Is Spectrum Sharing a Good Thing?

Is spectrum sharing a good thing? The answer generally depends on who you are, what you do in the Internet ecosystem, how you make your money and what sort of spectrum sharing is contemplated.

The specific answer is even more nuanced, and depends on what assets and demands an Internet service provider faces, how the specific trade-offs of costs and new capabilities and revenue are balanced.

Other broader political considerations also can come into play, as when a change of policy to support spectrum sharing is coming, whether specific entities like it or not.

In India, the immediate context is quite specific. Two mobile operators now are allowed to share spectrum they already are licensed to use, in the same frequency bands.

Such pooling of spectrum is a more limited case of spectrum sharing than other forms of dynamic spectrum allocation being considered around the world, but raises many of the basic business model issues.

Mobile operators have to balance higher spectrum fees (about 0.5 percent for use of shared spectrum) with the ability to generate new revenue from mobile data at 3G or 4G rates where it might not presently be possible.

In such cases, there are avoided capital costs (avoided backhaul and tower investments), higher operating costs (spectrum payments) but also revenue upside to consider.

So is spectrum sharing (pooling, in this case) a good thing, or not? The business case will drive the decision.

If both parties to a spectrum pooling agreement believe they benefit, it is a good thing.

But such pooling therefore is a bad thing from the perspective of other contenders in the same markets who cannot pool spectrum, since it makes stronger competitors of the two operators pooling spectrum.

Other forms of spectrum sharing raise more-profound issues, though. AT&T now holds spectrum licenses  worth more than $91 billion, estimates Goldman Sachs analyst Brett Feldman. He estimates the value of Verizon's spectrum at $79.4 billion.

Some have argued Dish Network’s equity value now substantially represents the value of spectrum licenses Dish owns. In fact, in late August 2015, Dish Network’s equity value was below that figure, at $28 billion.

Perhaps Dish Network’s spectrum is not worth $39 billion. But it might represent most of the current value of the company.

The point is that, traditionally, holders of licensed spectrum have valued it as a scarce asset. What one licensee has, no others may use. That drives up producer value.

In the case of Wi-Fi, the best understood example of license-exempt spectrum, there is no way to capture license value. That works fine for suppliers of devices and services that operate over the license-exempt spectrum, enabling services to be created at lower cost, since spectrum is not paid for.

And to the extent that license-exempt spectrum supports services that are similar to, or identical to, services provided over licensed spectrum, such spectrum arguably reduces the advantage of licensed spectrum.

In other cases, unlicensed spectrum helps service providers using  licensed spectrum. Mobile data offload and mobile device indoor coverage provide clear examples.

In other cases, as where commercial entities are allowed to use spectrum actually licensed to others, the acquired spectrum might have characteristics of licensed spectrum, especially when interference protection mechanisms keep others from using such capacity.

The biggest question likely is whether spectrum has to be paid for, or can be used on a license-exempt basis.

Generally speaking, device and app providers benefit from such access. Access providers gain some benefits, but also face devaluation of their licensed spectrum portfolios.