Why spectrum may be a bad investment

Posted in Financials, Globalstar, ICO/DBSD, LightSquared, Operators, Regulatory, Spectrum, TerreStar at 1:32 pm by timfarrar

The President has now announced his support for the proposal set out in the FCC’s National Broadband Plan (NBP) to free up 500MHz of spectrum for wireless broadband over the next decade. However, even though “Our nation’s ability to lead the world in innovation and technology is threatened by the lack of sufficient spectrum for wireless broadband applications and services” according to the FCC, it doesn’t necessarily mean that spectrum will prove to be a good investment over the next few years. (In this analysis, we’re looking at the opportunity for financial investors, as opposed to cellular operators – just like in the housing market, if you intend to use your spectrum, then that’s a different proposition from planning to flip it to someone else).

Indeed it is highly desirable from the FCC’s point of view that spectrum should become cheaper rather than more expensive, because that will enable more competition in the wireless market and result in lower prices for consumers. In the absence of lower prices for spectrum, it is likely that some cellular operators will be shut out entirely from 4G, or will be forced to merge with competitors in order to combine their spectrum holdings – not the outcome that the FCC wants to see. The FCC also doesn’t want to reward speculators – it would much prefer spectrum to be put to use, rather than see financial investors lock it up with a view to profiting from higher prices in the future. Although some might argue that the FCC also needs to raise money from auctions (not least to fund the buildout of public safety networks), it is far from clear that such motivations will weigh heavily in the FCC’s decisions (even if some members of Congress disagree). Certainly the concept of maximizing the proceeds of future auctions was not emphasized in the NBP.

From a historical perspective, despite this apparent crisis, spectrum prices in the US (on a per MHzPOP basis) are actually far lower than a decade ago. This is entirely logical: if it now takes 50MHz+ of spectrum to support $30 of monthly data services, whereas ten years ago operators required only 20MHz of spectrum to provide $50 of monthly voice services, then cellular operators simply can’t produce the same ROI from each MHz of spectrum as they did in the past.

Just the objective of freeing up 500MHz of spectrum (almost doubling the amount currently available for terrestrial cellular service) alone is likely to put a damper on spectrum prices. In recent months, we have also seen the FCC moving rapidly to finalize rules to enable use of 25MHz of WCS spectrum, and formulate policies to ensure that 90MHz of MSS-ATC spectrum is put to use. In addition, the FCC may also decide to limit the amount of additional spectrum that AT&T and Verizon (who accounted for the vast majority of spending during the 700MHz auction in 2008) can acquire in the future.

In my view, all of these developments point to lower spectrum prices in the next few years. In the short term, prices will be depressed further by the glut of spectrum owners seeking to monetize their holdings at the moment: Harbinger, Clearwire, NextWave and other MSS operators, to name just a few. This comes at a time when there is a relative lack of buyers, with most analysts hard pressed to name anyone other than T-Mobile that is an obvious partner for these companies. Investors who acquired undervalued spectrum assets a few years ago (particularly if that was prior to recent rule changes) may be OK, but new investors will need to be more cautious about the price they pay for these assets.

In summary, even if there is considerable long term demand for spectrum, it is a fallacy to equate this with increasing prices. In that regard, spectrum is like oil: you know there will be more demand in the future, but that tells you nothing about how the price will move in the next year or two. The short term price (and indeed the price in auctions) is determined by the balance of demand and supply today. That alone is a negative sign for investors in spectrum assets. However, when the FCC (unlike OPEC) would also prefer to see lower prices for spectrum, then it certainly looks like a risky bet to assume that prices will go higher anytime soon.

1 Comment »

  1. PCSTEL said,

    June 28, 2010 at 5:39 pm

    Spectrum is like oil, that being a commodity.

    Spectrum is a finite commodity, whereas, oil is a commodity whose supply may be ultimately finite, but whose long term reserves are vast and encompassing. Oil substitutes can be created through synthetics. There is no synthetic alternate for spectrum. Spectrum is the ultimate commodity. Because they are not making any more of it. They used to say that about Real Estate, but Palm Island developers in the UAE proved that claim as being incorrect.

    While it may be true that spectrum prices are based on current supply and demand demographics. The ultimate value of a given slice of spectrum is based more on technology, and regulation than supply and demand at the time of auction.

    A couple of case points would be FleetCall/Nextel and the creation of IDEN. Where creation of unique technology provided large financial increases for spectrum assets that at one point carried “zero value” on the original licensees financial statements.

    The second case is the valuation increases of non-paired spectrum between Auction 44 where Aloha Partners paid 4.6 million for the entire Western EAG (Channel 55 unpaired), and Auction 73 where Frontier paid 4.4 million (Channel 56 unpaired) for only Sacramento. It is difficult to argue with those increases in valuation metrics.

    In both cases, there was either Technological or Regulatory Rule changes that sparked the increased demand, hence prices, for the spectral resources over a short period of time.

    Spectrum is only valuable if there is Commercial-off-the-shelf infrastructure and user terminals to support that spectrum. Regionalized “one-off” spectrum offerings (disadvantaged) will be less likely to receive broad user terminal and infrastructure modem support. So spectrum pricing for non-standard “one-off” spectrum will continue to receive poor pricing support due to the lack of integrated modem support by the major chipset providers. Some spectrum like the 2Ghz MSS spectrum is already supported by several chipset manufactures, Globalstar’s 2.49Ghz spectrum also has support from Infrastructure (Alvarion), Beceem chipsets, and multiple CPE equipment manufactures.

    Perhaps the current Commission is a champion of competition as you suggest, but the reality is that we have gone through a period of consolidation over the past 5 years, all blessed by the then/current FCC administrations.

    In my view, spectrum prices will not decrease, except for disadvantaged spectrum that lacks sufficient commercial support. Sure the dogma of claims of 500Mhz of future availability sounds calming, but the reality of the ultimate availability of amounts anywhere near that figure seem doubtful in the near future to this commenter. It took 12 years to clear the Analog Broadcast Spectrum. I don’t see the process getting much easier, or faster. I tend to believe we will see a glut of spectrum seekers, versus a glut of spectrum sellers as you portend.

    Will the realities of demand by existing and new market entrants exceed the availability of spectrum with ready made COTS infrastructure and CPE/user terminal modem support? We are in the “cotton-gin” stage of Wireless Broadband Deployments, not only in this country. But, also worldwide. Time will tell.

    In other news, we see that the MSS Spectrum Flexibility finally got a spot on the FCC Open Meeting Agenda for July 15th. It appears that the FCC intends to issue both a NPRM and a NOI. I would suggest that the NPRM will be for L Band, Big LEO band licensees, and the NOI will be issued for the 2Ghz MSS licensees to stay in step with the original NBP timelines.


Leave a Comment

You must be logged in to post a comment.