Two weeks ago, the Progress and Freedom Foundation released two reports critical of Philadelphia's plan: Adam Thierer wrote one of these reports called "Risky Business: Philadelphia's Plan for Providing Wi-Fi Service." Based on articles from earlier in the year, the report was initially to be release by the Cato Institute, with which Thierer was associated. It was also drafted before Philadelphia's business plan and RFP was issued. The report was updated and released after Philly's plan--Thierer might have gotten advance copies as some people did--but it still bears many of the marks of ideas that were distributed about Philadelphia's plan that aren't actually in the plan.
When it first came out, I labeled the report more work from sock puppets. But I'm wrong: there's a fair amount of original thought in this. I don't agree with some of the assumptions made, but Thierer does spell out those assumptions which makes it easier to engage in a dialog about whether those assumptions are valid. His colleague Patrick Ross at PFF also pointed out that PFF has a variety of funders some of whom are backers of municipal networks, like Intel, which has taken a generally favorable stance, and may eventually lobby for it.
Fundamentally, I believe I come down on the idea that reducing corporate taxation and providing regulatory relief are arithmetically similar in my political philosophy to increasing the tax burden on individual taxpayers. Perhaps some of you can use that to pin a label on me, but I'm not sure one fits.
Interestingly, many of the new plans for broadband wireless are taking the tax issue into account and are requiring that none of the funding for the networks comes from taxpayer dollars. Offsetting the risk and putting money into the private sector takes some, but certainly not all, of the wind out of the sails of those who want entirely private sector competition.
One of the key issues I agree with Thierer on is that an underreported aspect of many municipal plans--including those that aren't disputed by incumbents or others--is that municipal facilities like light poles, conduits, and buildings are being made available at no cost to the entities that will build the networks, whether private or public. That no-cost basis actually has a price tag attached that prevents competitive entry into the market on equal terms. This is true in Philly's plan and in Minneapolis.
A state representative asked me a few weeks ago: what would keep SBC from coming into a city in Texas that decides to build its own municipal Wi-Fi network (on whatever basis)? I said, nothing. If they're using unlicensed spectrum, no one can really stop them. But the access to facilities for putting in access points is a killer: if SBC has to license for every private rooftop, like cell carriers do, and face opposition or stalling tactics from cities and towns on using poles and other utility facilities, then SBC can't build a network at the same cost or perhaps even one that has the same scope.
I have lengthy comments, and thus provide them after the jump. Download Thierer's report and then follow along. I've quoted small passages and provided page numbers for context.
Page 2:
"Moreover, there are serious fairness issues raised by Philadelphia's entry into the broadband marketplace. First, there is the question of how fair it is to other competitors in the field when governments gets involved in the provision of service. In particular, when government becomes a market participant, it can have a "crowding out" effect on private sector competition, innovation, and investment."
This is often stated, but I haven't seen concrete examples of long-standing networks driving out competition. In Tacoma, Washington, Tacoma Power's Click! network was established partly because of a failure of two incumbents (then TCI and US West) to offer a plan in the mid-1990s for Tacoma's infrastructure. Now, there is robust competition and choice among Comcast, Qwest, and Click! for broadband and between Comcast and Click! for cable. Click! itself sells broadband on a wholesale basis through three private residential ISPs and five private business ISPs. A Comcast spokesperson was quoted two years ago praising the competitive uplift that Click! provided to the city.
The definition of competition is extremely important in this context as well as a potential strawman. Many cities are building broadband when no robust infrastructure exists but incumbents want to preserve the future right to offer that infrastructure at a time and price they determine on a de facto monopoly basis. Competition has to be defined in reference to a body: the consumer or taxpayer, or the company and government. Later, we'll get into the taxpayer side of this equation.
"...if Philadelphia's bet doesn't work out, the costs for local taxpayers and consumers could be enormous. There is also the question of what else could be done with the money Philadelphia is proposing to spend to build its own public high-speed networks. Today's projected expenditures are also unlikely to account for the ongoing technological disruptions we can expect in the future."
Wireless technology has only become enormously cheaper year over year. WiMax is part of the Philly RFP. Philly includes $4.5 million in capital spending for year 2 to 5 to address technology change.
"In sum, in their rush to ensure that local residents are provided with affordable high-speed access, city officials may be doing them a great disservice by crowding out private sectors investments that do not require public bailouts should things go wrong. If private actors are willing to assume the risk, and expend the significant sums it takes to deploy high-speed networks, it makes little sense for local governments to intervene and force citizens to assume those risks instead."
This may have been written before Philadelphia released its plan and Minneapolis announced its approach. In both cases, risk is offset to private lenders and hands-length transactions.
"Before taking such an extreme step, lawmakers can work to reform regulations and tax policies..."
This gets to the heart of many of the anti-municipal broadband bills in state legislatures. Typically, the real issue is removing expensive universal service requirements and adding tax advantages. The viewpoint of this writer and most conservatives is that reducing taxation on companies does not equate to putting more of a burden on taxpayers because the companies plough the additional revenue they have into more investment which can result in reduced costs and improved services as well as a more robust business climate. That's a big assumption, but I don't take a particular stance. I would note that reducing fees that companies pay through regulation and taxation is the other side of taxpayer risk.
Page 3
" Also, means-tested financial assistance can be delivered directly to needy households should city officials feel those citizens cannot afford to pay the going market rate for broadband services. This has the advantage of not locking-in Philadelphia to any specific broadband technology. Instead, residents could use the equivalent of "broadband vouchers" to shop for service on their own from competing providers."
But that results in public and philathropic subsidy of private companies--transferring taxpayer money to private companies to pay for lower-income Internet access. This is precisely what I thought this report was coming out against: using taxpayer dollars for the benefit of a subset of citizens.
Page 4:
"The city would not directly manage this new utility, rather, the nonprofit corporation would oversee all operations and generate its startup funding from foundation grants, bank loans and other sources. At least in theory, therefore, the city would not be required to kick-in any of the initial funding for the project."
It's not in theory: it's stated explicitly in the plan and Mayor Street appears to have bet part of his political career on the city not providing a dollar in direct funds.
"In essence, this means the city would agree to give most (if not all) of its communications business to the non-profit corporation and grant it preferential access to publicly owned infrastructure that private firms would not have."
I agree that this is a significant problem. Minneapolis' plan has the same element although a private company will get this preferential treatment.
This is a method of creating a legal franchise for public infrastructure that's not available to all. It means that the non-profit has to act as the most non-discriminatory wholesale provider and keep up on technology to earn its right to hold that de facto monopoly franchise.
"Is Broadband a Public Good or a Natural Monopoly"
There's a whole set of economic theories here that I won't pretend to be qualified to address, and my interest is primarily in the technological and political implications. So without expressing support or disdain for this section, I leave commentary on it open to those with a richer background in this area than myself.
Page 6:
"* Of the 32.5 million "high-speed lines," 23.5 million provided "advanced services," (services with speeds exceeding 200 kbps in both directions)."
This definition has always been a sop to telcos who wanted to offer 256K DSL and have it deemed broadband. Barely, barely broadband.
"* Subscribers to high-speed services were reported in 94% of the nation's zip codes. In 81% of the nation's zip codes more than one provider reported having subscribers. And 99% of the country's population lives in the 94% of zip codes where a provider reports having at least one high-speed service subscriber."
Specious measure also used in other reports. Zip codes were originally assigned and continue to be added based on population density. They're related to carrier routes. Using Zip codes as a basis of DSL deployment doesn't say much. Texas, for instance, has dozens of counties in which DSL or cable modem service is available in a very tiny area in town centers. In some counties, it's not available at all. Hundreds of square miles might be encompassed by the least populated areas. And there's a post office in the middle of that. I'd like to see this measure dropped. It's like saying that 95 percent of the population lives within two hours of an international airport. What does that tell you about airport location versus where population lives?
Page 7:
" 'the U.S. already teems with wi-fi access,' notes Manhattan Institute telecom scholar Thomas Hazlett."
Hazlett's essay was down on municipally run Wi-Fi networks, but it was also clear that spectrum policy management failure is at the root of dissatisfaction with competition. Hazlett argue that cell carriers, denied enough spectrum to have true 3G, cannot be the robust 2nd or 3rd competitor in markets where that would actually produce real private-sector competition. Hazlett is opposed to government-run broadband, but it's also tricky to see when cell carriers will have enough spectrum to be effective national competitors for broadband.
Page 8
"Various terrestrial and satellite-delivered wireless services have also been proposed in recent years. Wi-fi is clearly the trailblazer on the terrestrial wireless side."
A weak point in this report is that this section should be much longer. One of the most effective counter-arguments to municipal broadband is that it's likely that a combination of licensed and unlicensed private sector networks will appear over the next one to three years that will offer true metropolitan-area point-to-point wireless broadband. There are still issues (see Hazlett comment earlier) about spectrum availability, but there's a lot of innovation. The biggest competitor to my eyes to a Philly-like network isn't wireline, but rather a Sprint/Nextel or Clearwire network or a company like TowerStream ramping up to be a business and residential carrier competitor.
Page 10:
"Even an investment in wi-fi along the lines of what Philadelphia is proposing, is a risky roll of the dice. And this is where the difference between private and public enterprise becomes crucial."
I find this statement fairly tenuous, but not specious. Wi-Fi will likely be replaced at some point in the future by something else. But the next generation of Wi-Fi (802.11n, due to be ratified by 2007, but with robust versions on the market by mid-2006) will extend its life by several years. There's no suggestion in the marketplace that the multi-billion-dollar investment in residential, commercial, and hotspot deployment can turn on a dime. Because Wi-Fi uses unlicensed spectrum and because its cost has dropped so far so fast (with future drops expected as it becomes integrated), Wi-Fi is not a bad bet for the five-year window of the network.
With $4.5 million in year 2 to 5 capital expense budgeted in Philly's plan, this would allow them in year 3 or 4 to replace all of the infrastructure without incurring expense above what they've budgeted. Some of the largest upfront expense is administrative and logical: setting up the network's parameters, billing, etc. And setting up power and other connections for the nodes of the network. If the raw cost of the network is $10 million today, replacing 100 percent of it should be $2 million in three to four years partly because you'll need less equipment to serve larger areas (higher throughput, better reflection absorption, lower costs) and partly because the fundamental setup was done in year 1.
Page 11
"Experience of Other Cities Does Not Bode Well for Philadelphia"
Most of the examples below have been thoroughly debunked. The references are to reports as far back as 1997 before systems were fully built. The Beacon Hill report cited relies on numbers from 2001 and earlier that didn't reflect the state of the systems in March 2004 when that report was released, and which misinterpret a number of publicly available financial details. (Beacon Hill didn't disclose in the report their funding by the New England Cable Telecommunications Association, either.)
The Free Press report from April 2005 is the most extensive; anyone who wants to cite Marietta, Tacoma, etc., needs to read that report first and respond on-point to its reporting:
http://www.freepress.net/docs/mb_telco_lies.pdf
Page 12:
"· Federal and state tax exemptions;
· Exemptions from various other taxes such as local property taxes, excise taxes,
franchising fees, or gross receipt ta xes;
· Tax-exempt debt securities;
· Low-interest government loans or loan guarantees."
Interestingly, the debate on these issues has shifted since the NMRC report was issued earlier this year. Both Philly and Minneapolis's plan would have these problems. Philly would create a non-profit that would be exempt from certain taxes like any other non-profit, but wouldn't have all the municipal tax benefits described.
Minneapolis's plan puts all risk and taxation on the shoulders of the private contractors who will build and operate their network, thus obtaining none of these advantages.
I haven't seen the anti-muni reaction to Minneapolis's plan yet.
"· Free and unfettered access to rights-of-way, easements, lampposts, water towers, and public buildings (including schools and libraries);"
Municipalities that outsource their networks may still be offering this advantage and at no cost to private or non-profit entities that build the network.
"· Exemptions from rate or price controls that private firms must live with."
Most broadband in most states isn't subject to price controls except in those states that require services to be sold on a non-predatory basis.
Page 13:
"In fact, the business plan assumes rapidly declining capital expenditures after Year 1. After the initial $10 million investment in Year 1, the plan posits that capital expenditures will plummet to just $200,000 in Year 2, $100,000 in both Years 3 and 4, and then fall to zero by Year 5."
This is confusing. I queried someone close to the plan who confirmed that the $4 million listed for gross capital reserve is part of the overall capital budget. The $500,000 figure is intended to come from initial funding while the $4 million comes from cash flow.
Page 14:
"Outsourcing the start-up costs will help insulate the city from some of the costs should things turn sour for the venture. Then again, should that occur, it seems unlikely that the city would just let the operation whither away."
This is just speculation: the city isn't just outsourcing, it's removing its control. If the network goes south, I could equally speculate that it would be political suicide for a city hall official to pick up the tab.
"The Wireless Philadelphia blueprint makes this clear when it notes many times that service rates are to be held below the prices charged by private sector competitors, and that any revenues generated by the network will be funneled to other programs of the city's choosing."
The non-profit will be charted with certain expectations, but once it's in operation, the city will have substantially less control of it than this implies. It will be a legally distinct institution.
"What this means is that, should things go wrong, the city (and, therefore, taxpayers) will be forced to pick up the pieces."
This is still speculation. My guess is that this part of the report was written before the city announced its plan, so some segue text was introduced to make the argument that what follows would make sense.
Page 15:
"After all, if the city is guaranteeing wi-fi access to the masses as the equivalent of a birthright entitlement, many citizens might come to expect that other services will be subsidized as well..."
That's a remarkable leap in logic.
"Again, capital expenditures would fall to merely $200,000 in Year 2, $100,000 in Years 3 and 4, and then zero in Year 5."
See earlier: this is incorrect, but a justifiably easy mistake to make given how the business plan was written.
Page 16:
"The blueprint does contain a short technical analysis of potential future wireless networking technologies, such as WiMax, but says very little about what sort of investments might be necessary to ensure those technologies are put into place if they come to supersede current generation wi-fi networks."
Actually, the plan states that WiMax should be considered as part of a proposal. WiMax could likely be used for backhaul and then gradually rolled out for point-to-point use as needed. Pretty simple.
"If things do not go well in the near term and the Wireless Philadelphia's initial $10 million investment (and whatever else they have plowed into it in subsequent years) turns sour, then the potential for a taxpayer bailout enters the picture. Obviously, should this occur, local residents will be left footing the bill for the failed experiment and they will lose financial resources that could be used for other public services. The municipality's bond ratings could also suffer as a result, making it more difficult and expensive to raise funds for other public projects in the future."
Again, we're getting years away and lots of speculation away from what's been proposed. It's not out of the question, of course, but this is the Malthus argument for the network: you know, population always increases indefinitely until we run out of all resources and we all die. Familiar to any debate club member.
"Thus, taxpayers could be stuck with greater tax burdens if private entities leave town and take their tax dollars with them."
Some telcos have threatened to take operations out of towns that plan to build muni networks.
Page 17
"Regardless of such concerns, many cities like Philadelphia will charge forward with wi-fi municipalization plans on the assumption that private firms will not abandon the expensive networks they have already built or planned."
See the earlier discussion on bundling. Because these current wireless plans don't offer the speed, consistency, or low-latency of most wired broadband networks, add-on and bundled services won't be offered. So as long as Comcast et al can offer cable, voice, and broadband on a single bill and promise 5 Mbps or much more in the future, that's not a hard sell against an unbundled service that runs at maybe 1 Mbps. Bundling will keep competitive wireline providers in the market.
"If the municipalization pus h is primarily about the price of service—as many local officials seem to argue it is—then it is likely that local officials will constantly apply pressure to keep the municipal broadband utility's rates lower than whatever the private competitors can offer."
I can agree with that completely. This is one of the risks that has been undercited elsewhere and one of the primary ways in which taxpayer-funded municipal networks could hit a wall.
"In antitrust terms, this referred to as "predatory pricing.""
Oregon apparently has these laws, and there is some concern that Ashland's fiber network has been pricing below cost according to a local newspaper.
Page 19:
"Such concerns partially explain why some state officials have mounted legal challenges to local municipalization efforts."
Now, please. This concern is largely funded by incumbent lobbyist. The fellow who introduced the bill in Texas, Rep. King, said he was unaware of the issue before he was told about and has consistently spoken about the scope of his bill incorrectly--which is why it's had such popular opposition. The original Texas bill would have put out of business all airport Wi-Fi providers, for instance. Further, enacting laws isn't the same as mounting a legal challenge--that would involve the courts, not the legislature.
"These state officials recognize that there are broader, long-term economic costs for the state and the nation should their local municipalities choose to substitute taxpayer-subsidized competition for market competition."
I would love this report to pieces if a reasonable explanation were in this paragraph, too, about the vested interests and the methods by which they have been involved and disguised their methods. The NMRC is one example, but the PFF co-released the other report in this pair with the NMRC, so it's unlikely they would be critical here.
Issue Dynamics, Verizon, and SBC are all deeply, deeply involved in getting these laws introduced. Some incumbents have spoken at length and openly about their super rosa role. I have writtea lot about the sub rosa role. Let's not be naive here and pretend state officials are driving this car.
In West Virginia, legislators wanted to change the bill that they introduced at the behest of incumbents. Here's the quote from a representative involved:
‘"We were moving toward a compromise with the industry," said Helmick, D-Pocahontas. "But they don't want the bill, and they're going to be against it."'
"encourage private investment and competition through:
· regulatory reform to diminish unnatural barriers to entry;
· reform of burdensome local taxes or fees;
· easier access to rights of way / easements / public land or buildings;
· if necessary, direct subsidies / vouchers for the neediest. (importantly, any
subsidies or vouchers should be means-tested."
All of these items transfer taxpayer money to private groups or reduce corporate tax receipts that go into government coffers to pay for services.
I actually like Thierer’s direct-to-consumer broadband voucher idea. This is a much more efficient way to get ubiquitous broadband to those who want but can’t afford it, if that’s Philadelphia’s goal.
The voucher system insulates taxpayers from risk by ensuring that government expenditure only occurs when a citizen actually demands the service. There's no $15 million infrastructure sitting out there underutilized.
But by granting a blanket subsidy to a private firm (or firms) to build a network and then praying for demand, the government grants benefits directly to a select group of vendors without guaranteeing that the public gets any benefit. It's great if you’re the vendor, but not if you're hoping to get more affordable broadband. So, under Philly's model you need a SECOND subsidy, in the form of special pricing, in order to stimulate demand among low-income consumers.
It's an inefficient, circuitous route – the city has to pay for the network’s construction, forgoes collection of right-of-way lease revenues, and then depends on cash flow from more financially stable customers before it can even subsidize computers - let alone broadband service - for low-income households.
Of course, if I'm a network provider, I'd much rather have a guaranteed $15 million contract to build the infrastructure without ever having to worry about subscribers; building the network with my own money and hoping for the direct-to-consumer subsidy to work is too much risk. But that’s not the basis for good public policy.