In today's ZIRP current income-starved world, WIN's 12% dividend yield indicates significant doubt regarding the sustainability of WIN's dividend and should be a red flag to any investor. Telecom investors have been scarred by the dividend cuts at FTR and ALSK and on the face of it WIN is just another RLEC whose business is in decline and whose dividend cuts are in the future. The stock performed poorly early in the year despite investor fervor towards dividends, and caught the post-election fiscal cliff selloff in dividend paying stocks anyway. As a result, the shorts who have been piling into WIN thinking it is another FTR or ALSK have made money without any hint of a dividend cut so far. However, I believe that WIN is actually a quality company whose management anticipated the difficulty in the RLEC business model and proactively transformed the company into an enterprise and broadband focused telecom on the cusp of showing revenue growth. A careful analysis indicates that the dividend should be sustainable for the foreseeable future and management has adamantly asserted the $1.00 dividend will be paid. More importantly, management actually seems to believe their own rhetoric about the dividend and have put their money where their mouth is with large insider purchases.  In my opinion the shorts have overstayed their welcome and may feel the pressure to cover given the now 12% annual carrying cost of a short position. Taking profits on a short seems all the more compelling after analyzing various potential scenarios, which indicates that a significant dividend cut is already priced in but 80%+ upside remains in the event that the market gains comfort around the sustainability of the dividend. I believe this could happen by Q3 or Q4 of 2013. Given this asymmetric risk/reward, I think WIN is a great long with a 12% tailwind that will accrue to investors even if nothing is resolved and the same controversy is raging a year from now.

Windstream was created in its current form in July 2006 after Alltel decided to focus solely on wireless, spinning off its landline division and merging it with Valor Communications. Windstream began life as a traditional rural incumbent local exchange telecom provider ("RLEC"), primarily serving residential customers in mostly rural areas of 16 states and owning a mere 24,000 route-miles of fiber. From inception, the company realized that it would face challenges maintaining its business in the face of wireless substitution for traditional land line local and long distance service as well as cable encroachment in broadband. Windstream dealt with these challenges first by increasing its scale through acquisitions of smaller RLECs, and then beginning in 2010 completed a series of acquisitions of business-focused competitive carriers. Windstream purchased NuVox in early 2010, marking its first move into IP-based competitive business telecom services. Windstream followed the NuVox acquisition by purchasing datacenter operator Hosted Solutions, regional transport provider Kentucky Data Link and CLEC Norlight, further expanding its presence in IP-based enterprise services and expanding its fiber-based network to make the NuVox business more profitable. In November 2011 Windstream made its largest acquisition to date, buying leading national competitive business telecom services provider PAETEC for over $2B. In the process, Windstream has transformed itself from a regional RLEC to a national service provider with operations in 48 states, 115,000 fiber route-miles and 21 data centers offering state-of-the-art managed hosting and cloud computing services. Most importantly, revenue concentration in its targeted strategic areas of business telecom and consumer broadband has increased from 42% in 2008 to 69% in Q3 of 2012. This portion of revenue grew 2.7% on a pro forma Y on Y basis last quarter, cutting the year on year organic revenue loss to less than 1% indicating that the return to organic revenue growth may be right around the corner. The less strategic consumer voice and wholesale revenue shrank 7.7% Y on Y.

 

To get more granular, Windstream divides its revenue into business services (58% of last quarter's revenue), which can be further subdivided into voice & long distance (21% of revenue), data and integrated services (25% of revenue), carrier (10% of revenue) and miscellaneous (2% of revenue). Note that within business services voice and long distance is declining modestly while the data and integrated services (which includes managed hosting and cloud computing services) and carrier (which includes fiber connections to cellular towers) are growing high single digits. Carrier revenue should accelerate to double digit growth in 2013 as the cellular towers that Windstream has been spending considerable capex to connect with fiber are turned on. Consumer services (22% of last quarter's revenue) can be divided into voice & long distance (12% of revenue), broadband (7% of revenue) and video/misc (2% of revenue). Voice & long distance has been shrinking at high single digits while broadband revenue is growing low single digits, resulting in overall consumer revenue shrinkage in the low single digits. Wholesale revenue (14% of last quarter's revenue) can be divided into Switched Access (5% of revenue), USF fees (7% of revenue) and other categories representing about 2% of wholesale revenue. Wholesale revenue should decline going forward, but a large part of the estimated 9% pro forma decline for 2012 is the result of one-time issues in the Switched Access category that I will discuss later. Given scheduled reductions in inter-carrier compensation, wholesale revenue declines should continue in the high single digits in 2013 and beyond. Finally, about 6% of revenue comes from product sales and other.

 

This year has been a rough ride for WIN investors. In February 2012 the company reported Q4 2011 results and guided Wall St. to expect $6.18B to $6.31B in revenue and $2.4B to $2.5B in pro forma OIBDA (company defined, includes impact of stock based comp). While current consensus expectations of $6.19B of revenue and $2.4B of OIBDA are not terribly different from initial guidance, a number of issues have made investors skittish this year and translated into a YTD 30% stock price decline. In the Q1 call, the company revealed that a PAETEC switched access wholesale product had received an adverse FCC ruling and would be discontinued, eventually leading to a drop in wholesale switched access revenue from $114M in Q4 2011 to $86M in Q2 2012. Switched access revenue is virtually 100% margin so this led the company to direct investors to the lower end of their OIBDA guidance, made investors question whether there were other shoes to drop at PAETEC and generated a stock price decline to the mid $9 range. The Q2 call was mostly uneventful as the company announced a restructuring to save costs but the quarterly dividend substantially exceeded the free cash flow. In Q3 Windstream reported yet another quarter where the dividend payment exceeded the cash flow, bringing the YTD dividend payout ratio to over 100%, and told investors it might be just below the low end of the OIBDA guidance. Despite the modest nature of the potential OIBDA miss and the highest sequential revenue improvement to date, the stock declined even further.

 

One key thing to note, and central to the bear thesis, is the difference between what the company defines as "adjusted free cash flow" and the free cash flow as you would calculate it from the cash flow statement in the SEC filings. According to the company's definition of adjusted free cash flow, the payout ratio is only 66% YTD, because adjusted free cash flow excludes one-time restructuring, merger integration expenses, merger integration capex, working capital changes and non-cash pension and stock based comp expenses. Besides the one-time merger integration expenses and capex and restructuring this year, the company is currently going through a heavier capex period as it connects cellular towers with fiber, builds out further rural broadband infrastructure with grants from the federal government and builds new data centers. The company has also suffered from uncharacteristically negative changes in working capital this year, making the entire picture worse. I believe the dividend payout ratio will continue to look very high for the next 3 quarters, but by Q3 2013 the payout ratio should be back to a sustainable level, potentially providing a catalyst for the stock. More detail on this issue below.

 

 

One thing you will note if you investigate WIN is the significant rise in short interest. What's interesting about the short interest in WIN is that it seems to be increasing, not decreasing, as the price declines. There were about 31M shares short at the beginning of the year when the stock was near $12 and 72M shares are short as of the end of October after a 30% YTD decline. My theory is that the shorts have been emboldened by their success at FTR and ALSK, which have continued to decline even after the initial dividend cuts. However, I believe they are wrong on WIN, the dividend will not be cut in the next 12 months, and at some point the stock will benefit from significant short covering as shorts realizing they are stuck paying a $1.00 annual dividend in anticipation of a dividend cut that will not occur. In this section, I will attempt to address the bear case point by point.

 

This was largely addressed above, but WIN has significantly changed its business mix. Currently nearly 70% of revenue is coming from business telecom services and consumer broadband, both growing industries. These segments collectively grew nearly 3% Y on Y last quarter. The declines in consumer and wholesale revenue more than erased those gains, but as the growing portion of revenue becomes a larger percentage of overall revenue, growth will eventually follow. Revenue grew sequentially in Q3 and is forecast to increase sequentially again in Q4. Q4 could be the quarter where the company crosses over into overall revenue growth, and the company has a shot at overall revenue growth in 2013. The decline curve of Windstream's consumer business is well understood at this point and has some of the best trends in the industry due to the relative age of WIN's customer base and the lack of good cable and wireless alternatives over much of their footprint.

 

Wholesale revenue has been WIN's fastest-declining sector, and delivered investors a nasty surprise in the Q1 earnings report when Windstream revealed that a PAETEC product classified under "switched access" was being discontinued after an adverse billing ruling, affecting approximately $10M per quarter in high-margin revenue. This is the primary reason why Windstream will come in at or below the low end of its OIBDA guidance range offered in the Q4 2011 call. Bears circulated a story of "other shoes to fall" at PAETEC after this, but so far none has and the company claims it has no further exposure. However, due to inter-carrier compensation reform instituted by the FCC, Windstream had to step down its access rates in July of 2012 and further reductions are mandated each July for several years afterward. To put this in context, I estimate wholesale switched access revenue will decline by about 21% in 2013, from around $340M this year to $270M in 2013. While this is mostly 100% margin revenue and represents a real headwind in 2013, it is important to keep in mind that (1) we are talking about only $70M of OIBDA or roughly 3% of a $2.4B base; (2) this will partially be offset by reductions in cost of services as the rates that WIN pays to other carriers will also be reduced. So switched access revenue will remain a headwind, but I believe a manageable one.

USF funding is a bit more complicated but less of a risk. The company will receive about $400M in USF funds this year, which roughly break down into $100M of Federal USF funds, $130M of state USF funds, $20M from a new federal broadband USF initiative known as the "Access Recovery Mechanism" or ARM and $150M of zero margin pass-through revenue that is collected by Windstream and remitted to the federal government. Obviously, potential reductions in the pass-through revenue don't matter. Federal USF funding is currently frozen at 2010 levels pending potential reforms, although given how late we are in the year nothing is likely to happen in 2012 so it will likely be frozen at these levels in 2013 as well. There is a possibility that reform will happen in 2013, but keep in mind the impetus behind the reform is to redirect funds to subsidize broadband instead of voice lines. Given Windstream's footprint as one of the leading rural DSL providers, the company is likely well positioned for any FCC reform and could actually increase federal USF revenue in the context of reform. As for the state USF funds, Windstream has publicly said it expects a slight decline driven by reductions in USF subsidies from Texas ($90M of the $130M or so in state USF funding) but that these reductions will largely be offset by service price increases.

 

This is the crux of the Goldman research note downgrading WIN to sell and lowering the analyst's price target from $11.50 to $7.50. In effect, the analyst argues that as the higher margin residential voice and regulatory revenue declines and it replaced with lower margin business services revenue, the OIBDA margin (and OIBDA in a flattish revenue scenario) will be under pressure. This makes some sense on the face of it, as back in 2010 WIN enjoyed near 50% OIBDA margins and PAETEC historically had OIBDA margins in the high teens. In fact, this is maybe the most cogent bear argument and represents the most difficult execution challenge for the company. The analyst makes some calculations and posits that ex-synergies and cost reductions, WIN's OIBDA margin has been under pressure and this will start to show later in 2013 and 2014 after the benefits from the PAETEC merger synergies and cost cuts fade.

However, Windstream has been dealing with falling high margin revenue for several years now, and so far has been up to the challenge. Digging deeper, the margins from the PAETEC business stand to get a lot better as the company is integrated by Windstream, as in many markets PAETEC will now be able to deliver its services over Winstream circuits instead of leasing from another wireline provider. I believe the company is targeting the PAETEC sales force to focus on deals where a great deal of th service delivery can take place on-net. Additionally, many of the new business services Windstream is selling, such as fiber-to-the-tower and managed hosting, carry OIBDA margins as high as legacy residential voice. 

As an aside, to me the Goldman note seemed like a classic sell-side closing of the barn door after the horse is out. The analyst's reductions in OIBDA were extremely minor compared to his 35% cut in target price and in the piece he admitted a dividend cut in 2013 was unlikely. The stock had already made 85% of the downside move contemplated in the price cut, I'll leave you to your own conspiracy theories as to why the Goldman telecom research team finally saw the light and downgraded to sell. There is even less in the recent Piper downgrade to sell (target $6.50), and the justification for the price target is a long-term DCF analysis that includes the questionable assumptions of a 4% UST 10 year rate (does this analyst have access to the Wall Street Journal?) and a 15% "minority discount" (somehow the Piper analysts covering CRM and FIO don't seem to include "minority discounts" in their valuations). The Piper downgrade came out as I was finishing this write-up, and I think the stock's reaction today (up 2%) indicates that at the very least the bear thesis is well known and even some of the investors who are short the stock are using these belated sell-side downgrades as covering opportunities.

At its simplest the bear case is that WIN's dividend is unsustainable, and inevitably going to be cut. Shorting on this thesis was successful at FTR and ALSK, and I believe many of the same investors have been piling into WIN thinking it will follow the same trajectory. In fact, WIN's dividend already looks unsustainable, as the company has reported YTD free cash flow of $434M vs. dividend payments of $441M. Further bolstering the shorts, WIN reports its own version of "adjusted free cash flow" in its IR supplemental information, making it appear that they are obfuscating the truth. Indeed, on a GAAP basis the company has paid out just over 100% of its YTD FCF in dividends, while on management's preferred metric the payout ratio is a high but sustainable 66%. While I am sympathetic to those who want to focus on GAAP metrics and ignore management's "pro forma" results, I think it is important to look at both in context and determine if the divergences are permanent or temporary. In this case there are a number of one-time events that are taking place in 2012 and are not going to recur in 2013. The highlights are:

• WIN purchased PAETEC in November of 2011. PAETEC was a sizable acquisition and has necessitated significant integration opex and capex spending. The company is spending $65M in opex to integrate PAETEC this year, and this figure will fall to $15M next year. Further, the company is spending about $50M in capex to integrate PAETEC's network, while I expect this will fall to only $5M or so in 2013. These two items alone (which are added back in the company's adjusted free cash flow) will improve actual free cash flow and reduce the divergence between GAAP free cash flow and adjusted free cash flow by $95M next year. WIN has made no other sizeable acquisitions since PAETEC so these charges will not recur.
• Carrier revenue is expected to be one of WIN's fastest growing revenue streams in the coming years, driven by the needs of AT&T, Sprint and Verizon to connect their cellular towers to fiber in order to support 3G and 4G broadband. WIN has won numerous contracts to upgrade the cell towers it serves (and some it doesn't currently serve) to fiber. This requires one-time expenditures to lay fiber to cell towers, after which WIN will enjoy recurring high margin revenue for the indefinite future. This is the largest portion of what WIN refers to as "initiative capex", approximately $250M of the $325M of the initiative capex I expect for this year. Next year this spending should decline to $125M, providing a further improvement to both GAAP and adjusted free cash flow (initiative capex is included when the company calculates adjusted free cash flow but broken out separately from core capex as of the most recent quarterly call). In 2014 this should fall to virtually nothing.
• WIN also won stimulus grants from the government to connect un-served rural areas to broadband. The government is paying 75% of the $240M costs, leaving WIN to pay $60M to connect 75K potential new broadband subs. Virtually the entire $60M is being spent in 2012, providing a further improvement to 2013 cash flow.
• WIN does not include working capital in its calculations of adjusted free cash flow because it has historically been a roughly working capital neutral business without a lot of movement given the stagnant top line. However working capital has worked against them by about $147M so far this year, primarily due to the closing of merger related liabilities and prepayments for discounts of certain contracts in Q1 and because the quarter end fell on a weekend (with no collections) in Q3. This should reverse in Q4 but more importantly the events that caused working capital buildup are unlikely to recur next year, although by a quirk of the calendar both Q1 and Q2 of 2013 will end on a Saturday-Sunday. This should improve cash flow by approximately $120M.
• The company announced a management restructuring that should save the company $40M in opex next year at a cash cost of $23M. No major restructurings are expected in 2013.
• I have identified a potential $423M improvement in GAAP FCF next year ($50M PAETEC M&I expense + $45M PAETEC integration capex + $125M less FTTT initiative capex + $60M less broadband stimulus capex + $120M from stabilization of working capital + $23M of restructuring = $423M)  Partially offsetting the above improvements will be a reversal in taxes, as WIN will receive about $93M from the government this year but pay an estimated $256M next year for a swing of $349M. FCF will also improve from around $60M less in cash interest.

The net effect is that I estimate the company will go from a dividend payout ratio of about 71% on its preferred adjusted free cash flow calculation and 99% on GAAP in 2012 to 75% on adjusted free cash flow and 77% on GAAP in 2013. While this dividend payout ratio still looks uncomfortably high, it is important to recognize that it will look very high in Q1 and Q2 as the last of the FTTT capex is made, and then ramp down to show a sustainable-looking 63% of GAAP by Q3 of next year. A table detailing the quarterly company defined free cash flow, GAAP free cash flow and a reconciliation between them is provided below. Showing a mere 63% GAAP dividend payout ratio in Q3 of next year could provide a significant catalyst to the stock, and may put to rest questions of dividend sustainability for the time being.

Management and Insider Behavior

I always try to pay more attention to what a company's management team does rather than what it says. On this front, WIN's management team seems to really believe that they have successfully transformed the business and can sustain the dividend going forward. In every public forum, CEO Jeff Gardner has adamantly stated that the dividend will not be cut. Since May of 2012 he has spent over $760K purchasing 83.5K shares in the open market at prices ranging from $10.00 to $8.27. CFO Anthony Thomas has purchased 35K shares for $340K. Chairman Dennis Foster has purchased 65K shares for over $621K. Maybe they were all just sick of earning less than 20 bps in their savings accounts and foolishly reached for yield. Clearly, many of the purchases were too early. However, all these executives are respected industry vets involved with Alltel prior to Windstream so they know where the bodies are buried.

Regarding a potential dividend cut, many investors seem to believe that a dividend cut may become a self-fulfilling prophecy if the stock price gets low enough, reasoning that a certain yield the company may decide it's cheaper to cut the dividend and buy back stock or de-lever. Other than the insider purchases and unequivocal public statements by Jeff Gardner, I would also point you to recent history. WIN's stock spent the financial crisis trading around current levels and even dipped below $7. The company had plenty of time to cut the dividend due to yield or just general fear of economic conditions, but did not. And this was before the company had a line-of-sight to eventual revenue growth.

It's also worth noting that the company has fairly liquid debt. All of the company's major debt issues trade over par according to Bloomberg and fixed income investors seem unconcerned with the company's financial stability. This is a great time to be an issuer of high yield debt and the company has access to plenty of liquidity. The company has $1.5B of debt capacity in the form of a secured facility, more than enough to pay off the $1.1B of maturities in 2013 although I personally think they should term out the debt and lock in these interest rates for a long time. They could use the revolver to pay the dividend if they had to and their IR has told me they have gotten zero pressure from the credit rating agencies to reduce the dividend.

Scenario Analysis and Conclusion

To simplify things a bit, I think there are three basic scenarios that could play out:

• The bullish case I outlined, where the market gains comfort in the sustainability of the dividend during the 2H of next year after the one-time cash flow hits from PAETEC, restructuring and discrete initiative capex are gone. If WIN goes to a 7% dividend yield (similar to CTL today) on a $1.00 dividend, the stock would be at roughly $14.00. Add in the $1.00 in dividends an investor would collect over the next year and you are looking at a total return of 77%. Note that at $14.00 the stock would be at a roughly 10% FCF yield on my GAAP 2013 FCF per share estimate.
• The bearish case outlined most recently by Piper, where the stock reaches $6.50 for a decline of around 23%. However, if this took a year investors would collect a $1.00 dividend in the interim, significantly offsetting the capital loss and turning the total loss to only 11% or so. And let's say after a year the dividend is cut by 40% to $0.60 annually. The stock is already trading at a greater than 7% yield under that scenario. In my opinion, a significant dividend cut is already priced in, although I don't think one will be forthcoming. If it was, I think the stock could actually be flat to up as it would ease questions about dividend sustainability and induce some shorts to cover. Even if the stock traded to a 9% yield (similar to FTR) on a $0.60 dividend, that is fully captured in the $6.50 bear case price.
• The company could also muddle though and a year from now Wall St. could still be wondering about the dividend. If the stock is flat investors will realize a 12% return from the dividend alone, not too shabby in today's environment.

While I think the bullish case is most likely to play out, averaging across the three scenarios still yields a significantly positive expected return of 26% (average of 77% bullish case + -11% bearish case + 12% status quo). To me, the risk/reward clearly favors being long. If the shorts ever decided they needed to get out of the way or get sick of paying a 12% annual dividend to be short, I think WIN could move up in a hurry.

• 2013 convergence between "adjusted free cash flow" and GAAP free cash flow undercuts the bear thesis
• Return to a 60% dividend payout ratio on GAAP free cash flow in Q3 2013
• Fiscal cliff negotiations avoid a large increase in the dividend tax rate and/or produce a reduction in corporate tax rate
• Shorts get tired of paying a 12% dividend on top of their rebate