The purchase thesis
Uniti Group (NASDAQ: UNIT) is the 8th largest fiber optic provider in the United States with 133,000 miles of route and 7.8 million miles of fiber strands. I think UNIT is undervalued both from an asset and cash flow perspective.
Assets are valuable for 2 reasons:
- Demand is increasing with the rollout of 5G and data usage continues to rise.
- The cost of replacement has become insurmountable
UNIT is valued at only a fraction of replacement cost and an abnormally high AFFO yield that gives the stock an edge through dividends, multiple expansion and/or a potential major asset or sale of the company.
Let’s start with the cash flow side of the analysis because the value of the asset is only relevant if the asset can generate income.
Revenue generation and growth prospects
Revenue metrics are guided as follows:
- 2022 revenue of $1.131 billion
- 2022 adjusted EBITDA of $896 million
- $450 million AFFO 2022
- AFFO 2022/share of $1.74
Each of these metrics is up moderately from 2021.
Against a stock price of $9.83, $1.74 of AFFO represents an AFFO return of 17.7%.
The average REIT has an AFFO yield of 4.23%, so as a runrate, the capital invested in UNIT generates about 4 times more cash flow.
If UNIT were to continue winning at this rate, it would obviously be undervalued. Obviously, the market thinks earnings will drop significantly and the obvious reason is the ongoing battle with Windstream.
Windstream’s contract revenue will continue to grow over time, but the lease will eventually be renegotiated with arbitration potentially scheduled for 2027.
It’s pretty uncertain what the end result will be. Uniti says rental income is fair while Windstream says it needs to come down significantly.
The basis of Windstream’s argument is that the copper portion of the network has depreciated significantly and that’s true, but the fiber portion of the network has become extremely valuable. Even though Uniti owns the assets, Windstream has spent $1.1 billion to build the fiber network, including $85 million already in 2022.
Investing $1.1 billion in your own network requires a certain return on investment, but investing $1.1 billion in someone else’s network requires a much higher rate of return.
So I don’t really buy Windstream’s argument that the network isn’t hugely valuable to them. If the network wasn’t generating serious revenue for Windstream, there’d be no way they were investing $1.1 billion in assets they can’t even keep.
Looking at the base cost of the network and the replacement cost of the network, I lean towards the current rental levels being fair. However, there is no doubt that it will get messy and bad public relations for everyone involved. I also do not deny the potential risk of a haircut around 30%.
It’s still a long way off and in the meantime, Uniti is growing its non-wind revenues quite well.
The growth comes from 2 sources:
- Rental of existing fiber
- The new made more economical by the existing fiber
Beginning in 2Q21, fiber rental volumes accelerated at a rate of approximately $0.9 million MRR (monthly recurring revenue) booked each quarter.
Since this is monthly revenue, this represents $10.8 million in annual revenue. 59% of this additional income comes from rentals, which represents a margin of almost 100% and the remaining 41% represents a high margin but not 100%. Of that $10.8 million in new revenue, about $9 million is passed through to AFFO, which has split over 267 million shares outstanding, implying about $0.033 increment each quarter or about 0, $13 per year.
Most non-wholesaler bookings consisted of corporate and electronic fares. Demand from these tenants is well oriented.
As 5G rolls out, I expect UNIT rental growth to maintain or increase in pace. $0.13 of AFFO/additional share represents significant growth for a stock below $10. Using a 10% capping rate on new revenue, this represents $1.30 of annual value growth.
So, while the Windstream mess will constantly weigh on UNIT’s share price, I want to point out that even in the worst-case scenario, UNIT still owns all of the fiber assets, so that growth through the pipeline rental remains intact.
In other words, it is only Windstream’s revenue that could potentially be discounted. The fiber sector is unrelated.
Value of fiber assets
Fiber is a bit difficult to assess as there are many route specific factors as well as add-ons such as nodes or network connections.
As such, it’s hard to get a perfect comparison, but there are some decent comparisons available which I’ve featured below.
The $14.3 billion purchase of Zayo by DigitalBridge (DBRG) is probably the cleanest example of a public takeover of a pure fiber company.
Zayo has 15 million strand miles and 130,000 road miles, which makes the purchase price equal to $953 per strand mile and $110,000 per road mile. Uniti trades at an EV of $8.439 billion or approximately $1081 per strand mile and $63,435 per road mile.
Thus, Uniti is significantly discounted compared to the Zayo transaction on an EV per road mile basis, but expensive on an EV per strand mile basis.
What is the difference?
Zayo’s fiber is long distance roads with dense urban areas, so it has large ducts with an average strand density of 115 fibers. It covers less area but has more capacity in the areas it covers.
Uniti’s fiber is a bit more urban Tier 2, so it doesn’t need as many strands in each conduit to handle capacity. Its average density is 59 strands per conduit. Suburban/rural fiber like that owned by Uniti tends to generate higher revenue per mile of strand, but lower revenue per mile of route.
Overall I would say Uniti is a little cheap compared to the ZAYO deal. Uniti and ZAYO are offered at extremely low prices compared to replacement cost.
Today, burying new fiber ducts is extremely expensive and time consuming. Perhaps the best measure of replacement cost is the capital invested by Crown Castle (CCI) to lay new fibres.
- In 1Q22, CCI spent $2.6 billion to build 13,000 miles of road
- In 2Q22, CCI spent $2.8 billion to build 13,900 miles of road
This represents a cost per road mile of $200,000 and $201,000, respectively.
Obviously there are location differences that would impact the replacement cost, but at $63,000 per road mile I think it’s clear that UNIT is trading well below the replacement cost of its fiber network. It is also trading below the cost of its own transactions in which UNIT bought Bluebird and TPX at $1,792 per mile of strand and $146,000 per mile of road, respectively.
Additional assets not included
The fiber value analysis above assumes that the rest of Uniti’s assets are worth $0.
Copper is no longer in vogue, but it is still used and generates revenue. As such, I believe there is material value in Uniti’s 230,000 miles of copper route. Even if the copper itself becomes obsolete, the conduits in place could reduce the cost of overbuilding this network with fiber.
Purchase Thesis Summary
Uniti is a messy story that drives investors away and the market price is low. However, the assets are in high demand and difficult to replace, which makes them valuable. I estimate that UNIT is trading at just under half NAV.
Leasing this fiber network paves the way for realizing the value of the asset. At the current rental rate, UNIT can add an additional $0.13 of AFFO each year, which, at a conservative multiple of 10X, represents $1.30 of incremental value. Note that 10X is less than the market multiple for pure-play fiber AFFO.
That $1.30 a year more than justifies Uniti’s sub-$10 market price, so the stock is undervalued even if the Windstream debacle ends up being bad news.
While waiting for the fiber to be leased, shareholders can take advantage of the dividend yield of 5.88%.
Catalyst: increase in the dividend
Uniti’s dividend was limited by covenants, but improved financial conditions removed these restrictions. To date, the payment represents only 90% of taxable income and 34% of AFFO. In order to maximize its tax advantages as a REIT, UNIT will have to increase its dividend.
Such dividend increases on already high-yielding stocks tend to increase investor interest.