Description

I am recommending the purchase of StarTek, Inc. (SRT) common stock based on what I believe is an overreaction of the market to the Company’s admittedly abrupt (and potentially troubling) announcement last week that its CEO had resigned effectively immediately. The CFO, who has himself been on board only since the first of the year, has assumed the interim CEO role. Thus, no one who has attested to the Company’s most recent financial statements remains with SRT, and I think there is a risk of the proverbial other shoe dropping sometime soon with some sort of restatement or accounting scandal, which likely is why the market reacted so violently to the news, dropping the shares from nearly $26 to its current level of $18.45. Such an outcome also would not be inconsistent with the “material weaknesses” in financial processes identified by SRT’s auditors as disclosed in the Q3 10-Q. Having said all that, and cognizant of the fact that the shares could get yet cheaper, I believe today’s price represents a good entry point at approximately 4.5x TEV/LTM EBITDA and a dividend yield of 9% (the dividend, at approx $24mm annually, also represents the preponderance of free cash flow). Normally, I would not go near a situation if I thought the integrity of the numbers was in question, however, I take comfort from the fact that SRT has paid $23 million in dividends in the past year while its net cash/debt position is essentially unchanged (after accounting for a roughly $4mm overpayment on taxes, which is being refunded this quarter); “material weaknesses” or not, the dividends presumably can’t be faked nor can cash and ST investment balances (but with Parmalat who really knows!). Furthermore, it’s hard to imagine that the board wouldn’t have been forced to make a disclosure of an “investigation” of this nature if any such thing had in fact been commenced already, but again, I’ve learned the hard way not to underestimate how dumb and/or dishonest some boards and managers can be, and SRT’s board hasn’t exactly handled recent communications well. The upshot is that the “other shoe” fear is a real one, though IMHO it is overblown and already priced into the stock. Finally, SRT has increased its quarterly dividend by one penny each for six consecutive quarters since it instituted the dividend in 2003, most recently earlier this month (and a scant few weeks before the CEO announcement, which indicates to me that if something really is amiss, it has come to light within the Company only very recently).

SRT is an outsourced service provider, principally to telecom companies, whose business process services include provisioning management, wireless telephone number porting, receivables management, wireless telephone activations, and technical support and customer care services. The Company also provides supply chain management services including packaging, fulfillment, marketing support and logistics services, though it is a much less significant piece of the business. SRT operates roughly 17 call centers in the US and Canada. Public comps include West Corp, Sykes (which I wrote up on VIC in April 2004 at $6.24; current price $7.51), TeleTech, and Convergys. West is the class of the group, with operating margins north of 20%, and historically has traded at a premium to the group. SRT also has better-than-average margins (13-14%) reflecting its focus on higher end customer care applications. SRT has enjoyed double-digit revenue and earnings growth since its IPO in 1997. There is considerable information available on SRT’s website (www.startek.com) so I will not belabor the factual stuff.

SRT, like many of these companies, has very significant customer concentration, with Cingular accounting for 42% of revenues, T-Mobile 25%, and Microsoft and AT&T another 10-11% each (for a total of roughly 87% for the top 4). Although the Company is working hard to diversify its revenues, this concentration obviously represents a significant ongoing risk. In fact, as recently as 2001, Microsoft represented over 40% of revenues ($88 million), but this business has declined by over $50 million since then (and will disappear altogether soon), so SRT’s overall revenue growth of over $50 million (~25%) over that time span is all the more remarkable (i.e. almost 50% revenue growth in three years from non-Microsoft business). The Cingular business had been with AT&T Wireless and currently is under contract through the end of 2006 (according to the 10-Q, this contract “is not subject to termination for convenience or without cause”). The contract was executed in June 2004 by AT&T Wireless/Cingular, which has its own in-house call centers that are unionized and likely very high cost, so there is a good chance Cingular will continue to outsource the lion’s share of this business. In fact, Cingular increased volumes with SRT in Q3, which has the effect of lowering margins since the contract provides for tiered pricing, with lower unit pricing at higher volume levels (Cingular revenues still managed to grow by 4.5% sequentially).

Set forth below are some summary financial stats on the business; all figures are LTM through September, are in millions except per share data, and are from continuing operations excluding nonrecurring/extraordinary items.

Revenues $261.8
EBITDA 50.8
Net Inc 27.1
EPS 1.83

Shares Out. 14.5
Price (2/24) 18.45
Market Cap 268
Net Cash 40
TEV 228

TEV/EBITDA 4.5x
P/E 10.1x

The shares had traded as high as $41 last year and came under pressure first from significant insider selling as the founder (and current Chairman) dumped nearly four million shares (almost half his stake) last year through a secondary offering in June at $33. The third quarter earnings then disappointed, with the stock drifting in the $26-28 range until the recent plunge. Although SRT grew the top-line by 14% in Q3, gross margins compressed significantly (~500 bps). However, about half the margin compression is attributable to increased costs that are one-time in nature and without which EBITDA would have been nearly flat yoy—disappointing but hardly disastrous. Of course, we need to carefully consider Q4 earnings and any other disclosures that come with them in the near future.

On that front, there has been rampant speculation about further bad news (just see the Yahoo message board over the past week), but at this point it is just speculation, and further clarity will have to await the earnings announcement, which I believe is scheduled for March 3, though even this is not entirely clear. The volatility in the stock is no doubt exacerbated because of its relatively small float and, until recently at least, a decent sized short position (though it stands to reason that many of the shorts have now covered). Of course, this volatility can work to the upside should the Company allay investors’ worst fears. Moreover, the Company is in its pre-earnings release quiet period so presumably cannot be buying in shares under the $25 million share repurchase program announced late last year, but Company buying should provide a floor for the stock, if not a lift, once SRT announces earnings and can purchase shares again. Given the risks inherent in this business, it never should have been a $41 stock, but at the current price, the risk/reward proposition appears pretty favorable.

Catalyst

Relief rally once Q4 earnings are announced with no further bad news
Company stock repurchases
Maintenance of juicy 9% dividend yield