Spectrum auctions: trade off and trade up

A winners’ curse is best avoided. Consider the 3G auctions in Germany and the UK where operators paid three and two times annual revenues respectively, betting their analogue house on a digital future. To make good on the auction pledges, plenty of debt had to be raised while financing for network deployment was constrained. Whereas Europe had been a global leader in mobile technology, some maintain that the 3G auctions forever blunted the continent’s edge.


A sellers’ market

High spectrum prices realised at auctions continue to be a sore point amongst bidders, not least when a new generation of technology, currently 5G, imposes hefty investment demands. Add in the more extensive political targets for the sector and potential outlays further increase. 

Auctions, though, can be a non-distortive means by which governments monetise a scarce resource and ensure its economic and social value add is delivered. The onus is on bidders to assess how the downstream market will develop by factoring in the license length, end-user demand, market structure, competitor positioning and other aspects. A calculation is made, where expected revenues are weighed up against the cost of supply. A positive difference is the total amount a bidder may be willing to pay for spectrum.

Going, going, gone wrong?

But spectrum auctions can be, and often are, flawed. In the pursuit of revenue maximisation, the spectrum monopolist risks lower take up, higher prices and reduced connectivity which stunts economic growth and diminishes the associated government tax receipts over the long term.

This happens when the seller exploits its own market power to increase revenues, for example by setting a high reserve price or by hoarding spectrum which restricts supply and, assuming constant demand, hikes the price. Another way to inflate value is by creating a winner and looser scenario, where spectrum is packaged into asymmetric lots so fewer firms can acquire meaningful quantities. 

Italy’s recent mid-band auction is a case in point. Not only was supply of the band’s 400 MHz halved but also the available spectrum was divided into unequal lots (2 x 80 MHz and 2 x 20 MHz) which meant only two of the four active bidders would triumph. Little wonder then that the auction design squeezed out record prices of €0.35 per MHz per population (€4.3 billion in total), some 7 times higher than a comparable Spanish auction, held a few months earlier.

Back seat drivers

While Europe may have a leading 5G vision, the commercial race is being won elsewhere. European policy makers need to decide which 5G variant they want to drive. For a limited version, in which 5G is a quicker and more efficient version of 4G, the path is relatively straight forward and well-trodden. But for an expansive version, where fast and low latency communications capacity is available everywhere and “digitizes” industry verticals not yet much penetrated by intelligent connectivity, policy makers need to charter a new course. If the destination is expansive, such cartography could propel an exponential boon in positive externalities, an asset government should value today.

The combination of 5G and the imminent potential of widespread digital transformation creates an opportunity for a step change in the uses to which connectivity can be put. To deliver the efficiencies and externalities associated with the wider application of the industrial internet, government should adapt auction processes and reconfigure its pricing objective.

Obligations to deploy networks, contained in auctioned licenses, have often been applied at the dawn of a new technology generation. Typically, these have been measured as a percentage of the population covered but they are increasingly becoming more granular, extending to geographic targets including roads, railways and inside buildings. Regulators are also considering obligations for consistent levels of quality and experience, currently measured by minimum speeds.[1]

There is a strong case for governments to stipulate more granular obligations. In doing so, they would accept less revenue upfront but attain a better and more equitable market outcome within a given timeframe. And in signalling their intent, policy makers may catalyse the market to move early.

Deutsche Telekom’s recent bold commitment goes considerably further than the proposed 5G auction requirements.[2] 

Watching the rear-view mirror

Spectrum renewal decisions are another opportunity to re-consider sectoral objectives. Often politicians are dissatisfied with the outputs generated during the license term. A lack of pervasive coverage is often a sore point. For reasons due to topography, demographics and other factors, not least local deployment barriers, some areas are either not covered or left uncovered by one or more operator.

Regulators have taken several ex-post measures to fill these gaps, but no example is quite as definitive as in France. Looking back, the newly elected French government concluded that their predecessor’s approach to spectrum licensing (revenue maximisation) had not delivered the desired outputs (geographic coverage). As a result, e-Government and other digital services were not functioning effectively. 

In a landmark new deal for mobile, the French government waived renewal fess for legally binding commitments to extend network coverage. The accord applies to licences in the 900, 1800 and 2100 MHz bands, expiring between 2021 and 2024, which were extended on condition that an immediate additional deployment programme commence. To provide ubiquitous 4G coverage, every operator will build 5,000 new cell sites, some of which will be shared, at an estimated cost of €3 billion.

This process is functionally equivalent to the government borrowing money from the license holders to achieve its deployment objectives now and repaying the loan via the removal of spectrum charges later. The implicit interest rate of the ‘loan’ might be the licence holder’s cost of capital. Since the loan is given in kind (increased deployment) and repaid in cash (fees abated), the borrowing side must have accurate information on the costs of incremental deployment to secure a good deal. 

The government’s approach established a new set of principles for spectrum management:

  • Coverage trumps revenue maximisation
  • The relevant metric is landmass and indoor coverage, not population percentage
  • Financing is exclusively from the private sector
  • Costly and lengthy deployment barriers will be removed

France would do well to maintain these principles as they consider how to auction off 5G frequencies. 

Hogging the highway

The United States aims to compound its 5G lead and ensure lasting superiority. Uncle Sam’s cunning plan easily fits onto one page. Recognising the externalities gained by being a 4G leader, an annual GDP dividend of some $100 billion, the U.S. is looking at an expansive variant of 5G to drive 3 million new jobs, $275 billion in private investment, and a whopping $500 billion in new economic growth. With their eyes on the prize, the plan has three gears:

  1. Double the amount of spectrum in the market by auctioning an additional 5GHz of mmWave spectrum by end-2019 and looking to make 844MHz of mid-band available too
  2. Remove deployment barriers by speeding up access to utility poles, establishing a 60-day shot clock for small cell approval with a cost recovery approach to fees
  3. Modernise regulation by removing outdated rules that add cost and complexity to incentive network-based innovation

As Europe fires up its 5G engine it should platoon with the juggernauts, but the spectrum fuel has been in short supply. Despite the 2020 deadline to release the three 5G pioneer bands, to date, only four of 28 member states have auctioned the low band, seven the mid-band (although three of the seven have kept some in reserve) and one the high band.

Auction traffic is set to get busy. While auctions have functioned better than beauty contests, the spectrum allocation system they replaced, a revenue maximising objective should be avoided. To trade up to the extensive external benefits that an expansive version of 5G can bring, obligations are an important trade off to consider.

[1] Obligations attached to the 700 MHz assignment in Italy call for operators to cover 99.4% of the population (4.5 years after the spectrum is released) and to cover the principal national roads and railways (3.5 years after spectrum is released) through reciprocal roaming agreements. The planned German auction calls for 98% population coverage with 100 Mbps and 50 Mbps along major transport routes by the end of 2022. 

[2] Deutsche Telekom will deliver 50 Mbit/s to 99 percent of the population by 2020, 100 Mbit/s by the end of 2022 and by 2025, deploy 5G to 99 percent of the population and across 90 percent of the country.