Description

Internet Gold Golden Lines (NASDAQ:IGLD) is a leveraged play on Bezeq, Israel’s incumbent telecommunications provider. In fact, it’s a little more complicated than that -  IGLD is a leveraged play on B-Communications (NASDAQ:BCOM), which is the actual controlling shareholder of Bezeq.

Two layers of publicly traded holding companies, both with debt and minority interests, is not something most value investors usually are excited with. It’s definitely a complex story with a lot of moving parts, and the controlled operating business at the bottom of this pyramid is a highly regulated, former-monopoly, telecom giant, that is in the middle of political and regulatory scrutiny.

So why bother? Well, the other side of that story is that you can get an 18% IRR over the next 3 years simply by watching the holding co. de-lever, even as Bezeq’s shares remain flat for the entire period.

IGLD is cheap, trading at 32% discount to its “real” NAV with some catalysts that should assist narrowing this discount in the next few years (including a hard catalyst that is still 3 years away - but more on that later). In addition, it may be an unpopular opinion these days but we still believe Bezeq is the highest quality & best managed telecom group in Israel, and we don’t see why this should change over the next several years.

Our base case, the one that results in the ~20% IRR for IGLD, has Bezeq delivering results that are below current sell-side consensus. In a more bullish scenario in which Bezeq executes according to current estimates, and the deleveraging of the holding structure is played as we think it will, IRR should be above 34% as IGLD shareholders should enjoy total return of 141% by the end of 2019.   

This thesis consists of two parts. First, we ought to demonstrate why Bezeq is cheap, and why it should at least retain its current earning power for the next few years under conservative assumptions. Second, we’ll discuss some catalysts that should narrow the discount to NAV, justifying an equity investment in a leveraged holding company whose main asset is another leveraged holding company.

Bezeq was written up for VIC a little more than a year ago by ataloma. We think ataloma did a very good job articulating Bezeq’s competitive advantage and defensible earnings, as well as most of the competitive dynamics in the Israeli telecom industry. However, entry price of ILS 8.41 per share now looks expensive in hindsight, as Bezeq’s shares have recently traded at ILS 6.5 (to be fair - price should be adjusted to reflect ILS 0.52 of dividends distributed during 2016). As the matter of fact - our ‘Bull case’ price target for Bezeq shares currently reflects a price per share of only 8 ILS, lower than the price Bezeq was written up at in December 2015, so any upside from there should translate to even higher IRR for IGLD.

We should note that while IGLD is thinly traded on the Nasdaq, it is far more liquid on Tel-Aviv Stock Exchange (“TASE”), so for those with access this is the better way to own the shares. Also, please note that all figures discussed and presented in this writeup will be in Israeli Shekels (unless specified otherwise), as this is the relevant currency for all three companies discussed in this writeup.

Part I - Bezeq

Background

Bezeq Israeli Telecommunication Corporation Ltd. (“Bezeq”) is Israel’s largest telecommunications group. Like many incumbent telecoms around the world, Bezeq was initially a government-owned corporation, before going through a privatization process and eventually being sold to a group of investors led by Haim Saban in 2005. Saban and his partners made a killing on their Bezeq investment, recording a gain of over 300% on their initial investment before selling the controlling stake to Eurocom Group, a private company controlled by Israeli Businessman Shaul Elovitch, in late 2009.

Eurocom structured the deal as a public LBO, buying the controlling stake via two layers of publicly traded holding companies - BCOM and IGLD. At the time, Eurocom paid ILS 8 per share for its stake in Bezeq, approximately 21% higher than where the share is trading today. However, Bezeq distributed almost ILS 15b in dividends during 2010-2016 (approximately 81% of current market cap), allowing BCOM and IGLD to rapidly delever what was originally seen as a shaky holding structure.

It was only four years ago that IGLD’s bonds were trading at YTM of 35%-40%, as investors were skeptics on the group’s ability to serve its huge debt load. The constant stream of dividends and some re-rating in Bezeq’s trading multiples, combined with some share sales across the group (IGLD sold some BCOM shares and BCOM sold some of its Bezeq shares to institutional investors) have most likely resolved the liquidity situation. IGLD’s bonds are now trading for YTM of 1.7%-2.0%, so clearly the market is no longer worried about insolvency. We believe that while IGLD’s debt holders were well payed in the past 3-4 years, the next 3 years would turn up to be be much more lucrative for the equity owners.

Below is the Group’s holding structure as of September 30, 2016:

Business Overview

Bezeq has the most comprehensive telecom offering in Israel, with the ability to offer an attractive quad-play once the structural separation that was forced by regulators is removed (more on that later).  According to management’s most recent guidance. Bezeq is expected to finish 2016 with ILS 10.1b in revenues, 4.2b of EBITDA and Operating FCF (OCF - Capex) of 2.2b. The current share price of ILS 6.6 reflects a 6.8x EV/EBITDA multiple, and 7.6% FCFE Yield.

Bezeq’s services are provided through four main business segments:

Domestic Fixed Line Telephony and Broadband

Through this segment Bezeq provides fixed line telephony and fixed line broadband internet access in Israel. Bezeq’s fixed network is the most advanced in Israel and includes fiber rollout that reaches approximately 1.3m households. Latest quarterly earnings showed average speed of 42 mb/s, more than double the 20 mb/s in 2014. This increase also helped push ARPU from ILS 82 at the beginning of 2014 to ILS 90 today.

Note that in Israel there’s a separation between the provision of internet and internet infrastructure (this structure was put in place originally to prevent Bezeq from achieving monopoly-like market share in the ISP market). Bezeq’s ISP offering is provided through a different segment and a different legal entity, called Bezeq International.

For many years Bezeq’s only competition in the broadband infrastructure market was HOT, a subsidiary of Altice. Compared with Bezeq, HOT has underinvested in its network as it was focusing its efforts on its payTV offering and on a low cost bundle (“Triple Play”). This helped Bezeq take market share in the broadband infrastructure market all the way to 67.6% in 2015. However, this duopoly situation pushed regulators, specifically the Minister of Communications (MoC) to implement a wholesale market freeform process for fixed telephony and broadband infrastructure. While the fixed line wholesale market wasn’t implemented yet as it ran into some technical difficulties, the ministry did set a wholesale broadband price of ILS 51/month in 2014, opening up Bezeq’s infrastructure to competitors - mainly wireless network operators Cellcom  (NYSE:CEL) and Partner Communications  (NASDAQ:PTNR).

The wholesale market only started to gain traction in Q3-15, the first quarter in which Bezeq experienced a y/y drop in the number of retail lines. Overall, since its peak in Q1-15 and until Q3-16, Bezeq lost 13.6% of its retail lines, however this trend seems to be slowing now, with Q3-16 showing only a 6.2% y/y decline and a loss of only 6k lines from the previous quarter.  

More importantly, although the transition of some customers from retail to wholesale hurt the segments’ top line, the ILS 51 price point is relatively high for providing only the infrastructure - with no customer support or acquisition costs needed. Gradual cost reduction in the segment was enough to support further EBITDA growth in 2016, after adjusting for a non-recurring gain from the sale of real-estate recorded in 2015. In fact, while Bezeq is expected lose approximately 5% of its internet retail customers in 2016, EBITDA growth is likely to accelerate to 4.3%, compared with a 2.5% growth in the previous year.

Regardless of the retail/wholesales distribution (which we showed has very little effect on actual EBITDA per line), the broadband infrastructure market in Israel continues to grow at mid-to-high single-digit rates (6.5%-8.5% over the last few years). This is supported by population growth of ~2% per annum, further improvement in penetration rates (which are very high at 77%), as well as high GDP growth (relative to most of the developed world) which supports further expansion of B2B offering. While we are aware that the penetration rates can not improve forever from their already high levels, we think Israel’s growing population, combined with further improvements in average speed, are supportive to the segments’ long-term growth prospects.

Mobile Telephony (Pelephone)

Pelephone, a wholly-owned subsidiary of Bezeq, operates a High-Speed 4G mobile network. Pelephone was long considered the best-managed mobile operator in Israel, in what has been for many years a three-horse race between Partner Communications, Cellcom and Pelephnoe.   

After a long period of excessive profitability between local mobile operators, in 2012 the Israeli government, led by Minister of Communication Moshe Kahlon pushed for more competition. The MoC ran a process to add two more MNOs and several more MVNOs, in addition to the existing three participants. This resulted in an oversaturated market and ARPU declines of ~30% between 2013 and 2015.

Throughout this period, Pelephone performed significantly better than Cellcom and Partner, as it cut almost ILS 800m in OPEX and managed to keep EBITDA margins around 20%-21%, despite the rapid decline in ARPU.

Following a prolonged period of declining fundamentals that started in 2013, several signs now point to ARPU finally bottoming in Q2-Q3 of 2016. First, Golan Telecom who was the “irrational” player who kept pushing prices down to gain market share despite not recording any profits, is now our of the game after selling its activity to Electra - a participant who we (and the market) expect to be much more “rational” going forward.

Second, ARPU was finally flat in Q3-16 on a sequential basis for Pelephone, and up 1.5% for Partner, and  EBITDA increased by 15% (Pelephone) and 14% (Partner). Both management teams indicated an improvement in the competitive landscape, supporting the thesis ARPU may have finally bottomed (Cellcom’s results for the quarter were distorted as it stopped recognizing revenue from Golan Telecom, whose future was unclear at the time).

Third, it appears that MoC Director General Shlomo Filber is aware that current ARPU does not support further investments in their networks by the big players. In a Bloomberg interview that ataloma linked to in his writeup, Filber argued that mobile prices were too low and that there should be fewer operators in the market.

Investors also seem to (speculatively?) bet that  ARPU has bottomed and will gradually rise from here. Since the Golan-Electra deal was announced around year-end 2016, shares Cellcom and Partner rallied 21%-26% on hopes that the new participant will be much more disciplined in terms of pricing.

Ironically, Bezeq shares actually lost 9% during the same period, as the market grew worry that the Ministry of Finance will interrupt with the removal of structural separation that the MoC is trying to promote (politics... more on that later).

The absurdity of the recent rally in Cellcom and Partner’s shares is that they now trade for ~6.8-7.0x EV/EBITDA (Q3-16 annualized), similar (or even a little higher) to Bezeq - whose other assets (fixed line infrastructure) deserve a much higher multiple and its MNO subsidiary only contributes about 11%-12% of EBITDA.  

The wireless market will likely stay highly competitive for the foreseeable future, but participants’ perception is that we’ve seen the worse in terms of ARPU and intensity of the competition. Also, Bezeq’s ability to sell ‘quad play’ once the structural separation is removed should provide a competitive advantage over Partner and Cellcom in the form of reduced churn rates. (It’s worth noting that HOT is already selling quad packages in Israel, but being a late joiner to the mobile market it’s hard to see them gaining significant market share unless they acquire either Cellcom or Partner - something that could probably happen but only at lower prices).

Multi-Channel Pay TV (YES)

YES offers payTV services through its DTH technology and is the only company in Israel licensed to provide multi-channel pay TV broadcasts via satellite. Again, the only competitor for many years has been HOT, with its cable offering. Exactly two years ago Cellcom launched its own payTV via OTT, and in recent months Partner also announced its intention to launch its own OTT payTV service soon.

Since Cellcom launched its PayTV offering in December 2014 both YES and HOT have started to experience a moderate decline in their subscriber numbers. However, YES has an offering that is based on higher-quality content and premium pricing, and as such the entrance of low-cost OTT providers had a smaller effect on its subscriber numbers. Since CellcomTV launched YES has lost 12k subscribers in aggregate (1.9% of total subs), compared with HOT’s 30k (3.5%).  

That said, the entrance of new participants made price increases more difficult for YES.  Despite improving its offering in recent years, ARPU has been stable at ILS 230-235 for the past three years.

Going forward, we aren’t too worried about further subscriber loss. YES’ positioning as the highest quality, premium-priced player will allow it to compete effectively with the new low-cost OTT offering. Israel’s growing population and the long-awaited removal of the structural separation should provide tailwind and support at least moderate growth in earnings.

ISP and Data Services (Bezeq International)

Bezeq International is the leading provider of ISP services in Israel and the only market participant operating a high-speed submarine optical fiber cable connecting Israel and Europe. Owning the submarine fiber supports Bezeq’s premium offering in terms of connection speed and fixed network quality.

However, this is a relatively small segment for Bezeq, responsible for only ~8% of the company’s EBITDA.

Other Assets

Bezeq owns some additional assets that only have marginal contribution to EBITDA. These include (among other smaller ones) Walla! - Israel’s 2nd largest news portal in terms of traffic, and Bezeq-on-line - a market leading call center services provider.  

Structural Separation

Due to its monopoly status in the fixed line business, a structural separation was forced on Bezeq by regulators. What this meant in practice is that Bezeq had to keep its operating units separated, preventing it from: i) eliminating unnecessary operating expenses, including duplication in managerial positions, and ii) offering “Triple” and “Quad” packages to its customers, realizing what is commonly referred to as “revenue synergies”.

As part of the wholesale freeform in fixed line and broadband infrastructure, the MoC will gradually remove the structural separation that forces Bezeq to spend hundreds of millions in duplicative back office infrastructure and other G&A.

Recently (Dec 21), the MoC granted Bezeq a preliminary authorization for what is considered the first step in the separation - the removal of the corporate structure separation. This was perceived well by the market, as Bezeq shares rallied 7% in the week following the announcement. However, on Jan 2,  the Ministry of Finance announced they object to taking this step before the wholesale reform of the fixed line market is completed as well, and the shares lost most of their gains in the next couple of days.

The mandate to handle this issue is within the MoC, not the MoF. However, the MoF can obviously delay the process if it wishes to, so it’s difficult to accurately outline a precise timeline at this point. Sell-side analysts estimate that at the end of the process, after merging YES, Pelephone and Bezeq International into a single quad-play telecom provider, Bezeq should see cost savings in excess of ILS 600m per annum. While this figure isn’t expected to be fully realized before 2019, it also doesn’t include any revenue synergies resulting from a better offering and reduced churn.

It’s also worth mentioning that Shaul Elovitch is considered close to Prime Minister Benjamin Netanyahu, and Netanyahu has in the past been supportive of Bezeq specifically and of the telecom sector generally.  

Valuation

We don’t like using a specific price target, especially when analyzing a company with so many moving parts. Bezeq is facing some headwind from increased competition for some of its products, but also enjoys competitive advantages in almost all of its business lines, and should particularly enjoy the eventual tailwind from the removal of the structural separation.  We believe that conservative assumptions support flat-to-LSD growth in EBITDA for the next 3-4 years, and as such - our best effort for putting a value on the business will look something like that:    

This implies a 26% upside from current price, and basically constitutes the valuation for Bezeq shares in our Bull Case. While the upside in Bezeq may not be huge, the attractive valuation explains why we prefer to not short Bezeq shares vs. our IGLD holding.   

Part II - NAV Discount

Why the Discount should narrow/close

Let’s jump to the end and explain why we believe IGLD’s NAV discount would be significantly reduced if not completely eliminated by the end of 2019.  In December 2013, the Israeli Knesset passed the “Israel’s Anti-Concentration Law”. The law is dealing with many issues surrounding the lack of competitiveness in several industries, as well as a high-concentration of power/holdings in the hands of a relatively small number of holding groups.

While implications of the new legislation are numerous, our focus here is on the Pyramidal Holding Structure part. According to which, pyramidal holding structures will have no more than two layers of publicly traded companies by the end of December 2019 (this includes private companies with publicly traded debt).

Currently, Eurocom’s holding structure in Bezeq has 3 layers: 1) IGLD; 2) BCOM, and 3) Bezeq. To comply with the new legislation, by 2019 one of the levels need to be collapsed or delist from TASE (equity + debt).

Bezeq is one of the largest publicly traded companies in Israel, with its equity and debt broadly held by basically all pension funds and institutional investors in Israel. Hence there’s practically no chance it will be delisted. BCOM recently issued ILS 1.9b of bonds through TASE at very attractive terms. These bonds will mature only on 2024 and carry a fixed rate of 3.6%, hence it’s very unlikely the group will have the will or resources to buy-back the debt before YE 2019.

This leaves us with IGLD - and there are basically two possible scenarios that we think are viable:

1. Merging IGLD and BCOM through a tax-free stock-swap. This is the ideal solution as we see it, since it requires no additional capital from Eurocom nor refinancing of current debt (the new merged co will assume IGLD’s outstanding debt as of Dec 31, 2019, expected to amount to ILS 446m). If this is the ultimate solution, one layer of NAV discount will be removed as we would expect such transaction will only takes place at a valuation very close to NAV. The only problem with this scenario is that with current holding ownership structure of the group, Eurocom will only hold 40% (61.9% * 64.8%) of the merged co. This is unlikely to happen as Eurocom won’t settle for a controlling stake lower than 50%. So for this scenario to materialize, one or more of the following needs to happen: 1) Eurocom to buy more IGLD shares; 2) IGLD to buy more BCOM shares; 3) BCOM to initiate a buyback program; 4) IGLD to initiate a buyback program (we think 4 is the most likely scenario).

2. Delisting IGLD from TASE, keeping IGLD traded only in NASDAQ. This is possible according to the legislation (weird, right?), but for this to happen IGLD’s bonds must be delisted as well, and that means refinance current bonds through private channels (bank loan, probably). Given IGLD’s current cost of debt is ~2%, a bank loan will almost certainly be much more expensive, and value destructing for shareholders. We think this is a much less attractive option for IGLD’s shareholders, and will do whatever we can to persuade management against it when time comes. Needless to say, if this scenario materializes, there’s a good possibility the discount to NAV in IGLD shares will remain.  

At this point, we rule out a straight-out acquisition of IGLD’s minority interests by Eurocom since we learn from the media that Eurocom does not possess the necessary liquidity (but we might be wrong, and things could change before 2019).  

Some “softer” catalysts that should help narrow NAV discount include:

1. initiation of dividends/buybacks in IGLD in late 2017 or early 2018 - once the retained earnings balance turns positive again (this is required by Israeli law in order to distribute earnings via dividends or buyback). In private conversations, management has indicated the willingness to buy-back shares at current discount to NAV once their allowed to.  

2. Eurocom making open market acquisitions - this may sound strange given Eurocom actually sold some IGLD shares in 2015, reportedly due to liquidity issues. However, Eurocom’s other publicly traded company, satellite company Space Communication (TLV:SCC), has been publicly shopping itself for the past several months. As of November the company reported it was in negotiations to be acquired by a Chinese corporation for approximately ILS 750m. While the deal may face some regulatory obstacles, if completed -Eurocom will receive approximately ILS 470m for its 64.7% stake in the company. We assume at least some of the money will be used to pay down private debt, but don’t rule out a possibility of Eurocom seizing the opportunity to increase its share in IGLD.  

Valuation

Below is our current NAV estimates for IGLD and BCOM:

IGLD is currently trading at ~11% discount to its NAV when using BCOM’s shares, or ~33% discount to NAV when using the value of the derivative Bezeq shares (referred to as “Real” NAV). As mentioned, we believe the structure will collapse to only one layer of holding co. by 2019, bringing down NAV discount to approximately 15% (the average discount BCOM has been trading for in the last 12 months).  

This writeup is long enough as it is, so we won’t go into the amortization schedule of BCOM and IGLD. We do believe that the current stream of dividends from Bezeq should be more than enough to support debt repayments and servicing in the foreseeable future (and it seems that debt holders concur, given the current yields on the traded bonds).

In terms of expected IRR, our base case assume: i) Bezeq’s current share price and dividend yield to remain flat for the next 3 years; ii) flat earnings for the next 3 years, lower than current sell-side consensus estimates, and iii) an eventual merger between IGLD and BCOM, causing the elimination of one layer of NAV discount. This results in a 71% total return for IGLD shareholders by YE 2019, or a 19.7% IRR.  

We would argue that our bull-case isn’t that bullish as it simply a combination of: i) sell-side consensus estimates for Bezeq; ii) a 7.2x EBITDA multiple on 2017 results,  and iii) the merger of IGLD and BCOM (still assuming a 15% discount on BCOM’s NAV). The bear case assumes Bezeq earnings and dividend decline at a 3% annual rate, and Bezeq’s share price falls further 25% from its current levels, reflecting a 10x multiple on 2019 EPS.

The table below presents expected return on investment in Bezeq, BCOM and IGLD shares during the next 3 years under the different scenarios.    

Risks to Thesis

• Structural Separation could be delayed for a long period. Putting your faith in the hands of government officials never proved to be a great idea, and there’s a risk that this process will be postponed again and again. However, the fact that Mr. Elovitch is considered close to Prime Minister Netanyahu somewhat mitigates that risk.

• Eurocom’s liquidation status is unclear. Eurocom is a private company so there’s no available data on its capital structure. According to local media reports, Eurocom faces some major repayments of its bank debt in the next few years. In the past, Eurocom was forced to sell some IGLD shares in big discount to NAV due to liquidity issues, and if the sale of Space Communications doesn’t go through, they may become forced sellers once again.

• IGLD and BCOM merge doesn’t go through.  This probably means Eurocom chooses the 2nd option we highlighted earlier, which is less accretive to IGLD shareholders and makes little sense to us. In that case, our base case IRR is reduced by 400-600 bps (resulting in 14%-16% IRR over the period).

• Sharp deterioration in Bezeq’s operations, hurting its ability to pay dividends. This will put BCOM and IGLD’s solvency at risk again, and IGLD’s equity could eventually be worthless. While this is the low-probability tail risk for the thesis, it is the main reason this isn’t one of our biggest positions.   

Catalyst

• Debt repayments
• Removal of structural separation
• Triggering regulation (2019)