It is no longer politically acceptable for governments to pick winners with corporate national champions, but interventionists are still inclined to favour some business models over others. Examples include open-source supply in public procurement and extensive forms of network neutrality. This month I focus on the dysfunctionality of the latter--particularly in the context of mobile broadband.
Fast Internet access for all Europeans is predictability among six goals presented by the new Commissioner for the Digital Agenda, Neelie Kroes. These are purported to support the wider European Commission strategy for smart, sustainable and inclusive growth. She's misguided about achieving these objectives with her severe network neutrality requirements.
Only a market in which investors are free to pursue the business models and pricing they choose with clarity and stability in regulatory conditions can adequately and rapidly supply the urgently needed investment. Demands for coordinated efforts at national and European levels are platitudes that duck the fundamentals on commercial incentives, economics and regulatory uncertainties. Broadband stimulus funding can help a bit, but returning the equivalent of just a small fraction of spectrum auction proceeds extracted from the telecoms sector to network operators will have little impact on requirements for hundreds of billions of Euros in new networks.
The European Parliament passed a telecoms reform package last November that favors self-regulation through competition and lays down minimal network neutrality conditions. Kroes has adopted this and says she welcomes consultations, but her threats are worrying. According to the Financial Times, she says she will not tolerate Internet service providers that restrict the speed of commercial websites and would take action if operators such as Telefonica and France Telecom demand payments in exchange for carrying bandwidth-guzzling services such as Google's YouTube. However, these fees could fund the construction of network capacity to support growth and increased performance of these services. Similarly, BBC's low-definition iPlayer service is a wonderful first step toward a world where as much TV programming might someday be consumed on-demand and in HD as it is today via satellite, cable and terrestrial broadcasts. But it is wildly unrealistic and economically destructive to impede that possibility by constraining ways in which network operator revenues can be generated. These are needed to support financing for the massive increases in broadband network reach, speed and capacity required.
A lowest-common-denominator-for-all approach, requiring all services and customers to have identical network access and at the same price with free content carriage in all circumstances, will not make Europe a world leader in broadband Internet or video delivery. Pricing constraints discourage service providers from investing to deliver the higher-performance services some customers crave and for which they will happily pay.
Tiered pricing for different classes of service is proven to be most effective in many industries. Postal services delay lower-priced second class mail service so that additional revenues can be collected from first class customers. Toll roads and business class air travel offer premium pricing for faster, more convenient and higher-quality services. With cafeterias and restaurants or taxis and limousines we select among different suppliers depending on services and prices.
Cable and satellite subscription revenues accrue to both content providers and network operators. These fees are supporting construction of fiber-to-the-home with IPTV in various nations. The same economics apply with broadband telecommunications and Internet infrastructure as with cable TV distribution.
Uncertain regulation is most damaging because it scares off investment. After years of wrangling, a U.S. appeals court ruling has scuppered attempts by the Democratic-controlled Federal Communications Commission to regulate broadband Internet services. Comcast successfully challenged the FCC's ability to dictate how Comcast manages its network including the basis upon which it carriers and throttles peer-to-peer content from service providers such as BitTorrent. What will be the final outcome is unclear given political motivations. Network neutrality was a centerpiece in Obama's presidential election manifesto. Rules might be changed.
It is also unclear how effective Kroes will be with her proposed network neutrality initiatives at the European and member state levels. What are the legal obligations of telecoms operators in terms of carriage and pricing? She should be wary with her previous experience as Competition Commissioner. After four years of fruitless investigations, her antitrust case against Qualcomm and its licensing practices collapsed with lack of legal or economic grounds.
Mobile networks would be even more damaged than fixed networks by severe network neutrality restrictions. Regulation would be particularly unwarranted due to the intense competition with several operators in every nation. Network neutrality conditions could wreak havoc with the business case for mobile broadband. Tiered and segmented pricing already exists with charges of around 10 cents for a 150-byte text message versus a penny or less for a kilobyte of data over GPRS or HSPA, and voice is priced differently altogether including variations according to time of day and place. The mobile Web is embryonic. Its development will arise from experimentation with various new business models, customer propositions and pricing with a wide array of applications and content services.
Extreme network neutrality conditions would harm mobile significantly because wireless capacity is inherently so constrained. Spectrum is a limited resource and relatively costly per bit transported in comparison to fixed connections with potentially limitless copper or fiber capacity. Fast and high-quality mobile access to the Internet can command premium prices that will incentivize investment if only it can be offered reliably. This is impossible with unbridled access to the scarce capacity available at any particular time and place.
And when wireless goes all IP, including voice, network neutrality conditions could be most damaging. LTE is an all-IP architecture that does not support circuit-switched connectivity. With current mobile technologies, voice is carried separately to data on circuit-switched connections. This ensures quality is not degraded by bandwidth-hogging services. It would be a tragedy to mess-up this relatively low-bandwidth and yet highly latency and jitter-sensitive application with the likes of video file sharing downloads.
Service providers need the latitude to offer an increasing variety of service types and pricing for various user profiles. With traffic shaping and policy management, additional network capacity investments can be justified and managed most efficiently.
There is no generally accepted definition of network neutrality. Whereas extreme interpretations would impair market development and customer interests as I have explained, some safeguards are needed to prevent outright anticompetitive discrimination or exclusion. Transparency is essential, as required by the EU telecoms reform package. Class-of-service parameters and quality-of-service achieved should be made clear to customers. Open access safeguards could be required in places where operators are monopolistic or so dominant that wholesale and retail customers and third party service providers have no choice in networks. But mandatory and free content transport, regardless of resource demands on the fixed and wireless networks? No.
Following consultations, national regulators in France, Germany, Italy, Spain and the UK must also decide what their network neutrality requirements will be by mid 2011.