Poor Ebix…the Company and its flamboyant, turnaround expert CEO have had a challenging ten months. First, on September 28, 2012, an Atlanta U.S. district court ruled that an investor lawsuit alleging false statements by the Company in its financial reports could proceed (see 2012 10-K, page 15). Next, came the revelation in November 2012 that the U.S. Securities and Exchange Commission (SEC) was investigating the Company for its accounting practices: revenue recognition and financial reporting internal controls. This was followed by news on June 14, 2013 that the U.S. Attorney in Atlanta is investigating the Company for intentional misconduct. And this last bit of news not only scuttled an offer by Goldman Sachs Group, Inc. to buy the Company for $780 million, but also wiped out almost $300 million in market capitalization. At least for the time being, Robin Raina’s dream of owning 29 percent in the “new” company, and silencing the short-sellers who have been hounding the Company for the past two years, is dead. But is bad accounting really to blame for Ebix’s recent misfortunes? Or could it be something else?
What you ask? Well, a recent column by Jean Eaglesham titled “Accounting Fraud Targeted” might just offer a clue. This article mentions the SEC’s use of new software to analyze the 10-K’s management discussion and analysis (MD&A) section looking for signs of possible earnings manipulation and other financial reporting fraud behaviors. So, could “bad writing” have tripped up the Company? After all, on the surface, the Company’s numbers seem just fine: increasing revenue, operating income, earnings per share, and assets. As you might expect, this grumpy old accountant just had to conduct his own “textual analysis.”
Without any fancy fraud detection software at my disposal, I decided to arm myself with the latest MBA jargon as an “enabler,” so that I might “drill down” into Ebix’s MD&A. I was not disappointed at all. On page two of the Company’s 2012 10-K alone, I was rewarded with such confusing and incomprehensible phrases as “powerhouse of backend insurance transactions; complimentary accretive acquisitions; carrying data from one end to another seamlessly; best of breed functionality; integrates seamlessly; best of breed solution; resources and infrastructure are leveraged; and acquisitive growth vs. organic revenue growth becomes rather obscure.”
And page three continued with “revenue based contingent purchase consideration; contingent reward; and consistent end-to-end vision.” By the time I got to page 5, I was running out of “bandwidth,” but still found references to “unique position; cutting edge solutions; and end-to-end solution.” I asked myself why a software company felt it necessary to obfuscate (that’s a grumpy old word) its strategy and business model. Is this the kind of writing that regulators are targeting with their software to ferret out “word shell games” by public companies? This writing style sure made my antennae go up!
While bad writing or suspicious language is not illegal per se, it does reduce reporting transparency, and actually made this grumpy old accountant want to take a closer look at the numbers. Yes, Ebix probably would probably accuse me of “boiling the ocean.”
Recently, the popular press has highlighted a number of concerns at Ebix that potentially affect the Company’s financial reporting: revenue recognition, accounting for income taxes, and auditor changes. It is not surprising that the SEC is interested in the Company’s revenue recognition practices given all the problems discovered with accelerated income recording by companies in the software and service industries. After all, many of these companies (including Ebix, see 2012 10-K, page 54) avail themselves of the subjective multiple deliverables revenue recognition rules which give management significant discretion over when sales revenues are reported. See “H-P Throws Its Accountants Under the Bus! But Why?” and “The Beauty of Internet Company Accounting” for more grumpiness on multiple deliverables accounting. So, it’s not really surprising that the regulators want to “peel the onion” on Ebix’s revenue recognition practices.
But is there more to the SEC’s revenue concerns than simply the proverb “tell me who you go with, and I’ll tell you who you are?” Well, there are a few quantitative hints that might be signaling revenue overstatements. First, the quality of revenues ratio (cash collected from customers divided by revenue) is less than one for all fiscal years from 2008 through 2012. Also, the quality of earnings ratio (operating cash flow (OCF) divided by net income) is consistently below one between fiscal years 2008 and 2011 before making a slight comeback in 2012 to 1.024. And finally, the excess cash margin metric (which compares OCF and operating earnings) not only declined precipitously from 2011 to 2012, but actually became negative, suggesting that operating earnings are growing more quickly or declining slower than OCF. So, a closer look at revenue seems a reasonable “actionable” item.
And then there are Ebix’s extraordinarily low effective tax rates, which some suggest may reflect an artificially lowered income tax expense, which inflated profits over the past several years. I used the conservatism ratio (reported income before taxes divided by taxable income) to examine the aggressiveness of the Company’s tax strategy. A higher ratio generally indicates that management is more aggressive with its tax policies. Well, Ebix’s conservatism ratio has steadily increased from 1.246 in 2010 to 1.452 in 2012, clearly confirming aggressiveness in tax reporting. The Company does try to explain this as follows (2012 10-K, page 72):
"The Company's consolidated worldwide effective tax rate is relatively low because of the effect of conducting operating activities in certain foreign jurisdiction with low tax rates and where a significant portions of its taxable income resides."
But one doesn’t fully appreciate the disparity between where income was earned and where it was taxed until one explicitly compares the following tables. Page 76 reports the sources of pre-tax income by country/region as follows:
Clearly, the bulk of 2012 revenues originated in the United States and Australia. Yet, according to page 73, pre-tax income by country/region was as follows:
Ebix’s tax planning strategy somehow transferred taxable revenues to the lowest tax paying jurisdictions, namely Singapore, India, and Sweden. But does the 10-K reader have any idea as to how this was done? No! Such techniques typically involve some type of transfer pricing technique where a subsidiary in a low tax jurisdiction, provides a service to the revenue generating subsidiary in a high tax country. The “low-tax” sub bills the “high-tax” sub at market rates for the services, thus lowering taxable income in the “high tax” jurisdiction and raising it in the “low tax” jurisdiction. But given the Company’s penchant for MBA-speak, why hasn’t the strategy been discussed in the tax disclosures, particularly given its dramatic impact on the bottom line and effective tax rate.
And while I’m on the effective tax rate (ETR), how about the unusual ETR reconciliation (presented below) which breaks out ETR from ongoing operations from ETR after discrete items? If the Company is willing to go to this extent to disaggregate ETR, why not describe the tax planning strategy and transfer pricing mechanics? Clearly this cannot be considered a “best practice.”
And then there is the issue of multiple auditors. According to the Audit Analytics data service, the Company has changed auditors a number of times moving from larger to smaller firms: KPMG from 2000 through 2003, BDO from 2004 through 2006, Habit Arogeti & Wynne from 2007 through 2008, and Cherry Bekaert & Holland from 2009 to the present. Now to be fair, one change did relate to a firm merger and another involved a Public Company Accounting Oversight Board (PCAOB) registration issue. Yet, it is quite obvious that the big firms were just not “drinking kool-aid.” And are we naïve enough to believe that the Cherry Bekaert auditors are really capable of auditing an international firm? And don’t you wonder just what audit procedures they applied to Ebix’s tax transfer pricing scheme to ascertain the propriety of income shifting from “high tax” to “low tax” jurisdictions.
But this grumpy old accountant has some of his own accounting issues with Ebix, namely accounts receivable and intangible asset valuation. The Company appears to be grossing up its balance sheet by including deferred revenues in accounts receivable (2012 10-K, page 54)…what’s with that? In the old days, you recorded deferred revenues when you received cash before income was earned. More troubling is the unexplained decline in bad debt expense and the allowance for doubtful accounts, as revenues and receivables increase (2012 10-K, pages 55 and 106).
Then there is the intangible asset issue. The vast amount of assets acquired in 2011 and 2012 were goodwill and intangibles, 83.1 percent and 91.8 percent, respectively. So significant are these “invisible” assets that they account for 79.35 percent of total assets at the end of 2012 (i.e., goodwill, intangibles, indefinite-live intangibles). This would definitely be a critical audit area for this grumpy old accountant. And it’s not just the magnitude of the intangibles that bothers me. OCF as a percentage of total assets has dropped significantly during 2012 from 17.18 percent in 2011 to 13.99 percent in 2012. Any guesses as to which assets are not generating cash? And for the past three fiscal years, OCF to goodwill has steadily declined from 28.67 percent in 2010 to 27.25 percent in 2011, to 22.13 percent in 2012. Sure sounds like an impairment issue to me. And don’t forget, OCF would even be lower were it not for the Company’s aggressive tax planning (i.e., income shifting) strategy. Which intangible is the problem?
Well, goodwill is the likely candidate given its impact on the balance sheet. But the other intangibles could be a problem as well. For example, take a look at contractual customer relationships included in indefinite lived intangibles. In this case, Ebix values these assets using discounted cash flow models over periods ranging from 15 to 20 years (2012 10-K, page 56). It would really be interesting to see the audit test work on this…heads up PCAOB! Oh, then there are the estimated lives for purchased intangibles which also appear excessive (2012 10-K, pages 56 and 62). Eleven year lives on customer relationships….please!
There also are a few hopefully minor points that I wish I could “dialogue” with the Company about. Why are there no acquisition expenses, integration costs, or restructuring charges given the Company’s rapid “inorganic” growth? Why the previous netting of excess tax benefits in the statement of cash flows and lack of discussion of impact on reported OCF (2012 10-K, page 50)? How can the Company justify one reporting segment when it has global operations? Why does such a successful company self-insure for health insurance? These are the types of questions that promote grumpiness…
And then there are the market pressures that may be providing incentives to management to engage in less than transparent reporting. For example, the following create performance pressures for managers that might promote aggressive reporting: the need to pay and increase dividends (2012 10-K, page 16); the history of rapid growth and profitability (2012 10-K, page 19); contractual incentives related to debt covenants which must be maintained (2012 10-K, page 31); and pressures to obtain capital to stay competitive (e.g. the Goldman deal). And the Company also triggers a fairly significant management and control red flag. Not only is the CEO the founder/rescuer of the current company, but he also has significant wealth tied up in it with his 9.5 percent ownership stake (2012 10-K, page 97). Finally, there are the issues of extreme competition and pressure to innovate that characterize the Company’s software industry.
So, is the Company “excessively” concerned with share price and earnings per share? Well, recent share repurchase activity and the authorization to proceed to the $100 million level may offer a clue (10-K, page 17). But the Beneish earnings manipulation model only provides a hint of earnings gamesmanship with a probability of manipulation of slightly more than one percent for fiscal years 2010 through 2012.
Well, there you have it…that’s my “elevator story” on Ebix. I really tried to avoid “scope creep” in my “proactive” efforts to “gain traction” in making my case for writing quality and financial reporting transparency at Ebix. Would a jargon-free and transparent approach to financial disclosure really have helped the Company’s cause? My “rough order of magnitude” suggests no…but you make the call.
This essay reflects the opinion of the author and not necessarily that of The American College, or Villanova University.