Not All (UK) Infrastructure Assets Are Alike

We tend to like global infrastructure assets. In the best case scenario, they often have captive demand and no regulation, which can lead to superior pricing power – think US towers or datacentres (see most recent report HERE). But infrastructure assets always tend to have captive demand, so even in the case of European tower infrastructure, less attractive contracts still afford investors good earnings visibility with upside optionality from infrastructure expansion (recent note on Inwit highlighting this HERE).

And then even where there is regulatory scrutiny on fixed-infrastructure assets, underpinned returns means infrastructure investors are willing to pay far higher multiples than current stock market investors, as their implied cost of capital is lower. Altice know this too and is aiming to use this to its advantage as it sells a stake in its regulatory-supported monopoly rural fibre projects (note HERE). It is also why Elliott is looking to spin off TI’s network asset which would likely be valued by investors at c.7.5x EV/ EBITDA, vs. the market multiple of 5.3x.

However, the key issue here is “captive demand”. Take that away, and the investment case on infrastructure assets becomes a whole different ball game. Just ask the Indian tower companies. Or going back to the 1980s/ 1990s, ask the original investors in CLECs, or UK cable companies.

This week we undertook a deeper dive into the UK overbuilder market (full note HERE), as this area of infrastructure deployment has been generating a significant degree of external investor interest. Goldman Sachs-backed private equity companies have bought out Cityfibre, Prudential has been investing in Gigaclear, and Soros/ Abu Dhabi have been investing in Hyperoptic. However, as our deep-dive shows, not all of these assets are alike – and in very few of these cases is the demand case clear. These new overbuilders will often be head-to-head with existing big retail brands like BT, TalkTalk, Sky and Virgin. Building a network is one thing. Getting customers to switch, especially in a low churn market like broadband, is another.

We believe that where there is an inherent cost advantage to deploy, ie in urban multi-dwelling units, for businesses like Hyperoptic, they only need to get 10% take-up to cover their costs and low deployment costs will also make the retail opportunity attractive. However, in the wider suburban sprawl of the UK which represents three quarters of the market, the overbuilder take-up needs to be 40% to cover the costs. BT and Virgin are highly unlikely to allow that to happen, so we think the CityFibre/ TalkTalk suburban build model will be too challenging. Hence we see less longer-term risk for BT and Virgin. But BT and Virgin need to be careful. As long as backers continue to fund these models, more infrastructure is put into the ground, and potentially the real winners from this could end up being the second generation of owners, as we saw with cable overbuild from the 1990s when it was rescued from bankruptcy. So, infrastructure can still be a good investment in the end, one just needs to get in at the right point in the cycle. (For all the details on these overbuilder economics, the full note is HERE).

For the full weekly review and updated comp sheets, see HERE.