The Big Question  

Should Telcos Invest in 21st Century infrastructure or focus on deploying new services across existing plant?Though many are still generating plenty of cash, boardrooms are wrestling with the question of just where they should invest it to maximize returns in the future – and coming up with very different answers. The digitization of everything, and everything over IP, have made this a very tough decision indeed – not just for telcos, but for regulators, vendors, and everyone else that makes up our industry.

In a way, investing in infrastructure is the easier path to take. It's what telcos have always done, and generally done pretty well. And there are plenty of ways to do so through the next five years: Deep fiber and FTTP, true mobile broadband, and all-IP NGNs all provide more than enough rationale to spend, spend, and spend again well into the next decade. And telcos appear ready to do so, (with the exception of ATT/Bellsouth who seems focused on layoffs and mergers rather than deployment and customer service).

Yet there's no escaping the fact that investing billions in new infrastructure tends to require that you take some kind of position on the services that you expect to run on it – and on the return those services will earn. If telcos splurge on fancy new hardware and software, and consumers then use it to effectively cut their spending rather than increase it, investors are liable to cast a pretty cold eye on further investment. And the fact is that some infrastructure spending clearly does imply lower spending by end users – for example, infrastructure that makes it easier to run VOIP. What to do? One view that has gathered support over the past few years says that infrastructure owners should stop being service providers altogether, and focus on creating the best possible platforms for third parties to innovate around. That is the lesson of the Internet, advocates say, and it could be extended right through the telco value chain.

This approach is exemplified in the so-called "open access" networks that have sprung up across northern Europe over the past few years. The best current example is in Vasteras, a small city about 100km west of the Swedish capital, Stockholm. The public utility in Vasteras, Mälarenergi Stadsnät AB , has invested almost $40 million in the construction of a city-wide FTTP/Ethernet/IP infrastructure, yet it provides no end-user services at all on the network. Instead, it sells access to the network to other service providers. The results are impressive: About one third of all households in Vasteras are now buying services on the network, choosing from more than 100 services offered by more than 20 service providers, including the incumbent telco TeliaSonera AB (Nasdaq: TLSN ). As a result, the town's fortunate citizens can – among other things – get 10-Mbit/s symmetric Internet access for as little as SEK125 per month ($19). And MalarEnergi itself is turning a profit. That's a compelling argument for open access, but it's not the only way to justify infrastructure investment. Another is to bully and bewilder regulators to the point where they allow you to re-monopolize your network.

The investment dilemma isn't so big if you don't have much competition on the services side – though you still, of course, need to persuade customers to buy. U.S. regulation of Verizon Communications Inc. (NYSE: VZ )'s FiOS network looks like a good example of this approach. That may well be good for Verizon, though whether it will also prove to be good for end users remains to be seen. A third approach, pitched between these two extremes, is to invest both in infrastructure and in software platforms that make it easier for third parties to create compelling new services that will run on the infrastructure you've created. That seems to be the approach BT Group plc (NYSE: BT - London: BTA) is taking with its Web 21C program, which allows third parties to create new services using the resources built into BT's 21CN infrastructure. Many of the proposed investments in QOS tools, policy tools, IMS and next-generation OSS, as well as in Web and Web 2.0 software, are built around this view of the future. Telcos don't provide all the services, but they do provide a lot of the value that underlies those services. In BT's case, though, it doesn't involve much fiber yet.

Mind Your Back
The truth is, there is no easy answer to the question at the head of this column. Mao Tse Tung, asked in the 1950s what the consequences of the French Revolution had been, famously replied that it was "too soon to tell." Likewise, the impact of ubiquitous digitalization and IP are only just beginning to play out, and it's much too soon to predict their ultimate effects. In that environment, it's vital for telcos and their suppliers to keep reviewing the options, and to take the widest possible view of market and technology developments. The investment dilemma isn't going away any time soon, but we at least hope to make it a little easier by keeping the industry abreast of all of the emerging developments that really matter – whether in infrastructure, in services, or in the technology that lies between the two.