I am recommending a long position in underappreciated Latin American telecom operator NII Holdings.  In my base case I have a price target of $20 on a 2-3 year horizon, with a bull-case scenario of $30 or more.  Additionally, I believe there is downside protection as the company’s assets include tower assets, a wide range of spectrum holdings, and a high-quality, high ARPU subscriber base and potential M&A.  As a bonus, a tower sale will be a short-term catalyst that further crystallizes the value of NIHD’s assets just as the announced sale of NIHD Peru to Entel Chile for $400 million did last month.  The company targets high-value business customers and offers best-of-class customer service and competes in the key markets of Brazil (48% of sales) and Mexico (34% of sales).   With nearly 40% of the float of NIHD shorted, this stock is not for the faint of heart, but provides an excellent risk reward profile and an underappreciated collection of assets. 

Although the stock has rebounded in recent months, it is still down over 80% in the past two years, 35% in the past twelve months. This is due to a variety of reasons, including delayed execution on their 3G network build-out in Brazil and Mexico, strategic and operational missteps, increased competition in Brazil, and questions as to whether the balance sheet was strong enough to make it through a 3G network build that requires extensive cash outlay.  However, the company is now poised to right the ship under the leadership of newly reappointed CEO Steve Shindler who has a solid track record of operational success and solid alignment with shareholders.

NIHD fell all the way down to $4.11 as recently as April 2013, which was significantly below our conservative $8-$13/share estimate of liquidation value (details below) of spectrum and tower assets even if ascribing no value to the company’s 11.5+million subscribers with an ARPU of $35.  This was due to fears that the balance sheet was not strong enough to finance the transition to 3G.  However, balance sheet fears have been assuaged as the company has successfully raised $700 million through the issuance of 7.875% Senior Notes due August 15, 2019.  The company currently has $1.9b in cash and investments along with $5.7b in total debt, resulting in net debt of $3.8b (with no significant maturities until 2016).  However, net debt is currently overstated by $200m due the timing of billings/collection in Brazil as the company transitions to a new system.  The company also has credit lines/vendor financing available totaling around $800 million.  Additionally, the company announced the sale of its NIHD Peru assets to Entel for $400 million which is expected to close in the second half of 2013.  Finally, the company has announced that it expects to sell a portion of its tower assets in mid 2013.  I expect the company to sell 4,500-5,000 of their 9,000 towers for $900m-$1.4b, providing even more cushion from a funding standpoint as the company continues to focus on its 3G rollout in its key markets of Mexico and Brazil. 

The aforementioned injection of liquidity through debt financing and asset sales is more than enough to finance the in-process 3G build out and launch in Mexico and Brazil in the coming year.  The company will likely see negative operational free cash flow of $350 million in 2013, compared to negative $550 million in 2012. However, I expect quarterly operational free cash flow to turn positive in 3Q 2014 as the completed 3G network begins to reach scale as customers are loaded on the network.  NIHD’s network will be the fastest wireless network in Brazil with coverage in all of the major population centers and travel routes.  Capex peaked in 2012 at over $1.5b and is expected to be down to $1b in 2013.

While the stock is highly volatile during this time of transition in the operations of NIHD, I believe there is a solid margin of safety due to the assets that the company holds.  Current spectrum assets include 21 MHz in Argentina, 55 MHz in Brazil, 75 MHz in Chile, 111 MHz in Mexico, and 161 MHz in Peru.  Not including the Peruvian spectrum (which will be sold for $400m along with the Peruvian business later this year), I estimate that these spectrum assets are conservatively worth $3 billion. 

I believe this is very conservative on spectrum asset value, though there have been no private spectrum sales in the region to use as a baseline.   Obviously, all spectrum is not created equal as there are differences in frequency and location.  For simplicity’s sake I have aggregated the spectrum and population numbers.  A $3.4 billion ($3b+$0.4b Peru sale) valuation implies:

1. $0.105 per MHz-pop if you ONLY count the spectrum and pop numbers in Mexico and Brazil (by far the most valuable spectrum) and ascribe no value to the Argentina, Chile, and Peru spectrum.
2. $0.0324 per mHz/pop if you include ALL the spectrum- all types in all places.

For perspective, Entel Peru will be paying $0.146 per MHz-pop at the purchase price of $400m.  This is not an exact comparison because Entel will be buying the entire business in Peru (a business which is losing money and draining cash for NIHD), but a data point worth mentioning nonetheless.  Were NII to sell all of its spectrum at a price of $0.146 per MHz-pop, the value would be $15.3b compared to my spectrum value of $3.4b.

Here are the most recent values paid per MHz-pop in the US.  While US spectrum is more valuable than Latin American spectrum since ARPU in the US is around $65 per sub vs $35 per sub at NIHD, this information provides a frame of reference for my spectrum valuation.

Sales from 2006 and before:

Bidder:                 Price per MHz-pop

SpectrumCo                       $0.45

Cingular                                $0.55

T-Mo                     $0.63

Verizon                                $0.73

Metro                                   $0.96

Sales in 2008 averaged $1.01 per MHz-pop

2011: Spectrum purchases by Verizon averaged $0.68 per MHz-pop.  The cost of this spectrum was lower than 2008 spectrum sales because it involved spectrum in 3^rd tier cities at less attractive frequencies.

The company has tower assets which include around 9,000 towers.  Using comps from previous tower sales in emerging markets, I estimate a value of $200k-$300k per tower, or $1.8-$2.7 billion.  

Including the $400m Peruvian sale, this conservatively values NIHD’s assets at $5.2-$6.1B, or $30.3-$35.55/share.  This compares to net debt of $3.8billion, or $22.15/share, at the end of 1Q13, giving the stock bare bones asset value of $8.15-$13.4 (or $9.30-$14.55 after adjusting for the $200 million overstatement in net debt due to the billings system transition in Brazil).  This assigns no value to the ongoing business operations or best-in-class customer base.

While the value of the company’s assets provides downside protection, it is the potential growth in the business enabled by the new 3G network that provides the engine for expected growth.  I believe the company has the potential to see ebitda go to $2b in the next 5 years from an expected $650m in 2013.  Last week, newly reappointed CEO Steve Shindler reiterated to me that $2b in ebitda is a realistic goal and one that the company will be intent on achieving.  The challenge will be continuing the company’s recent focus on executing operationally in both Mexico and Brazil to insure no further delays in the launch and growth of 3G.  To that end, Shindler has deployed oversight teams to key areas in both countries, established operational benchmarks for the management teams in each country, and brought back in-house the technological deployment of the network.  I expect these changes will result in NIHD having the best network and highest speed 3G data offerings in both Brazil and Mexico.  The opportunity remains huge, with only 26% of users in Brazil currently having 3G services. 

Although the market’s focus on NIHD has recently been on missteps and delays during a rocky transition, a close look at the details reveals a best-in-class subscriber base with significant potential for up-sale in 3G services.  The company has over 11.5 million subs, including 3.9 million in Brazil and 3.9 million in Mexico.  NIHD has the highest value customers in the region by far, with an average service revenue per user (ARPU) of $35 compared to $14 for the closest two competitors.  This ARPU could actually initially rise in local currency terms in both Mexico and Brazil as the current customer base is given the opportunity to add 3G services to their current plans.  As the 3G network is completely deployed, I expect slight ARPU pressure which will be more than compensated for by subscriber additions.   I still expect the company’s best-in-class service level, corporate customer base, higher network speeds, and  premium brand recognition in the region to result in the sustainability of its elevated APRU compared to competitors.

In short, NIHD offers a phenomenal risk-reward profile with an underappreciated collection of assets that provides a margin of safety.  We’re buying a company with improving operations, a focused and aligned management team, and a best-in-class customer base that is also a potential M&A target for below liquidation value of the assets.  Additionally, the likely sale of 4500-5000 of the company’s 9000 towers provides a short-term catalyst which should highlight the value of company’s assets.  While it’s likely to be a wild ride, I expect it to be a wild upward ride.