Interest Rates And Auction Policy –Why The FCC Should Move Quickly On A 5G Auction.

It is a measure of how much communications policy warps my brain that my thoughts about the rise in the Consumer Price Index (CPI) and the likelihood that the Federal Reserve will raise interest rates aggressively as a result have little to do with the impact on stocks, or even my credit card debt, but go directly to the impact on any future spectrum auction. Short version — nothing good. So if we needed another reason for the FCC to move quickly to schedule the next 5G Auction, the potential rise in interest rates is a good one.


I explain this in more detail below . . .

I am not a huge fan of auctions. They are, to borrow a familiar cliche, the single worst way to distribute exclusive spectrum rights, except for every other way ever tried. So as long as we are going to have exclusive spectrum rights, we are going to have auctions. And a long as we are going to have auctions, we ought to try to do them in ways that maximize the likelihood of competitors winning licenses. Further, while I don’t think we ought to try to maximize revenue to the government, I recognize a lot of people (particularly in Congress) do.


So with all those caveats out of the way, lets look at the current auction situation. The big news is, of course, “5G.” Setting aside that no one really knows what 5G is or will be yet, communications policy has become obsessed with making sure 5G happens. As always when we are being pushed to do something by folks with an interest in seeing something happen, everyone in Policyland has heard endlessly how (a) 5G will be the bestest most wonderfullest thing ever and will make our current 4G wireless technology look like stone knives and bear skins; and, (b) if we don’t get moving right away China or Europe or someone will eat our lunch, we will no longer lead the world, USA! USA! But even if we discount the hype, it’s pretty clear that we have an industry looking to do cool stuff with spectrum previously thought unusable. So whatever 5G is, we probably want to get some spectrum out there for it (including unlicensed spectrum for next-generation WiFi, LAA and other cool things).


We Need An Auction To Promote Competition.


FCC Commissioner Jessica Rosenworcel — who is probably the most spectrum-wonkish FCC Commissioner to have ever sat in an 8th Floor Office — has been pushing hard for the FCC to schedule a 5G auction. I agree, but for somewhat different reasons. For me, the compelling reason to try to get spectrum already approved for 5G auction set up and scheduled as quickly as possible is because the current spectrum allocated for 5G is primarily held by Verizon and AT&T. Several bands designated for 5G spectrum have lots of legacy licensees — folks who bought the licenses cheap years ago for a bunch of uses involving much older technology. So once the FCC designated these bands for 5G back in the first Notice of Proposed Rulemaking back in 2015 (with the actual Report and Order in 2016), the race was on to buy out the legacy licensees. Unsurprisingly, Verizon and AT&T bought up most of the available licenses (with a smidge acquired by T-Mo). As a consequence, Verizon and AT&T have a significant first mover advantage in both developing and deploying new technology.


To be clear, I am all for Verizon and AT&T competing to develop new cool stuff to sell to me and other consumers. But more competition would make it even more fun. After all, if I want a monopoly (or even duopoly) network provider I can always have the National Security Council build the network.  But the only way to get competition is to make more spectrum available before AT&T and Verizon cement their first mover advantage, and the only way to get substantial exclusive licensed spectrum for 5G spectrum available is by having an auction. So if we want competition in the development of 5G, rather than letting the two largest companies have a couple of years head start, we need an auction as soon as reasonable.


So What’s the Hold Up? I Thought Republicans Loved Auctions?


They do. The problem is that Chairman Ajit Pai has decided that the current auction statute has a major problem that prevents the FCC from moving forward. 47 U.S.C. 309(j)(8)(C) requires that when the FCC collects money in advance of the auction, the FCC must put that money “in an interest bearing account.” (The FCC requires that companies participating in the auction may a deposit, the size of which determines how much you can bid in the auction. This prevents entities that don’t have money from just running up the bidding.) Apparently, there is some problem with the government getting an interest bearing account. I have no idea what the actual problem is, but Pai is convinced that the FCC cannot run an auction unless the law is tweaked to remove this requirement.


Yes, such a legislative fix is easy, and is included in the FCC Reauthorization Bill the House E&C Committee just passed. And sure, it seems like the kind of stupid thing that could be included in the next continuing resolution or budget bill or whatever Congress will need to pass in the next few weeks to keep the government funded. But it hasn’t happened yet. And until it does, Pai won’t schedule an auction. Indeed, as far as I can tell, the FCC hasn’t even started working on the auction rules — which they could certainly do even without scheduling an auction.


Which is why I am bringing up the issue of interest rates as an additional reason for Congress to get it’s rear in gear and just pass some quick technical fix so we can get an auction scheduled. Additionally, while we wait for that to happen, the FCC could at least put out a public notice to start the process of fixing the auction rules.


How Do Interest Rates Impact Spectrum Auctions?


Not all spectrum auctions generate massive revenue, but some can get pretty pricey. Given the huge interest in getting 5G licensed spectrum, we can expect the auction to run up a couple of billion dollars. Most companies, particularly smaller companies, finance participation in spectrum auctions by borrowing the money. It is further complicated by the fact that companies may borrow a chunk of change up front to make the advance deposit, and secure a line of credit for even more money in the event they win. Because securing a line of credit costs the lending institution the opportunity cost of holding the money in the line of credit in reserve, the cost of securing the line of credit is also related to the interest rate.


When interest rates are low and money is cheap, companies find it a lot easier to secure lines of credit and can afford to bid more on spectrum. As interest rates go up, it becomes both more expensive to borrow money and makes bidders more reluctant to have debt on their balance sheets. This disproportionately hits smaller carriers and competing carriers, since they are seen as being less able to extract value from the licenses than the larger, dominant providers. In plain English, lending money to Verizon or AT&T for a spectrum auction is considered a safer bet than lending money to someone else for a spectrum auction, since Verizon and AT&T are the biggest carriers, have the most customers over which to spread the cost of repaying the loan, and are most likely to get economies of scale to lower cost. By contrast, other players are a riskier bet, and therefore get charged higher interest.


There is a further complication here because while spectrum is a safe bet as an asset in the long run, it is likely to be unclear for awhile just how 5G will work and what sort of revenue companies can expect. That makes it harder to assess short term risk. For commercial lenders, and potential bidders, this is extremely important. Banks don’t want to hold licenses, so they are not going to foreclose on spectrum licenses to sell them later like banks do with real estate. (It is also extremely hard to get licenses in bankruptcy, but I will not go down that rabbit hole). So while 5G is a good bet in the long run, it is a somewhat riskier bet than more recent spectrum auctions in terms of getting paid off quickly.


For the last 10 years or so, interest rates were fairly stable. You can see the history of the US prime rates here. But analysts are now expecting a fairly rapid rise in interest rates. That means delay potentially has significant impact on the cost of capital for spectrum auctions. It is already going to take us until end of 2018 to schedule an auction (it takes awhile to set the rules, then you have to give potential bidders time to figure out if they want to participate and raise the capital to do it). In that time, we could see a rise of a full point on the prime rate, according to some projections. Given inflation fears, many analysts are now projecting rate hikes to continue into 2019. I will leave it to the real financial analysts to figure out the likely cost of delay, but when we are talking borrowing billions of dollars, an increase of a full point is expensive.


The more expensive it is to participate in the auction, the fewer competitors can afford to compete in the auction and the less money they have available to bid. So whether you care about promoting competition or enhancing auction revenue (or both), delay in a time of rising interest rates is major bad news.




You don’t have to believe the hype about 5G to want an auction sooner rather than later. As I said above, I don’t particularly care about enhancing revenue, but I would rather see the money go to the Treasury in higher bids than go to lenders in higher interest payments. The fix on this is fairly simple, and the FCC could at least start the ball rolling so we don’t waste time once Congress passes a fix.


Stay tuned . . .

This entry was posted in Spectrum, Tales of the Sausage Factory and tagged , , , , , , . Bookmark the permalink. Both comments and trackbacks are currently closed.

One Comment

  1. Dave Dayanan says:

    Whoever wins the bid will be taking the pros and cons of the 5g. I think AT&T might have a good chance though.

  • Connect With Us

    Follow Wetmachine on Twitter!


If you do not have an account: Register