How does a seemingly unstoppable little software company with a 10 year track record of 40% operating margins, consistently high revenue growth, and mouthwatering free cash flow plummet from $30 a share to $8 on the heels of a possible merger with financial juggernaut Goldman Sachs Group (NYSE:GS)? This is precisely what occurred in the case of Ebix (NASDAQ:EBIX).
What started as meticulous fact checking by a short-seller under the pseudonym Copperfield Research has snowballed into a series of lawsuits and inquiries by both the IRS and SEC, the latter of which was cited by Goldman Sachs Group as the cause for terminating their $20 per share buyout offer.
On the surface, the future seems bleak for Ebix and its CEO Robin Raina, who owns approximately 12% of outstanding shares. Amidst the uncertainty and merger termination, shares nosedived to nearly $8 a share. It seems investors have already assumed the worst, and have priced in the presumed damage to the company's balance sheet and reputation.
Ebix has existed for decades as a leading developer and vendor of insurance industry software. The company has the competitive advantage of operating in 4 distinct areas of insurance software, thereby offering clients a complete and horizontal line of products, while competitors remain anchored to single areas of insurance software. The past decade has seen the company grow revenue from 14 million to 208 million a year, while maintaining an operating margin of nearly 40%, and free cash flow nearly the same.
The Drama Unfolds
On March 24th, 2011, a Seeking Alpha contributor operating under the anonymous pseudonym Copperfield Research published a highly detailed and compelling article which questioned, among other things, the financial reporting, corporate governance, revenue recognition, tax exemption, and asset balances found in the company's required reports with the SEC and international securities regulators (Ebix, Not a Chinese Fraud, But a House of Cards Nonetheless, Part I, Part II, Part III). The author's piece created a very persuasive case for investors to abandon all confidence in Ebix, which is exactly what they did. Another article by a short-seller operating under the pseudonym Gotham Research can be found here.
Through 2011 and 2012, a series of lawsuits were filed. The most influential was a class action lawsuit in the State of Georgia which alleges both fraud and Exchange Act Section 20(a) violation on the part of both Ebix's CEO and CFO on behalf of shareholders of record between 05/06/09 and 06/30/2011. Essentially, shareholders claim that company management intentionally misrepresented financial statements to artificially elevate stock prices. The lawsuit draws most of its accusations from the information disclosed in Copperfield Research's article.
In the midst of the allegations by shareholders, the IRS quietly began to make inquiries into the company's tax history. Early in 2012, the SEC also began its own investigation, but has also been very quiet about any findings. Early in 2013, Goldman Sachs Group and Ebix indicated they were in the latter stages of merger talks, only to have them abruptly terminated on June 19th, citing the SEC probe.
The above timeline has occupied the complete attention of shareholders lately, due to the scale and dramatic nature of its events. P/E valuation has fallen from 15 to just over 5. However, there are several points which seem to have escaped the attention of investors. They present major opportunities for future alpha.
A. Despite the merger termination, Goldman-Sachs and investors were fully aware of all of the above information beforehand. Ebix states in its 2012 10-K report, "The Company is cooperating fully with the SEC to provide the requested documents and otherwise assist in the SEC's review, and advised the Acquiring Parties of the existence of this matter as well as the open tax audit discussed in Note 6, prior to the execution of the Merger Agreement."
No new findings have come to light, nor has revenue or EPS performance in the first quarter given any reason to terminate the merger. Despite this, the day following the merger termination saw share prices drop from almost $20 to around $8. There is no reason that present information should warrant the current trading price of $11 when not too long ago, investors valued it around $20 with identical information. Reaction to failed merger-arbitrage positions has led to overselling Ebix.
B. Investors are valuing Ebix at below its worst-case-scenario fundamentals. If we assume that Ebix is found liable for dodged taxes, falsified financial statements, and forced to pay injunctive relief to class action plaintiffs, the company's fair value still exceeds its current value. See the worst-case valuation later in the article.
C. Earlier this year, 3 other lawsuits alleged breach of fiduciary duty on the part of the Board of Directors, claiming that Goldman's offer of $20 per share is actually TOO LOW to be in the best interest of shareholders. To put this in perspective, the 3 parties (2 individual shareholders and one investment firm) believe they have compelling evidence that the company is valued significantly higher than $20 per share. The pending cases (Shah, Llinares, and TBD Investments v. Ebix, Inc) are detailed in the company's 2012 10-K report.
D. For the Georgia class-action lawsuit to be successful, plaintiffs must prove scienter (malice). To successfully prove fraud and 20(a) Exchange Act violation, it must be proven that the CEO and/or CFO had the absolute intent of harming shareholders. Thus, even if it can be proven that the company misreported earnings and did so intentionally, the case will be utterly unsuccessful if it cannot be proven that they did so with the intent to harm shareholders. This information is difficult, if not impossible to acquire. The unique tax structure that Ebix enjoys is unique and may not even be understood by the company, which would imply there is no intent to harm shareholders.
It should be noted, however, that the SEC does NOT need to prove scienter (malice) to levy its judgment and punishment independent of civil court (the lawsuit). Proof of material misrepresentation of financial statements will likely bring about consequences from the SEC.
E. After 2 years, all civil suits and government investigations have yet to find evidence of wrongdoing by Ebix. While the aforementioned lawsuits and agency investigations are presumably still ongoing, there have been no conclusions, nor statements, concerning any liability on the part of Ebix. Even specific claims to material misrepresentation mentioned in Copperfield Research's articles have not been legitimized by any findings by the SEC or IRS.
F. Ebix has a recent track record of distributing cash, and plans to continue in the near future. Current valuation of Ebix shares indicates investors believe that due diligence will soon reveal phantom earnings and balance sheet cash that does not exist. Yet, it is difficult to purchase shares with 100 million dollars that you do not have (this is exactly what the company announced it will be doing over the next 12 months). At current prices, 100 million dollars in buybacks would reduce outstanding shares by almost 1/3. In the face of stagnant lawsuits and investigations, this seems to be the only way the company can prove to investors that its earnings are real. This is a major thorn in the side of any short-seller case.
A look at P/E history reveals that Ebix held a valuation closely in line with the market in 2010. In 2011, with the onset of allegations and investigations, P/E began a nosedive that has continued in 2012 and 2013.
(click to enlarge)
Before before 2011, Ebix's price appreciated at a staggering rate, tripling over the prior two years. After allegations surfaced, price fell from a high of nearly $30 to around $8 a month ago.
Over the same period, revenue and earnings grew convincingly and without any hiccups (continuing into Q1 2013 not pictured)...
The above graph can only be described as a Picasso to growth investors. Such a high level of growth and consistency would seem to warrant a P/E far above average, yet the company is sitting at an alarmingly low P/E of 6.13. From this vantage point, the current price is an enticing point of entry. However, the risks provided by the battles in which Ebix is currently embroiled require a more fundamental analysis.
Looking forward, even in the worst-case scenario, Ebix is far from worthless. Since allegations by short-sellers and plaintiffs have been extremely specific, their implications can actually be estimated quite easily.
Let's pretend all of investors' greatest Ebix nightmares are realized; the class-action suit fails and Ebix must pay 100 million in injunctive relief, the SEC/IRS investigations find conclusive evidence and levy 100 million in punitive measures and 50 million in back-taxes respectively. Let's also pretend that Ebix henceforth pays the highest possible taxes on future earnings (35%) and its current CAGR (compound annual growth rate) of revenue and earnings, which is around 40%, is cut in half to 20% for the next 5 years due to damaged reputation and indicted executives. Since over 85% of Ebix's revenue is recurring, its growth is nicely cushioned against this situation, but we will humor short-sellers and assume it anyways. Here is a breakdown of the assumed worst-case penalties...
Using a Two-Stage DCF Total Enterprise Value Model, Ebix's fair value can be calculated based on historical CAGRs and the penalties assumed above. The following is derived from Ebix's debt structure and appropriate betas. Cost of equity is calculated using the CAPM model and unlevered beta (derived from the levered beta of 1.35, as indicated by Google Finance). Cost of debt is taken from the rate at which Ebix currently pays to borrow (as indicated in its 2012 10-K report). Risk-free rate is the current yield of a US 10-year Treasury and market return is estimated based on the S&P 500.
To humor the short-sellers, I will again skew my assumptions by nearly doubling the required rate of return (OTC:WACC) to 10% from the calculated value of 5.78%, since this is the worst cast scenario.
Here are the relevant financials for the past 6 years and corresponding CAGRs implied (taken from the company's 2012 10-K report and Morningstar). Note that figures without a CAGR are derivative of another figure. Tax expenses have been restated to the maximum rate of 35% to reflect the assumption that prior tax reporting was inaccurate.
Past 5 Years (in thousands of dollars):
I have reduced revenue, EBIT, and operating cash flow each by 20% on an absolute basis, meaning that they will grow at less than half the rate they did the past 5 years.
Given these modifications, here are the projected cash flows for the next 5 years.
Next 5 Years, Worst-Case (in thousands of dollars):
(in thousands of dollars)
Next, we must calculate a terminal value for Ebix using a perpetual growth rate that it is assumed can be attained after the year 2017. To be consistent in assuming the worst case scenario, we will assume a conservative perpetual growth rate of 3%. Using this rate and the Gordon Growth Model for terminal value, we arrive at a present enterprise value of 520 million (after subtracting 148 million in current liabilities, given that we do not know how many assets may be restated, and subtracting 250 million in penalties).
Worst-Case Share Value (in thousands of dollars, except per share):
520 million dollars in present enterprise value net of liabilities and penalties gives us approximately $13.99 per share.
The fair valuation of $13.99 per share reflects approximately a 25% premium to the current market price of 11.02. Furthermore, this fair value takes into account every possible factor that could work against Ebix, including a WACC that over 4% higher than what can be fairly calculated, growth rates for the next 5 years at less than half the rates we have observed in the past 5 years, a very conservative perpetual growth rate for beyond 5 years, penalties amounting to 250 million, and the assumption that current assets have a value of zero.
To put the worst-case in perspective, I will repeat the above process, removing all impairing assumptions that were made. I will use the actual 5-year CAGRs (which are actually lower than 10-year CAGRs), the company's current effective tax rate (9.5%), the unmodified WACC (5.78%), the effects of the current share buyback program (100 million over 12 months), and the assumption that balance sheet assets are accurately stated.
Recalculated to reflect the unmodified 5-year CAGRs, here are the next 5-years' cash flows.
Next 5 Years, If Not-Liable:
In this scenario, prior data suggests that Ebix could attain a cash flow of 312 million by the end of 2017. Calculating terminal value in the same manner as in the worst-case scenario (3% perpetual rate of return after 2017), Ebix would achieve a future total enterprise value of 12,464 million (almost 12.5 billion). Discounting each cash flow by the required rate of return (4.78%) gives a present total enterprise value of 9,468 million (almost 9.5 billion). In addition, the use of 100 million in share buyback funds currently has the power to buy back 8.95 million of 37.17 million shares for a net balance of 28.22 million shares.
Fair Value, If Not Liable:
A valuation of $335.56 per share is what is produced by the projections. While it would be a rebound of historic proportions if Ebix rose to this value, the valuation serves to illustrate just how much upside exists if Ebix continues on its current trajectory for the next few years.
Strength of Assumptions
Certainly, both the worst-case and not-liable valuations make a great deal of assumptions. I believe it is important to support the assumptions made in both cases, since fair value is highly sensitive to several of these assumptions.
First, the assumption of growth is based on CAGR. Very frequently, future growth does not match past CAGR, especially when the growth rate is approaching 40%. However, a look at the past 10 years shows that this trend has occurred since the year 2003. In addition, growth is approximately half organic and half due to the acquisition of other businesses. Since Ebix remains a relatively small company, it seems quite unlikely that it will run out of strategic acquisitions to make. Furthermore, given that 78% of revenue is recurring (subscriptions, yearly service, etc.) and is derived from Ebix software, it seems rather unlikely that any customers will deem it prudent to scrap a partnership with Ebix and purchase new software.
Another assumption that sensitizes the price is share buybacks. Ebix's track record of share buybacks supports the notion that the company will honor their 100 million dollar buyback plan, especially with current prices at low levels. The following chart illustrates Ebix's buyback track record with the 2 programs it announced in 2011 (data taken from reported share purchases in Ebix's 2012 10-K report):
Ebix's announced 30 million buy-back program in early 2011 was fully honored within 183 days and the company's second buyback program of 50 million announced later in the year was 95% honored within 488 days. Note that the excess buybacks that occurred during this time were fulfilling earlier, much smaller, buyback programs. From this track record, Ebix spends each dollar of announced buybacks, on average, 169.51 days after program announcement. The company has explicitly stated that it plans to complete its current buyback program within 12 months, but this analysis reveals it may be completed in as little as half a year.
Other major assumptions include cost of equity and WACC by extension, long term growth rate, possible penalties from lawsuits and investigations. While I have attempted to under or overstate these correspondingly in the worst-case scenario, they have no factual or historical support and I draw their values from what I believe most investors would likely perceive.
Ebix is undervalued, no matter which model you choose to apply. This undervaluation is due entirely to the looming threat of penalties and the effects they may have on the company. However, as I have shown, current share prices are below what should be expected, even if these looming penalties were certain, which they are not. Conversely, if the company is vindicated, its shares stand to be worth many times the price at which they currently trade.