Description
Atlantic Tele-Network (AMEX: ANK) is a Caribbean-based telecom monopoly available at an EV-to-earnings of 7x, a price-to-book of 1.34x, paying a dividend yield of 3.7% and with a six-year track record of increasing dividends. The past three dividend hikes have been announced in September, so we may be only a few months away from another boost to the yield.
ANK’s primary operations are located in Guyana (on the northern coast of South America), an economically poor country with questionable government. As unsettling as that may seem, ANK has a unique stronghold in its market -- an 80% interest in Guyana Telephone & Telegraph (GT&T), Guyana's monopoly provider of local, long-distance, and international telephone service. And who is the other 20% partner, you ask? None other than the government of Guyana itself. ANK’s relationship with the Guyanese government hasn’t been without its squabbles, but nothing helps to strengthen a monopolist's position than having the government in on the take.
My core investment thesis here is that the market has overestimated the risk of operating in Guyana, given that most of ANK’s profitable business is earned in US$ and other hard currency. ANK is priced as if it’s a stagnant or even declining business, but in fact the company has produced steady organic growth including EBITDA growth of 8% annually from 2000 to 2004. Over the past five years, net margins have averaged just over 13% and ROIC has averaged 18.5% -- and this in spite of a 55% tax rate and some unprofitable business lines. Finally, unlike most telcos that are typically laden with debt, ANK has net cash of $37 million ($7.39/share) representing 25% of the market cap. ANK also has some hidden assets, most notably a minority ownership interest in Bermuda Digital Communications, the leading wireless provider in Bermuda with over 20K customers and solid profitability.
Another major tenet of my thesis is that ANK is led by a heavily-incentivized owner-operator, Chairman & CEO Cornelius Prior Jr., who owns 58% of company. His communication with investors is consistently thorough and conservative-leaning. I also like the fact that stock options are practically nil.
I’ll warn you up-front that ANK has limited liquidity. With ownership that’s 59% in the hands of insiders, that leaves just 2.05 million shares of float. Average daily trading volume is a little over 2K shares, so this idea may only be of interest to smaller funds or those with individual accounts.
BUSINESS OVERVIEW
Headquartered in St. Thomas, U.S. Virgin Islands, Atlantic Tele-Network was established in 1987 and went public in 1991. It was in that year the company bought its 80% interest in GT&T, which had previously been 100% owned by the government. The terms of the deal (known as the “GT&T Agreement”) provided Atlantic Tele-Network with an exclusive 20-year franchise to provide Guyana with public telephone, radio telephone, pay station telephone, and national and international voice and data transmission. This license is renewable for an additional 20 years at the option of GT&T, so in effect this extends ANK’s monopoly rights to 2031. Even at the end of that period, the GT&T Agreement calls for the government of Guyana to buy back ANK’s 80% GT&T interest at a price to be determined by international arbitration, so that alleviates any serious concern of ANK being left high and dry in the end.
Since ANK acquired its interest in GT&T in January 1991, GT&T has significantly rebuilt and further expanded its telecommunications network. The number of fixed access lines has increased from approximately 13,000 working lines in January 1991 to over 103,000 lines as of December 31, 2004, all of which are now digitally-switched. GT&T first introduced cellular service in 1992 and has expanded its cellular service dramatically in recent years. In the fourth quarter of 2004, GT&T launched services on a GSM overlay across most of its existing TDMA wireless network.
The most lucrative portion of ANK’s business is GT&T’s international long-distance (ILD) segment, which had 2004 revenue of $46.9 million and gross margins near 90%. This segment by itself accounts for more than 100% of ANK’s gross profit (due to other segment losses). The ILD segment bears some explaining, as there are very different profit characteristics of in-bound versus out-bound traffic. In-bound has near 100% gross margins, whereas out-bound has very minimal gross margins. Fortunately for ANK, its ILD traffic mix is heavily weighted towards in-bound calls (I’ll explain why in a minute). The margin difference is due to the following: For in-bound traffic GT&T has no significant direct expenses except for international transmission systems costs which are applicable to all of GT&T’s international traffic. In contrast, for out-bound traffic GT&T must typically pay the foreign carrier a settlement rate payment equal to one-half of the applicable international accounting rate, and GT&T collects from its subscriber a rate that is regulated by the Public Utilities Commission of Guyana, which has in the past provided for only a minimal profit.
About 85% of overall ILD traffic is in-bound to Guyana from the US, Canada, and UK. One very important benefit here is that these in-bound payments are denominated in “hard currency,” principally US$. So for those of you worried about the currency risk of investing in Guyana, take note: The vast majority of the ILD income stream, and therefore the vast majority of ANK’s overall profits, are in US$. This significantly alleviates the currency risk for US investors. (By the way, all of ANK’s financial results are stated in US$, which is the company’s functional currency.)
So why is the ILD mix so heavily weighted towards in-bound traffic? These in-bound calls primarily consist of personal calls between Guyanese expatriates and their friends and family in Guyana. By way of context, CEO Cornelius Prior Jr. told The Wall Street Transcript in June 2003 that “…there has been a large diaspora that presently results in a Guyanese population in New York City equal to the population of the capital city of Guyana, Georgetown.” And as you might expect, the expats overwhelmingly tend to initiate these calls because they’re so much more affluent than the average Guyanese resident.
As an aside, ANK’s in-bound traffic has sometimes been rumored as being the beneficiary of 1-900 sex calls. Like most rumors, there’s an element of truth in this accusation, but only in a historical context. As late as the mid-1990s, ANK received a significant portion of its ILD revenue from “audiotext providers.” As stated in the company’s 2000 10-K, “Audiotext providers offer telephone information services comparable to those available in the United States on an area code 900 basis. By making a telephone call, the caller can obtain information (generally in the form of a recorded message) on subjects such as weather, sports, business news or material of a sexual nature.” But the 2000 10-K also says: “However the volume and profit margins of this traffic sharply declined in the past three years and audiotext traffic no longer contributes significantly to the profitability of the Company. Management attributes the continuing decline in the audiotext traffic to increased competition from other terminating country carriers, from domestic audiotext traffic and from the internet.” So this audiotext-related revenue was already history as of 2000.
Below is a table that presents a five-year snapshot of ANK’s ILD segment revenue and gross profits. You’ll notice the sharp decline in revenue in 2002. This is because on January 1, 2002, the US Federal Communications Commission reduced the settlement rate for US-Guyana traffic from $0.85 per minute to $0.23 per minute, where it has remained ever since. As you can see in the table below, the 2002 rate reduction initially caused a severe blow to the ILD business, but while also spurring a much higher call volume which eventually helped to offset the lower rate. In addition, ANK has benefited from lower telecom costs and a continued mix shift towards in-bound calling, both of which have been influential in driving up gross margins in this segment (all figures in millions):
ILD Segment 2000 2001 2002 2003 2004
Revenue $62.4 $62.5 $39.7 $42.0 $46.9
Gross Profits $42.0 $45.8 $30.4 $34.9 $41.8
Gross Margin 67.4% 73.3% 76.5% 83.1% 89.2%
Inbound Minutes 77.1 77.7 100.2 124.3 150.1
Outbound Minutes 19.0 19.6 18.9 25.6 27.1
Total Minutes 96.1 97.2 119.1 150.0 177.2
% Inbound 80.2% 79.9% 84.2% 82.9% 84.7%
ANK’s second segment is its GT&T Local Exchange business, which had 2004 revenue of $37.1 million but with a gross margin of only 13.7%. The low profitability of this segment is due to the Guyana Public Utility Commission’s stubborn refusal to allow ANK to price its service at a level commensurate with the 15% minimum rate of return provided for in the GT&T Agreement. That’s been a point of contention for years and unfortunately shows no sign of being resolved anytime soon. As such, this segment will likely remain a modest profit contributor – but there are a few opportunities for improvement.
One opportunity is for GT&T Local to at least gain some scale as it expands is subscriber base, which has grown at a 34% compound annual rate over the past four years. GT&T’s total base of wireline and wireless customers was 254K at year-end 2004. If nothing else, these customers provide a wider audience of potential in-bound ILD business, on which ANK can make good profits.
A second growth opportunity for the GT&T Local business is its new GSM/GPRS mobile wireless network, which launched in Q4 ‘04 alongside its existing TDMA network. GSM/GPRS is said to be a more advanced mobile wireless digital service than TDMA, allowing GT&T to offer richer handset features and certain mobile data services, while increasing GT&T’s network capacity. The revenue opportunity here is that the GSM network has allowed GT&T to enter into roaming agreements with wireless carriers in a number of other countries, including some of the largest carriers in the U.S. and Caribbean, enabling GT&T’s subscribers to use their handsets in other countries and allowing some visitors to use their cell phones while in Guyana. GT&T expects to enter into roaming agreements with additional carriers in 2005. While Guyana is by no means a magnet for affluent foreign travelers (like Bermuda, say), these roaming agreements certainly present at least a modest incremental growth opportunity.
Below I present the GT&T Local segment results (in US$ millions) and subscriber counts (in thousands):
GT&T Local Segment 2000 2001 2002 2003 2004
Revenue $14.2 $21.2 $31.1 $36.8 $37.1
Gross Profits -$8.6 -$6.3 $2.2 $6.2 $5.1
Gross Margin -60.7% -29.9% 7.1% 16.7% 13.7%
Wireline Subs 71.7 79.9 86.2 92.7 103.3
Wireless Subs 7.8 39.2 79.9 118.7 150.9
Total Subs 79.5 119.1 166.2 211.3 254.1
ANK’s third segment is called Other Operations, and it includes a hodgepodge of development-stage Caribbean businesses which have perpetually lost money. In 2004, the operating loss in this segment was $6.3 million, up from a loss of $5.5 million in 2003. I’ve always believed management could unlock some value simply by closing down these money pits. On this score, I counted it as very good news when earlier this year ANK announced it would cease operations for two of the three subsidiaries in this segment (among those closed, one was a call center operation, the other a collect call business – both clearly inferior businesses). These eliminations should go a long way towards stemming this segment’s losses.
The one remaining “other operation” is actually pretty interesting. Choice Communications provides wireless cable TV (wCATV) and internet access throughout the US Virgin Islands. Choice is still losing money, and will probably continue to do so in 2005, but at least it represents a business with unique assets and good growth potential. Choice’s 2004 revenue increased 15% to about $5.1 million, but was unprofitable due to investments in future growth. Choice television subscribership increased by over 40% in 2004 while its internet dial-up subscribers decreased by about 17%. The internet dial-up business is naturally facing decline, but the company plans to introduce new technology in 2005 to support broader deployment of its wireless DSL (wDSL) services. Over the past year, Choice has significantly expanded its wireless network service area, and this expanded network infrastructure will help support continued growth in both wCATV and wDSL services.
Finally, there’s the hidden asset I mentioned earlier, which shows up below the operating income line as part of “other income.” ANK has a 44% minority ownership interest in Bermuda Digital Communications (BDC), the largest cellular telephone service provider in Bermuda, doing business under the name Cellular One. ANK paid a total of $2.2 million cash for this interest, which was purchased in 1998 (including warrants which were exercised in 2002). In 2004, BDC contributed $2.57 million in equity earnings, up 26.5% YOY. BDC accounted for nearly 8% of ANK’s 2004 pre-tax income. BDC ended the year with over 20,000 subscribers, up 12% YOY versus approximately 17,800 subscribers at the end of 2003.
FINANCIAL RESULTS / CASH FLOW
Here’s a snapshot of ANK’s overall financial results from the past five years ($ millions):
ANK Income Statement 2000 2001 2002 2003 2004
Revenue $80.4 $88.1 $74.7 $83.3 $89.3
Gross Profits 32.1 36.0 27.6 35.6 40.6
EBIT 27.7 30.3 21.8 29.0 34.0
Tax Rate 49.4% 54.3% 52.1% 50.4% 54.9%
Net Income $12.3 $9.2 $9.5 $12.2 $16.0*
Diluted Shares (M) 4.91 5.01 5.07 5.03 5.03
EPS $2.51 $1.83 $1.87 $2.43 $3.19
(* = 2004 Net Income is after adding back two non-cash charges – an asset impairment and an uncollectible loan charge – totaling $3.9 million.)
ANK Cash Flow 2000 2001 2002 2003 2004
OCF $27.8 $28.4 $29.2 $26.5 $41.6
Capex 18.3 21.4 15.5 16.2 25.3
D&A 8.1 9.4 10.8 12.5 14.7
EBITDA 35.8 39.7 32.6 41.5 48.7
FCF (OCF – Capex) 9.5 7.0 13.7 10.4 16.3
Dividends 3.5 4.0 4.2 4.6 5.1
As you can see in the cash flow table above, ANK has been a consistent producer of FCF. This has been in spite of heavy growth capex in building out Guyana’s telecom infrastructure. Notice specifically that capex has significantly exceeded D&A each of the past five years. ANK’s CEO has said that the company’s D&A is sufficient to fund the upgrade and maintenance of its existing capacity, so any capex in excess of ANK’s D&A can rightfully be classified as growth capex.
Capex was particularly high in 2004 at $25.3 million (versus D&A of $14.7 million) due to the build-out of GT&T’s new GSM wireless network. With those investments now complete, I expect capex over the next few years should fall back in-line with D&A. For 2005, ANK is anticipating capex of $15-18 million, which should be roughly equal to D&A. This means FCF in 2005 and going forward should more closely approximate earnings.
VALUATION
I value ANK based on its earnings which are about equal to FCF not including growth capex or working capital fluctuations. TTM earnings thru Q1 ’05 were $15.9 million. (This is after adding back non-cash, one-time charges of $3.9 million, which includes a $3.3 million asset impairment charge and a $0.6 million loan write-off). At a current stock price of $29.23, ANK’s market cap is $146.6 million. Less $37.0 million in net cash, the EV is $109.5 million. That puts the EV-to-earnings at 6.9x.
It gets better though. Now that ANK has pulled the plug on two of its money-losing “other operations,” I think there’s a good chance of that segment getting to breakeven within the next few years. If ANK can eliminate those losses, that alone will provide a nice boost to earnings. To quantify this possibility, consider that “other operations” losses over the TTM thru Q1 ’05 were $6.6 million. At a 55% tax rate, this is lost earnings of nearly $3 million. Adding that back gets us to total earnings power of $18.9 million. On that basis, the EV-to-earnings is only 5.8x.
What’s the stock worth? Even on a no-growth basis, I think a 9x multiple is justified (i.e. 11% returns on a perpetuity basis). On current earnings of $15.9 million, that equates to a stock price of nearly $36 (math: EPS of $3.17 x 9, plus net cash of $7.38/share). On potential earnings of $18.9 million assuming no further losses from “other operations,” that would be a stock price of $41 (math: EPS of $3.77 x 9, plus net cash of $7.38/share). So on a conservative basis, I see the fair value range as $36 - $41.
DIVIDENDS AND BUYBACKS
I mentioned at the outset that ANK has paid dividends at an increasing rate for the past six years. I also mentioned that ANK has typically raised its dividend in September. Here’s a chronology of the company’s dividend actions:
Dec ‘98 – Adopted policy of paying quarterly dividends of $0.15 per share
Mar ‘00 – Increased quarterly dividend to $0.175 per share
Mar ‘01 – Increased quarterly dividend to $0.20 per share
Sep ‘02 – Increased quarterly dividend to $0.225 per share
Sep ‘03 – Increased quarterly dividend to $0.25 per share
Sep ’04 – Increased quarterly dividend to $0.275 per share
Given ANK’s solid ongoing profitability and bountiful excess cash, I see no reason there shouldn’t be another dividend hike come September.
As for share buybacks, ANK has made occasional repurchases but it’s only amounted to $2.4 million over the past five years. The company did, however, authorize a new $5 million program in September 2004 when the stock was around $26. No shares were repurchased in the two reported quarters since then, but I’m hopeful we might find out that the company bought back some shares in this current quarter.
RISKS
I see three main risks to ANK’s business:
1) Any further reductions in the US-Guyana long-distance settlement rate would hurt ANK’s ILD business. I have no reason to suspect this might happen, but it can’t be ruled out. As mentioned earlier, the FCC last reduced this rate in January 2002, from $0.85 to $0.23 per minute. Since then, according to ANK’s 10-K, in 2002 and again in 2003, AT&T proposed further reductions in the settlement rate benchmarks for many low-income countries, including Guyana, and requested that the FCC initiate a rule-making to consider the issue. In early 2004, the FCC rejected AT&T’s request but indicated that it will continue to monitor and evaluate settlement rate benchmarks. There’s been no further word on the issue.
2) Internet telephony also poses a potential risk to ANK’s ILD business. So far, however, there’s not been any visible decline in the company’s long-distance minutes. In 2004, total long-distance minutes increased 18.8%. Even in the latest quarter, which ended March ’05, total long-distance minutes increased 8.8% YOY, including an 8.1% YOY increase in the highly profitable in-bound traffic. As another check on this risk, I googled “call to Guyana,” and it produced tons of offers for discount phone cards, but nothing about free Internet calling.
3) A lesser concern is the likelihood of increased wireless competition in Guyana. Right now, GT&T Local faces only one other wireless competitor in Guyana. While no new competition is imminent at this point, it does look like a possibility within the next few years. Apparently, a regulatory body in Guyana has proposed that the GSM spectrum be divided from two to four bands, which, if approved, would open the door to two new competitors. But keep in mind that ANK derives less than 15% of its overall gross profits from this segment, so even successful competitor encroachment would have only a limited impact on net profits.
CONCLUSION
ANK at 6-7x earnings and paying a 3.7% yield that’s likely to increase over time represents a solid risk/reward opportunity. The company’s rock-solid balance sheet leaves open the possibility of a significant share buyback or large one-time dividend as well. Finally, CEO Prior’s 58% stake in the company ensures that shareholders have a watchdog who will do everything possible to maximize ANK’s value.
Catalyst
1) Possible dividend hike in September
2) Elimination of unprofitable segments
3) Share buybacks