Ultimate Software and the Art of Non-GAAP Earnings

| About: Ultimate Software (ULTI)

Ultimate Software (NASDAQ:ULTI) prides itself on being wonderful to its employees, and it consistently places highly in “best places to work” surveys. I have no doubt this is true, as a careful reading of the 10-K and some basic analysis reveals that all of the economic profits from ULTI's business accure to its employees, leaving nothing left for the actual shareholder owners of the company. In my opinion ULTI's valuation and share price performance is a testament to just how speculative and disconnected from economic reality today's tech stock market is.

Whether or not stock-based compensation should be counted as an expense has long been a source of controversy among Wall St. observers. While any thinking business owner recognizes that employee compensation is a real cost no matter what form it takes, tech companies in particular enjoyed being able to richly pay their employees without recognizing any expense.

Wall St. and corporate America fought the FASB’s attempts to require stock-based compensation to be recognized as an expense tooth and nail. As Warren Buffet observed in 1998:

A distressing number of both CEOs and auditors have in recent years bitterly fought FASB’s attempts to replace fiction with truth, and virtually none of have spoken out in support of FASB.

FAS 123 was finally adopted at the end of 2004 against howls of protest. To blunt its effects, many companies began issuing “pro forma” results which simply ignored stock-based compensation, and Wall St. analysts dutifully followed suit. At first this made some sense, as prior-year comparisons would have been muddled by the change in accounting standards. I would also point out that the real value of stock options is difficult to estimate, since their actual cost can either end up being nil or enormous depending on the subsequent performance of the stock. Thus they are meant to be a bonus contingent on value creation.

However, as the years march on many tech companies continue to train their Wall St. analysts to look past all stock compensation expense. ULTI represents an extreme of this practice.

In 2010 ULTI reported $0.11 in diluted GAAP EPS from continuing operations on a meager 4% operating margin. That's not a lot of earnings for a $60 stock. In order to make the stock seem more attractive to investors, management has convinced Wall St. analysts to focus on its non-GAAP (i.e. excluding stock compensation expense) 2010 EPS of $0.48 per share, giving ULTI a multiple of “only” 125x trailing and 83x 2011 estimated non-GAAP EPS.

However, ULTI’s non-GAAP earnings ignore significant, real expenses borne by its shareholders. ULTI no longer gives out employee stock options, rather grants restricted stock and restricted stock units. Importantly, employees receive their restricted shares upon vesting and can sell them for cash no matter whether the stock goes up or down. This means that shareholder dilution is inevitable.

In 2010, ULTI granted 166,000 restricted shares at a fair value of $40.51 and 200,000 restricted stock units, presumably around the same price. This is nearly $15M of stock given out to employees (more at today’s prices) and more than 100% of 2010's $12.8M of non-GAAP net income according to the company's press release. In other words, ULTI essentially earns no economic profits for shareholders. These shares have real cash value. But somehow, as long as ULTI pays employees in stock grants instead of cash, we are supposed to believe it isn’t a real expense. The 10-K description of the restricted stock units even indicates that the award can be paid by the company in cash instead of stock on the vesting date!

There is another way of illustrating the same point. In 2010, ULTI reported net cash provided by operating activities of $25.4 million and capital expenditures of $5M for free cash flow of $20.4M. At today’s enterprise value, the stock trades for almost 70x trailing free cash flow, which is quite expensive by just about any standard. However, the company spent nearly all of this cash ($19.8M to be exact) to buy back 610,000 shares at an average of little more than half of the current price. At the same time, basic shares outstanding went from 24.5M to 25M in 2010. In other words, despite plowing 100% of the cash it generated back into buying shares at much lower prices than today, ULTI’s shareholders still experienced dilution. As of the March filing of its 2010 10-K ULTI’s basic shares outstanding were up even further to 25.6M.

ULTI treats its employees very well and their software appears to provide real value for at least some modest sub-set of American businesses. However, I think that investors and analysts are foolish to take management's suggestion of focusing on non-GAAP earnings that ignore substantial compensation paid to ULTI employees simply because it is in shares instead of cash. If the company decided to pay all of its expenses in stock, would analysts dutifully model 100% operating margins?

I also find the great share price performance amazing given the fact that even the analysts’ non-GAAP earnings estimates for 2011 have been reduced over the last year. The Wall St. analysts whose reports I see justify their bullish price targets by comparing ULTI's revenue multiple to SAAS market darlings like Salesforce.com (NYSE:CRM) and SuccessFactors (NYSE:SFSF), despite their drastically lower deferred revenue growth, 14% in 2010 vs. 33% and 29%, respectively.

But there are indications that ULTI management may also be surprised at the valuation its business has achieved. At conferences, they have announced they are going to “build cash” instead of continuing to plow what meager cash flow they earn into stock buybacks. According to January rumors ULTI attempted to sell itself in the early part of the year. But apparently either the rumors were false or no one would pay a sufficient premium to its stock price (about $50 at the time).

Senior executives have been consistent sellers of the stock they grant themselves, while telling Wall St. to ignore the stock expense. The significant non-10b5-1 sales in the February open window indicate the company doesn’t expect a buyout to happen any time soon. Of course, in the current QE2-cloud-computing-speculative melt-up, it's anyone's guess where the share price will go.

Disclosure: I am short ULTI.

Additional disclosure: No position in CRM or SFSF