Description

Sprint is a company burning over $4 billion a year of cash and will be burning close to that amount for the years to come. With $34 billion of debt, it should not have a market cap in excess of $3 billion in spite of its valuable spectrum position. (Think Charter in 2002-2007 or Sears today)

Sprint has been a perennial turnaround story since its ill-fated merger with Nextel and most investors are no stranger to its story. Because the wireless industry is such a well-covered and well-understood business, I will try to add value by focusing on why I believe the Street was wrong on the cashflow projections of Sprint. The disconnect hinges on a misunderstanding of the cashflow pattern of handset leasing.

The challenge investors face with Sprint is that the popular metric for tracking telecom business trend, EBITDA, has become very noisy due to the advent of the EIP and handset leasing models. In investors’ attempts to make convoluted adjustments, they fell prey to the spin deftly performed by Sprint’s management and end up making mistakes that cloud their perception of Sprint’s future cashflow.

There are 2 ways of analyzing the company that we like. One way is to think through the very convoluted accounting adjustments to EBITDA in order to get to “True EBITDA” and then use it to project future free cash flow. It is incredibly easy to miss something and get to the wrong results. But we still do that just so we can track the quarterly performance trend of the underlying business. The other approach gets you to the same answer and is much less error-prone – just track the Free Cash Flow. Not a lot of people adopt this simple approach, partly because the Company has been trying to make people believe that most of the current cash burn is “temporary working capital swing”. It turns out, once you think through all the dynamics of EIP and handset leasing, you would realize that the current run-rate of FCF is the normalized FCF. The only downside of this approach is that quarterly FCF is subject to quarter-to-quarter volatility as is common in any industry, especially in one where sporadic securitization of receivables takes place. There is no reason why the trailing 12 months FCF burn shouldn’t give you a good sense of where the business is at though. The company is currently burning $4.2 billion a year and that is what Sprint will be burning even after all subsidy subscribers are converted to the leasing model. Below is a simplified version of the company's cashflow trend.

The company introduced handset leasing in August 2014 as a way to differentiate from competitors. They then argued that handset leasing heavily consumes working capital and want Wall Street to overlook its “temporary cash burn”. They say that once subscribers were transitioned from the subsidy model to the leasing model, the working capital swing should moderate and the company would then be on its way to positive free cash flow. So why is this wrong? Much has been published by analysts about the ins and outs of subsidy vs. EIP vs. leasing so I’ll only present  my simplified version of it, which aims to illustrate the important point that there is no meaningful difference between the 3 models from a cash flow perspective.

You should be able to see that there is nothing “temporary” about the leasing model working capital buildup. It merely is a movement of handset subsidy expense from the Income Statement to Cashflow Statement. (Handset subsidy expense used to run at a rate of about $6 billion through the income statement but a big chunk has since migrated to the cashflow statement since the advent of EIP and leasing model. For example, the handset subsidy expense for the 12 months ending 12/31/15 is about $2.8 billion. The $3 billion of so handset subsidy expense that migrated to the cashflow statement is exactly what Sprint would like you to believe is temporary.) Sprint does about 17 million of postpaid handset sales in the last 12 months and whether it sells those handsets under the subsidy or EIP or leasing makes no difference to its cashflow trajectory.

In the above illustration I also embedded my view of how the replacement cycle of a subscriber is affected by the choice of model but it should not matter to the thesis even if your views are somewhat different from mine.

The only way the Company can find line of sight toward positive FCF is to cut costs and add subscribers at high contribution margin by so much that it adds over $4 billion to its bottom line. It also has to do that fast enough so that its interest expense increase doesn’t outrun its operational improvement. I don’t believe Sprint can do that. My view is the company may realistically be able to cut at most $1 billion cost without jeopardizing the operation (plenty of Sprint executives have tried for many years), but that will be largely offset by the margin dilution from its latest aggressive 50%-off marketing strategy. It has been replacing high-margin churned subscribers with very low-margin new subscribers and we are about to see the dilution effect. Whether it be MLS, Network LeaseCo or Secured Debt backed by spectrum, the company will have to borrow over $10 billion in the next 3 years at way over its current 6.7% cost of debt to fund its FCF burn, not to mention the $8 billion of maturing debt that has to be rolled at market rate. Sprint also hasn’t found a way to densify its network to deploy its 2.5GHz spectrum without spending more than its $5 billion of capex budget. Many analysts have been led by the Company to assume that Sprint can defy the laws of finance and thus can borrow from the Network LeaseCo and MLS without incurring implicit or explicit interest expense.  I am quite baffled by this level of faith in the magic of financial engineering which I am willing to bet against.

Liquidity

The analyses I’ve seen on Sprint spent too much time on whether the Company can get liquidity through various innovative structures. However, if you came to the conclusion that the $4.2 billion of FCF burn does not have a temporary component to it, whether the Company can scramble for more liquidity makes no difference to its equity value, unless you think that the Company can extend the optionality long enough so that it can enter into a merger agreement with T-Mobile again. Bear in mind that Verizon and AT&T will never be allowed to buy Sprint. And even if a more pro-consolidation administration is elected into office in 2017, there is only a hope but not certainty that a Sprint/T-Mobile merger would be allowed. And Comcast, Google or other strategic entrants would choose to buy T-Mobile instead of Sprint, if only because T-Mobile is cheaper than Sprint (T-Mobile TEV is currently $50 billion while Sprint TEV is currently $47 billion, but growing due to FCF burn). We are long T-Mobile to hedge that risk but I believe the risk of Sprint getting taken over is exceedingly low that the short idea can stand on itself.

Why the opportunity exist?

-        Billionaire hope premium. Masayoshi Son at Softbank has the money and will to make Sprint work. Even billionaire doesn’t have unlimited resources to fund an operation that burns over $4 billion a year, and nobody can defy the laws of finance to get debt funding without incurring funding costs.

-        The cashflow pattern of handset leasing plans are widely misunderstood. Most analysts projected Sprint’s FCF swing from negative $4.2 billion to less than negative $1 billion over the next few years, because they let themselves believe that $3 billion of the “working capital consumption” will normalize. The dynamics and accounting for EIP/leasing/subsidy is complicated, which makes the EBITDA approach prone to error. Nonetheless it can be done right and should get you to the same answer as the FCF approach. Telecom business has always been less demanding on analysts’ accounting skills and I think this Sprint situation stretched the limit of many telecom analysts’ analytical rigor.

Timing

Timing is always important for a short. If Wall Street gets overly exuberant on this hope-based story the stock might trade to $5 but the market hasn’t bid it above $5 ever since the merger with T-Mobile was blocked by DOJ. I like the risk reward at $4 given my view that Sprint has no equity value.

Asset Value

Sprint has spectrum that can be sold for $100 billion in a competitive auction and thus a long case can be made if there weren’t such deep operational issues. But competitive auction on their spectrum is not in the cards, both because of regulatory spectrum cap and the fact that there are 45 million subscribers on networks built on those spectrum. Sprint could sell a slice of its 2.5GHz spectrum for liquidity but that doesn’t get them closer to solving for a $4 billion burn, let alone the fact that the huge swath of 2.5GHz spectrum are worth a lot more whole than sum of its parts. Sprint could sell the entire 2.5GHz spectrum but that is explicitly ruled out for now. In the end, the only carrier that will be allowed to buy the entire portfolio of 2.5GHz would be T-Mobile and one bidder does not a competitive auction made. That’s the pickle that an asset-rich company like Sprint is stuck in.

Catalyst

- Each quarterly number will demonstrate the dire liquidity situation the company is in

- Investors will gain understanding into the ineffectiveness of financial engineering such as MLS and Network LeaseCo