Description

LATAM telecom name Millicom (TIGO or the company) has been written up several times in the past – I would suggest reviewing previous write-ups for a more complete background.  The company is in the middle of a rights offering which was first announced back in November of 2021 when the company announced the acquisition of the remaining 45 percent of the Guatemala business that it does not own.  By any objective measure, the Guatemala deal is highly attractive and makes a great deal of strategic sense.  But, the TIGO rights issuance has come to market at a difficult time, the dilution is far greater than many existing investors anticipated, and several have questioned the size of the offering, especially considering current bombed-out valuation levels.  Many long-standing TIGO investors are sitting on large losses and the fatigue with the name is palpable.  And therein lies the opportunity – even taking a more conservative view of forward free cash flow guidance, the stock trades at 4.2x/10.5x 2023E EBITDA/free cash flow even though leverage is anticipated to fall below 3x by the end of 2023.  Valuation levels look too low and there seems to be a good chance the stock rallies hard once the offering is completed.  

As this write-up is focused on the trade associated with the current rights offering, I will not detail each of TIGO’s markets (See company presentations.  I would note that multiple markets (El Salvador, Honduras, Bolivia among others) have troubling macroeconomic headlines.  That said, the underlying businesses generate consistent revenue streams given the strong leadership positions in these 2-3 player markets.  TIGO’s management team also constantly points out: 

1. Its markets have lower broadband penetration rates, which are expected to rise

2. The demographic base of its markets is better than LATAM as a whole, offering further opportunities for growth

3. Remittances from the US make the underlying currencies in TIGO markets far more stable than other LATAM markets (Including appreciation in Guatemala relative to the dollar over the past decade).

Like Liberty Latin America (LILAK), TIGO suffers something of a complexity discount given the multiple minority interests in several of its markets including: 

-33% Honduras

-20% Panama

-50% Colombia

-45% Guatemala (repurchased November 2022)

These minority interests cause considerable confusion as to the amount of “owned” EBITDA, free cash flow and net debt and this confusion (combined with the confusion about taxes, lease payments vs. interest costs and spectrum costs) is likely a factor behind the lower multiple investors ascribe to a growing business.  For this reason, the acquisition of the 45% of Guatemala meaningfully simplifies TIGO’s story and moves “owned” free cash flow much closer to actual free cash flow.  The deal was done at an attractive price as TIGO paid ~6.2x/8x/11x 2021 EBITDA/EBITDA less Capex/Free Cash flow.     TIGO holds the leading total broadband and mobile market share (64% and 42% respectively) in this two-player market (AMX is the other large player) and the business has grown EBITDA at ~6% CAGR over the past 6 years.  The vast majority of TIGO’s free cash flow will be generated from Guatemala for the next several years.  

The Guatemala acquisition also gave TIGO 100% control of some attractive infrastructure assets – 4,400 (10,000 in total within all of TIGO)  towers, two tier 3 data centers and 21,000 km of fiber.    TIGO has said that towers and data centers are non-core assets, and it will explore strategic alternatives over the next 12-18 months.  With a stock trading at ~4x forward EBITDA basis, there certainly would seem to be some room to monetize the assets.  Following conversations with the company, however, I do believe that some TIGO shareholders may be overestimating the value of these assets.  The ultimate tenant ratio will be considerably lower than other tower markets.  Additionally, TIGO would have to take into consideration that different owners would be incentivized to attract competitors to TIGO’s primary market, and TIGO would also have to ensure that it could secure long-term leases at attractive rates for any disposition.  If TIGO was trading at multiples closer to US cable peers, I think it is less clear the company would monetize these assets.  But with a bombed-out stock price, it could be highly accretive to sell some minority stake to a third party even at multiples 30-50% below the 15-20x levels attained in more attractive tower markets.    

TIGO announced the Guatemala acquisition in late November of 2021 but surprisingly did not actually finalize rights terms until May 18 of this year – why the delay?  According to the company, TIGO needed audited financial results for required rights documentation and therefore the company waited until full year results were fully available.  Additionally, dual listings on NASDAQ and NASDAQ Stockholm caused additional complications/delays.  TIGO was finally ready to come to market in early March…just in time for Russia’s Ukraine invasion.  But, at the advice of its bankers, TIGO delayed moving until May and the stock declined a further ~25% over this time.    Another complication is that, unlike LILAK, TIGO does not have a lead shareholder backer for the rights offering.  JP Morgan (JPM) and a consortium of banks have stepped into this position, but this is a less ideal situation.  While it is more likely that other shareholders will subscribe for the bulk of any oversubscription allocation, the share drop in TIGO’s share price may curtail investor appetite.  Should JPM and other banks be forced to intervene, they would not hold shares long and therefore a forced sale is conceivable (and result in an incredible buying opportunity).  

According to the terms of the right offering, each TIGO shareholder as of May 23, 2022, received 1 right, with 10 rights required to purchase 7 shares.  Assuming the completion of the rights offering, TIGO will increase its share base by a whopping 70% from pre-rights levels – 170 million pro-forma shares.  With rights currently trading at ~$4.54, a rights buyer is effectively purchasing TIGO at~$17.10 ($10.61+.7*4.54), roughly even with the current share price.   Using the 3-year guidance provided at the company’s February analyst day and talking through the numbers with the company, pro-forma numbers would look something like the following:

The above projections are predicated on the company growing operating cash flow by ~10% annually – this assumes a roughly $1 billion capex envelope (this follows heavy network investment over the past several years) and implies roughly mid-single digit EBITDA growth.  It is less likely that the company spends a full $50 million on both working capital and “other” and therefore upside to the $1 billion cumulative 3-year guidance is certainly possible ($800 million-$1 billion is the target – above shows $862 million or $925 million excluding any Colombia/Panama leakage).  It is also possible that the company does not spend a full $150 million annually on spectrum.  Furthermore, Honduras (not consolidated – 0 and $24 million paid in 21/20) dividends (now called “repatriation”) could be higher as the operations would appear to have the capacity to pay closer to ~$75 million annually – TIGO’s free cash flow guidance does not assume any dividends from Honduras.  While many shareholders have argued that the guidance was far too conservative, I suspect there may have been confusion about the true spending on interest and lease payments – several sell-side multiples were not properly capturing the latter number – and many likely were including larger “owned” free cash flow from Honduras .  The company claims it has also noticed overly aggressive free cash flow projections and believes current guidance will help correct certain modeling issues.  It is certainly possible that TIGO grows faster than the above levels.  But, even using the lower end of its free cash flow guidance, TIGO should trade substantially higher from debt paydown and capital return.

When looking at the above numbers, one certainly could argue that TIGO could/should have scaled back the size of the rights offering – or possibly even avoided it completely.  If one simplistically assumes: 

-TIGO upsized the size of its Guatemala bond offering to $1.35 billion and been forced to pay 9.125% versus 5.125% and assumed some small uptick in other borrowing costs, total company interest costs would rise roughly 70 basis points to 6.2%

-No Rights offering but all company cash flow for the next 4 years is used to repay debt

Absent a rights offering, total free cash flow per share would be approximately 33% higher and total net leverage would have been below 3x by the end of 2024 and <2.5x by 2025.  When I pressed TIGO on this point, the company noted that it likely will buy out the 20 percent of Panama it does not own, resulting in nearly $300 million in funding needs (~0.15x leverage impact).  Additionally, it is possible that TIGO will have the opportunity to acquire the ~33 percent minority interest in Honduras which could be ~$400-$500 million drain on cash (~0.2x leverage impact) but again would appear manageable given the steadiness of the business.  Certainly, one can argue that another round of COVID shutdowns could cause results to materially deteriorate.  That said, the 2020-2021 experience probably makes most LATAM countries hesitant to repeat the lockdowns that crippled local economies.  Regardless of the merit, the reality is that TIGO’s shareholder base likely doesn’t support elevated leverage for any material amount of time and TIGO’s management team wanted flexibility for further minority purchases (which would likely be widely accretive).  

While the rights offering may prove to be dilutive on a free cash flow per share basis, technical selling pressure should ease once the offering is completed.  Furthermore, given the investor fatigue in the name, current rights holders may have a legitimate opportunity for a meaningful oversubscription allocation and thus ultimately accrue more value per right purchased.  It is also possible that the ultimate upside could be substantially more than anticipated.  The entire LATAM cable sector trades at bombed- out valuation levels and previous predictions of multiple expansion have been severely misguided.  That said, it doesn’t seem crazy to think that TIGO’s entire operations are worth somewhere near the multiple that TIGO paid for its minority interest in Guatemala.  Using the above projections and a 6x multiple would imply a share price over $50 by 2025 – now this type of multiple would likely require a complete sale.  Given the drop in US cable valuations, it is harder to believe that TIGO trades much beyond 10-12x free cash flow per share (despite potentially faster growth) suggesting values by 2025 of between $32-$39 assuming the lower end of company guidance and $38-$46 assuming $100 million higher cash flow from limited working capital/other spend.  Multiple investors who can consider owning LATAM names will likely be attracted to TIGO’s lower overall debt ratios and therefore I believe that investors will reconsider the name once the rights offering is completed.  Reasonable people can have differing views on the ultimate upside in the name, but the current valuation looks too low, and I believe a material rerating post offering is likely.  A partial monetization of tower assets is also possible.  As noted, however, this is not an easy transaction and would likely be done at a much lower valuation than what many investors envision.  That said, a transaction would accelerate debt paydown and likely allow some capital return.  A third-party investment in TIGO Money is also possible – it is unlikely such an investment will result in any capital return to shareholders, but a third party could help grow the business and a high implied valuation could be supportive for the stock.  

Certainly, there are the typical emerging market risks associated with TIGO.  Higher inflation rates/depreciating currencies/political instability are all part and parcel for this part of the world.  Mauricio does an admirable job of making US investors feel bullish about population/income trends in a volatile part of the world.  While the business has proved remarkably stable despite operating in more volatile countries, regime changes are possible.  Even dictators prefer robust telecom investment but it is possible that telecom rules look very different under a new administration.  More immediately, it is also possible that investor indifference continues following the rights offering – the entire point of the write-up is to highlight the strong possibility that forced selling pressure eases and there is a hard bounce once the rights offering is completed.  While the above valuation levels appear to make little sense relative to other transactions, free cash flow to come, etc., it is possible that emerging market outflows/other redemptions continue and TIGO shares continue to be left for dead.  TIGO will likely be less active buying its stock (vs. say LILAK) in the coming 3 years and therefore returns may be backend weighted.  It is also possible that another round of tax loss selling drives shares down later in 2022.  And as previously noted, there is a scenario whereby other investors do not fully exercise oversubscription rights and the JPM/other banks start dumping stock – this is unlikely but conceivable.  Again, these risks appear well discounted.  For those who can take a dispassionate look at the current setup or were simply never previously involved in the name, the current risk/reward looks compelling. 

Catalyst

-Completion of rights offering

-Higher oversubscription allocation

-Debt paydown

-Tower monetization/TIGO Money investment