10 ways to create (and avoid destroying) value in independent towers

Whereas once the tower transaction pipeline in CALA flowed freely, the volume of deals has slowed in the past months as a growing gap has opened up between the valuation expectations of independent developers and those of the strategic investors to whom many had hoped to sell. How do the strategic buyers value independent developer portfolios? How can that value be enhanced? What portfolio characteristics erode value, and how can they be avoided? SBA Communications’ VP International David Porte hosted a roundtable on this topic at the TowerXchange Meetup Americas 2016 – here’s what we learned.

The towerco market can be divided into three tiers:

i) small to medium-sized towerco developers that are seeking to create value over a three to five year period before sale to the next tier of investor;

ii) mid-tier aggregator companies that do some development but their main business plan is to roll up other’s assets; and

iii) large towercos with long term investment horizons sometimes called strategic buyers.

In CALA, the primary prospective strategic buyers in the latter group are the large international towercos; listed entities like SBA Communications and American Tower. Some entrepreneurs hope that Telesites and Telxius may also be prospective strategic acquirers, although neither has yet executed a transaction beyond their parent MNOs’ portfolios.

It is dangerous to assume that if someone simply creates a towerco it will be worth a lot of money. Barriers to entry can be low and there is a lot of money chasing the segment: private equity firms are seeing an ever-growing number of would-be tower entrepreneurs who claim they “know a guy who can get me 500 great search rings.” But not all tower builds and towerco business plans are investible. It’s tempting to think that if a tower is built for US$80,000, it will be possible to quickly increase tower cash flow to US$1,500 and then achieve a successful exit. But in reality things do not operate as simply as that and a growing number of financial investors are finding out the hard way that their projected investment returns may not become reality.

The past frothiness of the tower market, particularly in the Americas, risks causing some investors to either fall asleep at the wheel or make fundamental valuation errors when trying to create value from startup to exit. Too many investors and independent developers are focused on growing their tower count, rather than focusing on what creates high return on invested capital.

Strategics don’t buy towers by tower count, they buy them based off of their individual return. Two towers built for US$80,000 can be worth vastly different exit values even when they produce the same initial cash-flow. Much of the actual value comes from projected future cash flows conditioned on the risks that endanger those projections. If those projected future cash flows are low and/or the risks are medium to high, a company’s assets may end up being worth less than the initial development cost.

The mathematics of evaluating a tower business seem deceptively simple. They are not.

Many investors are duped by the pitch that a company’s exit value is simply equal to Current TCF multiplied by the trading multiple of public tower companies. What investors need to understand is that all cashflows are not created equal; tower companies base their offer on growth in cashflows, not just current cashflows. And investors need to understand that public company tower multiples have little relation to the value of acquisitions since public tower stock multiples are based on a completely different set of assets than those being acquired. Very few deals in emerging markets have ever traded at multiples close to the public towerco values.

What most developers do not understand is that the multiple is an OUTPUT of the model, not an input, which is why every tower transaction trades at a different multiple. When evaluating an acquisition, the price paid is based on the current plus projected cashflows conditioned by the opportunity and risk factors of the market and currency.

Every tower is different and every transaction tends to be different; the comps are seldom really comparable. Some companies will purchase one tower portfolio at 10x TCF and another at 19x TCF depending upon a wide variety of circumstances. Multiple comps must be referenced to the same tower in the same kind of place with the same kind of agreements – a task that is impossible for the investor to perform. Hence, multiples signify the output of an entire process and are meaningless in isolation, when projecting a future valuation.

Here is a quick example that shows just how deceptive a multiple may be.  A brand new tower with one tenant, well built, with solid agreements, ground leases and permits may trade for a multiple higher than a publically traded company tracks.  The same tower three years later will trade for a significantly lower multiple if the owner has already captured the potential growth via adding additional tenants and now there is no growth left for the acquirer. The absolute dollars the seller receives will be higher but the overall multiple will be lower. Further, a mid tier aggregator may purchase a newly built site for a 19x multiple only to see an offer of 14x or less later on because the acquirer’s valuation model is built on projected cashflow – not on some magical multiple that is always paid for tower cashflow.

In essence, when evaluating towerco cashflows, the anchor tenant rate is important if it is too low.

SBA has seen some anchor rates be as aggressive as 30x what it takes to recover the capital to build the site. However you see many developers cut very low-cost deals with operators to get sites and be surprised that at a realistic exit multiple they have covered less than half the initial capital to construct the site.

In some cases, to simply break even, some developers have to add a second tenant to every other site since the sites with one tenant are worth less than what they paid to build them. This might leave the towerco needing to achieve a tenancy ratio of two to satisfy investors’ return expectations – and that may not be realistic in the given timescale and market conditions where the maximum realistic lease-up is only two. This can also lead to cannibalisation of co-location lease rates as some developers become desperate to add any tenants they can find at any price, making it even harder to get any return. This challenging situation is compounded by the pressure that towercos are under to build in lower quality locations where a second tenant is difficult to secure. This means investor capital will need to be more patient and the lower the time-based return is to the investor.

While the anchor tenant rate is obviously a driving factor in TCF and valuation, many other factors are considered when valuing a portfolio.

A number of concessions given by developers to anchor tenants in the recent past dramatically drive the value of the site down when they go to sell it.  Some quick examples:

 - The anchor tenant is allowed almost unlimited space on the tower, 1) making the effective space available on the tower for other tenants less than optimal and 2) zeroing out all potential growth from lease amendments from the anchor tenant. In one case a developer gave an anchor a right of first refusal on all space with a very low cost, effectively giving the anchor veto rights on all new tenants!

- The anchor tenant receiving a share of additional revenues from a second or third tenant driving down the potential future cashflows.

- Having non-first-tier operators as anchor tenants, reducing the long-term expected cashflow security for the site.

- Free RAN sharing allowing the anchor to gain all value from certain co-locations with the towerco receiving no value from their investment.

- Early termination rights that put the cashflows at risk and reduce their value dramatically.

When valuing a portfolio, strategics look closely at the location of a site to determine its potential lease-up.

A tower built in a rural area that must be covered by only one operator may be worth very little. A tower built next to another tower may be worth less than the cost to build. A tower built in an area that has few restrictions on new site builds is worth less because someone else can build next to it in the future.  Once again, the value of the tower is based on its uniqueness and desirability.

For example, in some markets, to build as many towers as they can to satisfy the investors targets for sites built, some developers take rings from operators that are closely adjacent to existing sites. Operators like to do this because they can often get a better rate as an anchor tenant than from an existing site and better terms since the small developer wants to grow.  What actually happens is that the new tower is a money-loser for the investor because the potential lease-up for that site is zero with the tower worth much less than the cost to build.

Towercos need to be aware of market rates and not over-charge to try to make up for what they lost on the anchor rate. Some towercos charge rates as high as US$2,500 over ground.  In many instances the strategics have learned that operators will wait until the portfolio is sold and then demand price concessions from the new owner. In one market it is rumoured that one towerco is having to give away sub-$100 lease rates for co-locations to average-down the overall rate from a previous portfolio purchase. Strategics have learned this lesson and apply a heavy discount for over-market rates.

In general, strategic buyers are reluctant to value over-market lease rates at face value, as they feel they will probably be normalised eventually. If it’s too good to be true, towercos and their investors should expect to be called on it.

Could the towerco mitigate this risk by creating a different brand or subsidiary? This suggestion was met with some scepticism: the wireless industry is too small to “play cute”.

Towercos also need to be aware of exploding swap and other MLA provisions. For example, if the anchor tenant is granted 10% transportability within a towerco’s portfolio, that may not seem problematic within the context of the independent developer’s small portfolio. But if that independent developer wants to sell to a larger strategic towerco like SBA or American Tower, that transportability clause might then apply across a much larger portfolio. A strategic buyer would then have to enter into complicated negotiations post-closing to mitigate the potential damage, but generally demand the seller re-negotiate their leases to remove this provision. In some cases, towers with this provision are stranded with the developer and can never be sold.

Investors and independent developers should always consider the importance of permits.

In some markets it is possible to build and operate sites without permits until someone asks, but in others permits are absolutely essential from the outset.

An investor should never think they will receive close to full value for a tower that does not have all its necessary permits, even if that is a standard practice in the country. Without proper permits, the future cashflows and leaseup capability of a site are severely at risk, once again making the site worth far less than if the company had obtained the permits in due course.

Investors should never believe a developer that says “don’t worry about it – no sites have permits.” Brazil is a fine example. The same operators that built their sites without permits and sold these towers to towercos and now demanding to see the permits of the sites they sold without them. In another instance, an investor was assured by management that the company had “all the permits necessary to start construction” which was technically true since you can always build a site without permits – but you may have to take it down at some point and/or expect a heavy purchase price discount when the site is put up for sale.

Therefore, it is advisable for towercos to spend both time and money permitting all sites before construction, or quickly fix unpermitted sites once built. You need to have a comprehensive checklist of the permits required for each site because your perspective buyer will. A large towerco might spend US$15,000 or more per site obtaining a complete set of permits – but the cost of cleaning up a portfolio can be only one small part of the value destruction. For example, one towerco that didn’t have permits is paying huge per-month fines, while another was forced to take down 100 sites. In addition, in some countries towercos are unable to add additional tenants to their towers without a full set of permits.

Under extreme pressure by their investors to build sites, too many developers are giving away RANsharing rights for no additional fee. TowerXchange is aware of at least one operator only offering build to suit contracts if unlimited RANsharing is granted in perpetuity. A prospective buyer would forecast significantly depressed future lease up potential of a tower built under such terms, resulting in a very low economic value. Investors need to understand the value implications of granting such concessions to operators given that this essentially transfers almost all of the potential economic value to the operator for free.

In some markets operators try to use technical language to obtain RANsharing without the developer noticing. Thus it is critical to clearly define RANsharing, otherwise towercos run the risk that their tenants claim that what they are doing is not really RANsharing. In the end, any allowance for a tenant to use their lease to host another operator, either physically or virtually, takes away from the value of the owner of the site. Developers and their investors need to understand that frequency-hosting is as economically damaging as physical sharing and if they allow it, the site value is dramatically lower.

Good ground leases are the foundation on which the TCF of the tower is secured. Ground leases have to be long-term and renewable solely at the option of the tower owner. Rates need to be within market and not renegotiable. A tower where you cannot secure the long-term rights to the ground is worth far less with some developers seeing their sites be valued at 5x because that is all that is left on the ground lease.

Likewise, agreements that give the landlord a share of future revenues may make securing the ground easier for the developer, but then all future cashflows are discounted in the acquisition model and so the site does not receive full value.

Securing a right of first refusal on the purchase of the underlying property and not allowing for the assignment of credit rights is also important to mitigate against the current or future threat of ground lease aggregators. Without these provisions strategics will pay less.

Developers must include provisions for the assignment of the ground leases without landlord consent - especially if the seller will need to sell their towers in an asset transaction.  If the developer forgot this important provision (and we have seen this more than once) the value of the tower is stuck until the developer can get the leases assigned and has to incur all the costs and risks associated with the ground owner having another bite at the apple.

Another major risk factor facing towercos in many countries is foreign exchange. There have been because a number of failed or aborted transactions because the investor valuation expectations weren’t met because of negative foreign exchange movements at the time of the proposed sale or the devaluation of their portfolio based on future negative foreign exchange movements against the USD. Buyers, sellers and investors should condition their models to foreign exchange risk.

In acquisition models, strategics factor in forward projections of currency movements and so in a country that has good organic growth of 10% with a foreign currency exchange devaluation projected to be 7%, the effective USD growth becomes 3%. If the spreads are even narrower the forecast might even be negative growth.  Developer investors need to realise they are investing as much in the currency as they are in the assets.

On top of FX risks, taxation, regulation, and siting restrictions play a very large part in how much the portfolio is worth. In countries with the inability to increase the depreciable value of the sites through the acquisition or tax towerco revenues heavily, the TCF will always be recalculated to include all those taxable costs.

Market structure is so important that it sometimes baffles the strategics that investors often ignore it! If you invest in a market with one dominant operator and two struggling operators, it is inconceivable that a majority of the sites will ever get all three tenants. A portfolio’s lease-up is made up of an aggregate of sites.  Some get only one tenant, some get two and in some markets with three or four strong operators they may get three.  But averaged out a portfolio may have only 1.5 TPT (tenants per tower) with the three-tenant sites helping the averages a lot.  What is generally seen in markets dominated by a single operator, the maximum expected tenant count may be less than 1.5 TPT limiting the value of the portfolio severely.

The Foreign Corrupt Practices Act (FCPA) matters; this is a binary issue for listed towercos – they won’t buy a portfolio that lacks 100% FCPA compliance.

If there are small deviations from FCPA, it is best to disclose them in advance because they can and will derail a sale process. In countries where there are weak legal and regulatory frameworks there may be a temptation to accelerate permitting and build processes via backhanders – towercos must resist acting in this manner as it can prohibit the divestiture of their entire portfolio and strand the investment.

To date, consolidation of the developer tier sites into the middle aggregation tier of companies has been patchy.

There is growing concern that in a growing number of cases the aggregation buyers are paying substantially more than the strategics are willing to pay. Both the strategics and the aggregators bid on the same portfolios. This calls into question the aggregator’s business plans.  This imbalance suggests that the aggregator’s assumptions are divergent from the strategics and many will have to wait a substantial amount of time to see if their more aggressive assumptions hold true with a high cost of funds. For example, some purchasers are paying 19x earnings when the strategics and past market are only willing to transact at 13x in CALA. This indicates that some buyers are betting on the market rebounding – unless they have a different means of calculating value.

“We have outbid a listed towerco twice,” countered one independent developer, “we felt we had a better view of co-location potential, and we’re happy with the outcome. A mid-tier aggregator towerco can make good returns if they have better information.”

The aggregator is faced with two significant issues that their investors should be wary of:

1) Why does my management team believe they can grow the portfolio faster than the strategics?

2) Why would the strategic pay a significantly higher multiple for the same portfolio three to four years from now if they aren’t willing to pay that multiple now?

In his session, Porte contended that there was a substantial flaw in this argument that goes to the very core of calculating value. A quick example will suffice. If a towerco is willing to buy a portfolio at 13x and an aggregator buys it at 19x, to earn back the differential and aggregator would have to add 46% more TCF than was in the strategic’s acquisition model.  If the strategic’s model contained the assumption that a tower portfolio would grow from 1.3 TPT to 2 TPT in five years, the aggregator would need to have three TPT to break even. This assumption is not believable in markets with three or less possible tenants.

Aggregators that rely on the special sauce that they know more about the market than the strategics who generally operate in the same market, need to be challenged by the investor.  While strategic investors may be more conservative, they generally are well within the band of expected values for growth.  And, as proven above, the amount the strategics have to be wrong by to make up for the valuation difference is too large to play a significant factor in the final transaction value.

But if the action is not in the growth assumptions, but mainly based off of a belief that strategics are willing to pay a higher multiple to an aggregator than to a developer, this is where the model really breaks down.

What drives the business model for the aggregator is really the promise that 1,000 sites should intrinsically be valued more than 100 sites. That site quality, security of cash flows and growth doesn’t really matter. At SBA we caution investors that while there may be a value premium associated with aggregation, when they assess deals they should value the exit at the multiple the strategics bid for the portfolio or at times slightly less if the aggregator paid for the future growth the strategic won’t see. The quality of the underlying asset doesn’t change because of scale.

Since there have been so few aggregators trade, there are not a lot of data points to rely on.  However, in one instance a player has demonstrated that strategics were willing to pay only a small premium for aggregated sites. After aggregating two large portfolios and one small acquisition for a blended cost of slightly above 10x, the company ran a sale process and sold to the winning bidder – a strategic – for slightly more than 12x TCF (tower cash flow).  The important data point is that the strategic bidders for the portfolio were trading at multiples close to 19x but only offered 12x or less.  The bids were based on the quality of the portfolio, not on the trading multiple of the acquirer. Investors should base their exit plans on how good their sites are, not on how public towercos are valued.

There has yet to be any other aggregator of size trade in Latin America, while a number of processes have been delayed or aborted due primarily to the fact the sellers couldn’t meet their target return thresholds.

Since the value created by aggregators over and above the value included in the models the strategics use when bidding on the same portfolio is not an appreciable driver, and the assumption that strategics will pay a significantly higher multiple for aggregated sites over-and-above what they used in the initial sale process is equally flawed, it is surprising that so much money is chasing this segment with no data available to predict a positive outcome.

Because of the past, many investors see the title of “towerco” and throw money at the business plans believing that no one loses money at towers. They fall asleep at the wheel and won’t realise until too late that they may not be investing in towers that are worth anywhere close to what they thought. They pressure teams to deliver site counts rather than quality returning assets worth a lot to a strategic.

Investors in developer and aggregator companies should closely examine and govern the capital allocation of their investments, understand the details and actively participate in the investment because in the end, strategics want to pay good money for good towers.

---

1. The value of a towerco is as much in the contracts as it is in the steel and dirt

2. Valuation comps can be misleading: every portfolio – every tower – is unique

3. A lot of the concessions anchor tenants will ask for can be value destructive, particularly those concerning reserve space on the tower and on the land

4. Towercos must have well-drafted clauses governing RANsharing

5. Buyers, sellers and investors must condition their models against foreign exchange risk

6. Cutting corners on permitting, build and maintenance to improve margins doesn’t increase valuations: every large towerco understands what towers really cost

7. 100% FCPA compliance is a MUST if you wish to sell to a US listed entity

8. The aggregator business model is risky given the gap in valuation assumptions strategics have for the same portfolios

9. All prospective acquisitions are in competition for capital from large towercos – not just with other tower portfolios for sale, but also in competition with towercos’ opportunity to pay down debt or buy back stock, so returns have to be strong to the acquirer

10. The internal rate of return (IRR) on capital invested is important but it’s not the sole factor in defining M&A strategy. Country risk and foreign exchange risk might mean a 15% IRR is palatable in one market versus 23% in another

Buyers will tend to seek out high quality opportunities. They want to buy quality assets at a fair price. A better quality of asset translates into a long stream of cash flow with fewer headaches. Buyers and sellers should appreciate that they are operating in an industry where valuations can be quite baffling.

---