athenahealth's (NASDAQ:ATHN) CEO Jonathan Bush usually describes his company in buzzwords, acronyms and phrases: "the health internet", "the national health information backbone", "SES appeal" and so on. SES by the way stands for software-enabled services, which Bush uses to liken athenahealth to Amazon (NASDAQ:AMZN) and to distinguish it from SaaS (software as a service) providers. David Einhorn of Greenlight Capital, a hedge fund manager, had a lot of fun with all this in his speech at this year's Sohn Investment conference.
Some of Einhorn's criticism was a bit unfair. Jonathan Bush's use of the word "cloud" is certainly nebulous, but he (like many other health sector activists) does appear to have a sincere vision for how health information should follow patients seamlessly and securely as they move between payers and providers throughout the world.
The trouble is not with the words surrounding athenahealth, but the numbers. It is insanely overvalued by the stock market using virtually any valuation metric you can think of - whether intrinsic valuation, price relative to earnings or price relative to sales. It's a very graphic example of just how desperate investors are for any kind of "growth" stock in these somewhat economically morbid times.
In July 2014, athenahealth reported second-quarter earnings. This was widely reported as a great (even "knock-out") quarter, with media and analysts commenting on how the company had outperformed on both top-line revenue and bottom-line earnings estimates. Some even wondered why the stock had not responded more forcefully to such good news (more recently, the stock has indeed climbed sharply from about $120 post-July earnings to about $144 at the end of the summer trading season).
But if we analyze this earnings report a bit further and set it alongside athenahealth's historical financial performance, an altogether different story emerges. Athenahealth is a company with slowing sales growth, contracting margins and rising capital expenditures, with a hugely inflated stock price that looks increasingly set for implosion.
Start with sales growth. After six successive years of sales growth in excess of 30%, both consensus estimates and the company's fiscal guidance indicate that sales growth in 2014 will slow down to around 25%. For the first two quarters of 2014, Athena reported an increase in sales of 28% over the same period in 2013, but the average for 2014 as a whole will be lower as collections slow down over the holiday period. Beyond 2014, although there is no reason to believe Athena won't continue to expand its core practice management and revenue collection business amongst independent and smaller-scale practices (its products are well-liked and its customer retention rate is high), it faces considerably greater challenges in winning business from hospitals and larger practices. As for electronic health records (its second material revenue earner), the company has benefitted nicely from the HITECH Act 2009 incentive payments to physicians to make "meaningful use" of electronic health records. But even here, there is competition from providers willing to provide such systems for free, and the incentive payments in any case are due to stop altogether from 2015.
Move on next to margins. Gross, EBITDA and net margins all rose smartly from its IPO in 2007 to a peak in 2011, when they stood at 62.1%, 15.3% and 5.9% respectively. Since then, Athena's margins have begun to erode. Gross margins have come down to just under 60% in 2013, while EBITDA and net margins have collapsed to 8.3% and 0.4% respectively. The picture has actually worsened in the first two quarters of 2014: the second quarter statement for instance shows that gross margins contracted further to under 58%, while EBITDA margins collapsed to just 4.8% and the company made a net loss.
Management blamed this worsening performance on two key factors. First, stock-based compensation has pushed up expenses due to a rapidly rising share price, and so is depressing operating margins. And second, GAAP accounting adjustments for acquisition investments (such as its takeover of the medical applications developer Epocrates) have increased amortization charges, hitting net income. The company published non-GAAP measures of earnings and margins that excluded these factors, which converted falling gross and EBITDA margins into a stable and rising margin respectively, and a net loss of about $10m for the period into a net profit of $16.5m.
I have no sympathy with this approach. Prof. Damodaran has recently pointed out that excluding stock-based compensation from the calculation of margins and earnings is misleading and wrong. They are a part of an employee's compensation package, and should be treated exactly as if the company had decided to pay the same value to the employee in cash, issuing new shares to finance the payment, because the economic effect on existing shareholders is identical. And management can hardly claim that stock-based compensation represent a one-off, extraordinary factor that distorts operating results; athenahealth has made SBC payments every year since it went public, and these payments have been rising, not falling. They are very much a part of Athena's operating expenses. Its GAAP results therefore paint the more accurate picture: the business is suffering from sharply contracting operating margins.
Similarly, GAAP accounting requires companies to amortize acquisition costs because it is recognised and understood that (a bit like capitalised R&D), these costs will only yield results over time, and so expensing them entirely at the time they are incurred could unfairly distort operating results. But the quid pro quo is that they must be properly expensed over time! Otherwise, management is not held to account if it paid too much for the acquisition, or integration turned out to be too costly relative to any contribution in profits from the acquired business. In the case of the Epocrates acquisition for instance, amortizing over time the acquisition premium that Athena paid and the integration costs it incurred will give investors a fair view of whether or not the acquisition was worth it in the end. If it was, Athena's GAAP net income should (all else equal) rise over time. If it wasn't, it won't. Excluding these costs obscures this important judgement.
Let's reflect finally on capital expenditure relative to sales, or what I call the capital intensity of the business. Athena's capital intensity has averaged about 9% over the past six years, although it declined to a low of 6.8% in 2009 and since then has risen to about 11.3% in 2013. The first two quarters of 2014 show a sharp acceleration in capital intensity to 15.8% of sales. Reflect on what this means for shareholders. The higher the capital intensity of the business in steady state, the lower the free cash flow available to shareholders from a given level of sales and profits.
Some analysts shrug off rising capital intensity by invoking the "Athena is investing in the future" argument. Einhorn pointed out the trouble with this argument - Athena is not a "build it and they will come" business with network effects and operating leverage. Most of its capital expenditure is on what I would describe as "semi-fixed" costs, viz. purchasing space and servers for data centres and offices and equipment for staff, both of which tend to grow as its business grows. Moreover, software and hardware have to be replaced every few years because they depreciate so quickly, so a large element of capital expenditure tends to recur even if sales are flat.
What does all this mean for athenahealth's valuation? Even if we were to assume that sales will continue to grow at more than 20% for the next ten years, and even if we assumed that its EBITDA margins will recover to their historical peak of 15.2%, and even if we assume that its capital intensity will fall back to its historical average of 9%, the free cash flow that emerges from these assumptions would (discounted even at a generously low WACC of 9% and a generously high terminal value growth rate of 5%) produce a share price of only $88 per share.
The bear case - where current trends continue - barely bears thinking about. For instance, if EBITDA margins fail to recover and stabilise at the 2013 level of 8.3% and if capital intensity remains at the 2013 level of 11.3%, then even with sales growth in excess of 20% per annum, athenahealth will likely continue to generate net losses into the future and its shares would be worth nothing.
Disclosure: The author is short ATHN.
The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.