With effect from 1 January 2019, new international lease accounting rules will remove the distinction between finance leases and operating leases, requiring conventional tower leases to be brought on to MNO balance sheets. At first glance, this appears to compromise the sale and leaseback of towers as an option for MNOs seeking to delever, but the tower industry believes it has devised a work-around: the Master Service Agreement. Will it work? And will IFRS 16 Leases prompt more MNOs to carve out rather sell and leaseback their towers? TowerXchange spoke to two of the leading investment bankers, who have advised on multiple tower transactions in Europe and worldwide, to explore the implications.
What the tower industry need to know about the accounting standards change
Historically, lease fees paid to towercos were typically classified by lessee MNOs as operating leases, rather than financial leases. This is because the MNO tenant was leasing only part of the tower, not the whole asset. As such, the operational lease enabled tower transactions to be conducted off balance sheet, often by MNOs seeking to use the capital from the sale of their towers to delever.
From 1 January 2019, under the new international accounting standard IFRS 16 Leases, existing long term, non-cancellable leases, including tower leases, must be brought on balance sheet. As a result tower leases may amplify rather than relieve debt, placing more pressure on MNOs’ debt covenants. However, because the lease goes on the books, leasing towers won’t adversely impact EBITDA like it does at present.
Some towercos feel that restructuring Master Lease Agreements as Master Service Agreements would enable their tenants to avoid bringing towerco fees back onto balance sheets.
Introduction
After initial anxiety, towerco M&A and legal experts believe they can mitigate the risk that IFRS 16 Leases could disincentivise tower sale and leasebacks by restructuring Master Lease Agreements as Master Service Agreements. But will that fully relieve the uncertainty, thus leave the motivation for indebted MNOs to monetise their towers uncompromised?
“Tower transactions have never been purely driven by debt reduction. The more fundamental drivers have been the imperative for capital reallocation from non-core passive infrastructure to higher return, core business functions as well as access to a less capital intensive network rollout in an environment of continued ARPU decline,” said Aldo Cardoso, a Vice President in the Global Communications Group at Citi. “Another key driver remains the difference in trading multiples between MNOs, often trading at 4-7x, and towercos, typically trading at 10-21x.”
“The independent towerco business model has deepened network sharing, enabling MNOs to efficiently accelerate the network rollout, modernisation and densification needed to crystalise the growth in mobile data. So a change in accounting standards, wherein no material changes will take place, is unlikely to slow MNOs’ migration to an era and philosophy of shared towers,” added Cardoso.
The independent towerco business model has deepened network sharing, enabling MNOs to efficiently accelerate the network rollout, modernisation and densification needed to cystalise the growth in mobile data. So a change in accounting standards, wherein no material changes will take place, is unlikely to slow MNOs’ migration to an era and philosophy of shared towers – Aldo Cardoso, Citi
With the increasing complexities of managing tower assets as ground rent continues to rise, compounded by the activity of ground lease aggregators, MNOs need to find proven partners to mitigate these risks, and to maximise the efficiencies of infrastructure sharing. Towercos add value to lease management and can optimise the management of operating costs.
“At the crux of tower sale and leasebacks is the arbitrage between the upfront cash and the capitalisation of the leases by the agencies,” said Alexandre Lucas, an Executive Director at Goldman Sachs who heads up towers M&A for EMEA . “Today tower carve outs and sale and leasebacks can be done off the P&L as an operating lease – so you sign a ten year agreement, then apply the present value of leases, and you can capitalise at 6x yet sell for 12x. So it’s a cheap source of financing.”
Will this change have a different effect on the tower transaction pipeline for sale and leasebacks than it will have on the pipeline of MNO towerco carve-outs and IPOs?
“While IFRS 16 may be an impediment to full disposals, it will have much less impact on transactions where the towerco remains consolidated by the MNO – the only difference will be if the telco continues to consolidate the business but is a non-recourse entity,” said Goldman Sachs’ Lucas.
“The pipeline of sale and leasebacks may slow shortly before and after the implementation of IFRS 16 as MNOs assess how the investment community assimilate the change,” said Citi’s Cardoso, “but the effect will be short-lived.”
TowerXchange foresees a finite pipeline of tower sale and leasebacks in Europe after the current Sunrise process in Switzerland. While there is no shortage of motivated buyers (Cellnex, American Tower, Digital Bridge, Tillman, Wireless Infrastructure Group, Global Tower, SBA Communications et cetera), there are currently few similarly motivated sellers in Europe – unless you expand the footprint East where Veon (formerly VimpelCom) is seeking to monetise towers in Eurasia. Even Veon seem to be cooling on the idea of selling and leasing back their Russian towers, although MegaFon and Tele2 could both monetise their towers in the country.
Some MNOs have decided to carve-out and retain control of their towers, rather than sell and leaseback their towers, but the implications of IFRS 16 Leases may have been a secondary driver at the most. Telefónica retained a majority stake in carve-out towerco Telxius primarily to retain control, while securing partial monetisation through a minority stake to KKR. TIM’s philosophy was similar when carving out INWIT: TIM also retained a controlling stake, while partially de-levering through the IPO of 40% of equity. The prevailing view seems to be that Deutsche Telekom may follow a similar path and sell a minority stake in their 30,000+ site towerco Deutsche Funkturm.
Could a strategic investor like a towerco bid for these minority stakes? While there is little motivation for Cellnex or American Tower to deploy capital and share their expertise with a venture in which they were a minority stakeholder and which they could not consolidate, a private investor like Sanjiv Ahuja’s Tillman or Marc Ganzi and Ben Jenkins’ Digital Bridge could consider acquiring 40% of a carve out towerco. However, they may find more compelling investment opportunities where they can attain control.
Is a new formula required for MNOs seeking to partially monetise their towers? Cellnex, American Tower and their peers can be valuable strategic partners – they may be more generous in their valuations than financial investors because they know how to maximise the efficiencies of infrastructure sharing.
One thing is for sure: with competitive processes driving valuations to a healthy premium, and with Europe’s continuing low interest rate bonanza, European towers continues to be a sellers market.
Highly levered MNOs have historically been more inclined to monetise their towers for obvious reasons – if they can generate returns from the disposal of their towers greater than their cost of debt, then selling their towers enables them to partially delever with no material increase in their cost base.
Looking ahead in European towers, large MNOs will still be able to carve out and sell 40% stakes, but it may become more difficult to dispose of 100% during the normalisation phase after IFRS 16. However, MNOs will continue to restructure the ownership of their towers, from carve outs to joint ventures, because the efficiencies that can be secured are substantial.
Break clauses and renewals critical
“With their current methods, the credit rating agencies do an NPV of disclosed leases at the point where the first break clause takes effect. But even if you have a ten year lease, if the anchor tenant has sold and is leasing back the majority of their portfolio, then that tenant effectively has to renew – the lease cannot be broken,” said Alexandre Lucas of Goldman Sachs. “So for example if that initial ten year agreement was extendable by a further ten years, the probability of cancelling would be so low that it should be measured as 20 years and 13x which eliminates the arbitrage. So the sentences in the Master Lease Agreement – or Master Service Agreement – concerning renewal have become critical!”
“With the new accounting standard, operating leases will be more scrutinised – and auditors will take a prudent view on corresponding net debt. In particular they will assess the probability of renewal, and any provision for pricing to be re-negotiated, which would introduce the risk that the valuation arbitrage becomes less,” concluded Lucas.
There may be a useful precedent in anticipating how leases, and specifically break clauses, may have to be disclosed post-IFRS 16. Auditors in the U.S. forced towercos to disclose liabilities where ground leases might have less duration, or a break clause, which could make them shorter than tenancies, for example if they have a tower with two tenants on 15 year leases but a ground lease with a break clause after eight years.
A marginal not a massive change
“If a towerco is using a Master Lease Agreement, lease capitalisation would increase under IFRS 16, so the overall debt burden on MNOs will increase, and the rating arbitrage becomes less attractive,” suggested Goldman Sachs’ Alexandre Lucas.
“However the impact on arbitrage will be less than people think. Imagine you are an MNO which owns a 10,000 tower network with ground leases and other operating and maintenance costs – typically a managed services agreement with a tier one OEM. There is lots of passive infrastructure opex already in your cost base. Then you sell and leaseback your towers. Before the sale you had ground leases only in the operating leases, now the total direct cost is perhaps X% higher with the towerco margin. But the MNO has significant influence on the value of X when they determine the price at which they will agree to sell the towers: if they want to maximise capital released, they have to give the towerco a higher lease fee and more margin, but equally they could drive the opex cost down at the expense of up-front capital. The O&M costs themselves are not significantly changing, and the markup may not be as big as feared relative to the previous cost base of a managed service contract. That will blur the impact and perception of agencies and investors.”
“IFRS 16 will bring about a change in methodology, but if you apply the new accounting standards to an MNO that retains ownership of their towers, you will find significant costs of in-house network management anyway – a sale and leaseback will only make some difference, but not a massive difference,” concluded Lucas.
Will MNOs who wanted to avoid taking a hit to their EBITDA reconsider monetising their towers from 2019?
“Under IFRS 16, the impact of a tower transaction on Status-Quo EBITDA is fairly limited as only the (relatively small) service portion of the lease payment will be accounted above the EBITDA line, so MNOs with leverage headroom may be more likely monetise their towers now,” said Aldo Cardoso of Citi.
Today tower trades reduce headline debt but many investors don’t undertake a sophisticated enough analysis to do a normalisation of the lease capitalisation impact. Rating agencies, on the other hand, will undertake these analyses, but even they use a wide range of multiples: some multiply by three, others by five, others eight – and the level of disclosure isn’t sufficient to normalise EBITDA and net debt.
The focus for valuations in Europe have moved from EBITDA to Equity Free Cash Flow – which is less affected by dividend policy. While EBITDA valuations may be out of fashion in Europe, they remain more important in emerging markets where it’s more difficult to normalise.
Given that these are only accounting standards changes, nothing is materially changing, will banks be inclined to allow MNOs more room in their debt covenants?
“Fundamentally nothing has changed from a cash flow and tax point of view,” concurred Goldman Sachs’ Alexandre Lucas. “It’s an accounting treatment, there is not more interest to pay.”
“Covenants will probably have to be adjusted to reflect the change in accounting treatment, otherwise we are likely to see covenants being breached across EMEA,” added Citi’s Aldo Cardoso. “Telecommunications is only one industry affected, others, like the Airline industry which leases a significant proportion of its fleet, will feel the effects of IFRS 16 even more acutely if no flexibility is allowed in covenants.”
“So many things will be capitalised under IFRS 16 that it will reset how people look at debt,” concluded Lucas, “and I think people will still differentiate between financial net debt (bank loans, bonds et cetera) and debt which comes from capitalising leases.”
Many MNOs may take the view that after the implementation of IFRS 16 in 2019 nothing is materially changing – it is just an accounting treatment, and they’ll state both comparable approaches in their disclosures.
Will this parallel reporting approach convince the ratings agencies? Will banks provide more flexibility in covenants? Have highly-leveraged MNOs that have monetised their towers risked subjecting themselves to greater liabilities than they thought? Or has the tower industry devised a strategy to mitigate this risk: the Master Service Agreement?
Adjusting the structure of lease agreements
Cellnex’s MSA with Wind is effectively a 30 year agreement, so it could have huge impact if capitalised. But it is a Master Service Agreement, not a Master Lease Agreement, and it seems that, to date, Cellnex’s deals with Wind Italy and Bouygues are treated like service agreements – they don’t appear at all in the notes.
TowerXchange spoke to one advisor, who preferred not to be named, who was cautious about the Master Service Agreement solution, saying: “Some towercos have engineered contracts to avoid capitalisation of MNO lease fees. But these agreements have not been tested until the new accounting standard regime is implemented – and the IFRS is clear that if an agreement can be renewed you have to take it into account. This makes the differentiation and identification of the service and lease components critical. While the theory may be that short term leases may not be subject to capitalisation, they could still be capitalised if there is obvious intent to extend.”
“While the accounting standards change applies to most long term agreements, potentially the tower industry could consider not only moving from lease to service agreements, but also shortening the duration of Agreements and adjust the optionality around renewal to avoid treatment under IFRS 16,” suggested Aldo Cardoso. “But of course it’s in the buyer’s best interest to have a long term contract; it reduces risk while increasing certainty of cash flow. And this has been the key selling point to the investment community. However if the MNO is selling the majority of their tower network, they have little option but to renew – perhaps a towerco could accept a shorter term to avoid the effects of the change in accounting treatment if they see that investors truly believe in the lack of MNO optionality at the time of contract renewal.”
Conclusions
“Until the investment community adapts to the new accounting treatment I’d imagine MNOs, and other industries that lease a substantial amount of equipment, will have a window during which they will disclose two levels of leverage: as would have been presented before and after the accounting standards change,” said Aldo Cardoso of Citi. “Rating agencies have already reclassified all operating leases as financial leases – so the credit ratings of MNOs already take tower leases into account and towercos will still be valued higher than MNOs – so that fundamental driver will survive the accounting standards change.”
“So in the short term we may see some MNOs waiting to see how investors react to the change in accounting treatment before monetising their towers, but I expect IFRS 16 to have a neutral effect on tower transactions in the medium to long term as nothing fundamental has changed in the operating model. MNOs still need a more capital efficient way to densify their networks in order to be able to monetise the expected further increase in mobile data traffic” concluded Citi’s Cardoso.
“The difference in the impact of IFRS 16 on MNOs that have sold and who now lease towers, and those who retain their towers, will not be as significant because the long term costs of operating the in-house network also need to be capitalised if the auditors do their jobs properly – otherwise they risk skewing financial decisions based on accounting principles, which would be wrong,” said Alexandre Lucas of Goldman Sachs. “Whether an MNO owns or leases it’s tower network, IFRS 16 is a big change – the magnitude of the change is defined by the margin granted to the towerco or managed service provider. Ultimately, neither MNOs that lease nor those who outsource their towers can cut 20% of their network to reduce costs, quite the opposite.”
I anticipate IFRS 16 creating some upward pressure on tower valuations needed to get a deal done, some downward pressure on lease contract duration, and some pressure on prospective sellers to wait… We could see 5G spectrum auctioned in 2021 so the early twenties will herald massive investment in 5G, including demand for many more towers and small cells – Alexandre Lucas, Goldman Sachs
“I anticipate IFRS 16 creating some upward pressure on tower valuations needed to get a deal done, some downward pressure on lease contract duration, and some pressure on prospective sellers to wait. It may create uncertainty and a slowdown in transactions over the next three to four years as 4G rollouts draw toward a close, and network investment slows,” continued Lucas.
“Overall the towerco model still makes a lot of sense, although it needs to adapt to urban areas networks. We could see 5G spectrum auctioned in 2021 so the early twenties will herald massive investment in 5G, including demand for many more towers and small cells. If I were the CEO of a European MNO, I’d rather wait for the accounting standards change to shake out before I monetise my towers, especially if I didn’t need to do the deal until after 2020,” concluded Goldman Sachs’ Lucas.
By requiring the treatment of tower leases as financial leases rather than operating leases, IFRS 16 Leases will, from January 1 2019, bring the cost of leasing towers onto MNO balance sheet. The majority of towercos are confident that utilising Master Service Agreements, rather than Master Lease Agreements, means their MNO clients can keep tower leases ‘off the books’. One towerco TowerXchange spoken to had been assured by the Big Four accountancy firm who drafted their Master Service Agreement that the contract would not force client MNOs to restructure their balance sheets from 2019. Other stakeholders TowerXchange have spoken to suggest that one cannot be certain how the investment community will respond until the accounting standards change is implemented.
While IFRS 16 Leases could temporarily slow tower sale and leaseback deal flow, it may have little or no effect on the drivers to carve and retain captive towercos, or to monetise minority stakes. And the timing is fortunate as the ‘new normal’ post-IFRS 16 will be better understood by the time MNOs need cash to invest in 5G spectrum and rollout.
Ultimately it should be re-emphasised that this is only an accounting standards change: nothing is materially changing in the operation of lessees’ businesses.