

Interestingly, much of the technology behind the product was created by my friend Julian Rosenman at UNC. Brilliant fellow with a devastating sense of humor.

As promised, here is Stuart’s response:

Wow. could our approaches be more different, Robert.
I talked to a couple of oncs before I dipped a foot in ARAY.
The technology is the best — no comparison with TOMO and VAR.
Kudos to your friend.
I really think ARAY gets taken over in the next two years — possibly near 20. (It could probably be bought now for 15-16).
There is almost a revolt going on now against ARAY management. There are profits being generated but management is eating them up with their stock options.
Eight cents last quarter became 1 cent after provision for the options.

So, which of us right? I suspect Stuart is, but, with a son ready to enter college and preparing to turn 50 in June, I am more risk adverse. In fact, the last time I took a significant risk was 21 years ago when my bride sealed the deal with a heart-warming “I do.” Perhaps I should try speculating more often.

Actually, the better method is to combine the two approaches (journalistic investigation with fundamental analysis), but, even then, I suspect Stuart and I would arrive at decidedly different conclusions on this and most other investments.

++++++++++++++++++++++++++++++++++++++++

In the message board discussions that followed, one poster expressed concern over the technological viability of such a new product.  I responded:

The concept and approach is valid and has been in use since the 90’s with 3D treatment planning (Rosenman’s advance) and conformal treatment of the patient — where no singularly lethal dose is delivered from a multitude of angles targeting the tumor and, collectively, deliver a lethal level of radiation to the tumor site. This allows the practitioner to avoid harm to healthy tissue and avoid areas (such as spine and heart) where even low levels of radiation should be avoided. Where the Accuray product is superior is in the smaller beams (less secondary bleed-over effect) and the use of robotics and computer systems. The robotics provide greater speed in delivery of treatment, while the computer system makes it less necessary to restrain the patient from moving during treatment.

The reason I say the general approach is proven is that patient receiving treatment with earlier approaches are surviving much longer (often with adjuvant therapies — post operative — and with combination chemo / rads regimens). For example, stage III non-small cell lung cancer patients treated with combination chemo / rads (using 3D conformal treatment) have seen life expectancy improve to levels previously only experienced by stage II patients. This was a follow-on discovery / advance by Rosenman and Mark Socinski at UNC. In fact, life expectancy is so improved from use of this radiation approach that many patients are dieing of secondary cancers attributable to their earlier treatments with radiation — with some arguing that many patients receiving radiation should also receive cranial radiation to preemptively address the potential for secondary brain cancers. This is why surgery has long been considered the only viable treatment for cure — but this is changing.

This means the product has the superiority of a new advance but it doesn’t represent as quantum leap an advance as the original approach. This makes it ripe for replacement by the next incremental improvement (i.e., a smaller barrier to entry). They haven’t, in other words, reinvented the wheel, but they have introduced steel-belted radials. The key to determining value is to research the patents for their impregnability and, more importantly, to determine the competition’s ability to supplant it with a moderate improvement. With healthcare, as well, it is necessary to consider the patent life remaining, given the long lead times before FDA approval — which reduces the period of exclusivity.

Stuart added:

Robert, you certainly understand the technology. You expressed it perfectly.
But consider:
– It is likely that improvements in the technology will come from ARAY itself rather than from competitors. In fact, we are seeing new treatment updates now from ARAY. (In ophthalmology, products like multifocal lenses and excimer lasers are usually improved by the original innovator).
– It is likely that more oncs will be trained on and become comfortable with the CyberKnife technology, which in terms of applications is probably in its infancy.
– the CK User’s Groups meetings are growing rapidly in terms of attendance and in the number of new research papers presented.
To me, these are all signs of a growing rather than a mature technology.
Penn is now building a proton-beam radiation center fore about $200 million. That will be one of a kind, I believe. The CK at $4 million is much more practical for more widespread adoption.
The exit scenario for current ARAY shareholders will probably be a buyout at a nice premium, I believe.

[Note: Much of the content follows from my work with Joe Ponzio, founder and part-owner of the Meridian Group in Chicago and the gentleman behind www.FWallStreet.com -- a website of incalculable intrinsic value for value investors.]