Recently in Networks Category

Michael Giberson

Tom Weber said that neither word in the phrase "congestion pricing" is too upbeat, and strung together "the combination evokes thoughts of opening one's wallet while suffering a sinus headache." He suggests that the unappealing phrase may have had something to do with the failure of Mayor Michael Bloomberg's congestion pricing plan for New York City.

Zubin Jelveh, at Odd Numbers, collects some alternatives from branding experts and others, including "EZ-Zone," "FreeFlow," and "StreetSmart." But Jelveh notes that most congestion pricing plans in place around the world don't seem to need an appealing brand name to succeed: in London the phrase used is "Congestion Charge," and in Singapore their program is "Electronic Road Pricing."

On the other hand, Jelveh notes that Milan has styled their plan as "EcoPass."

Beginning in January this year, to enter the central district of Milan with an automobile required purchase of a pass, with the fee tied to the emissions level of the car. One early report suggests the plan is working to improve air quality in the city. (Of course, improving air quality is easy -- economists would want to know how much it is costing to achieve the improvements.) I was unable to locate a more current report in English.

RELATED: In 2006, Lynne posted on Road Congestion Pricing in Stockholm. Last October, I argued that, "Economists do not understand the opposition to congestion pricing."

Michael Giberson

If you, like me, were looking forward to the FCC's auction of prime 700 MHz wireless spectrum -- and frankly who wouldn't be interested in seeing how the FCC's new hierarchical package bidding rules worked out in the C-block auction -- you may have noticed the distinct lack of news on the topic in the major press.

Jump to the FCC's auction website, however, and you'll see the auction action continues. Auction round 175 will start in about 30 minutes.

Only 13 new bids were received in round 174, with licenses still contested for places like Minot, ND and Salisbury, MD. Just about all of the major results are already resolved, except that technically all auctions remain open until they all are ready to close. But C-block hasn't seen a new bid since round 90. D-block, the spectrum loaded with public service obligations, received a single bid back in round 1.

There is light at the end of the tunnel. The FCC has ramped up the pace to 10 rounds per day, and an industry analyst was quoted by Reuters as saying the auction should be wrapped up within a week.

Michael Giberson

From the EU Energy Policy Blog, Thomas-Olivier Léautier contemplates the factors that contribute to efficient investment and management of transmission systems:

As power engineers and economists have known for a long-time, the transmission grid is essential to the operation of well-functioning electric power markets. Yet, grid expansion in several regions has been nil or slow. By reviewing the main prescriptions from academic literature and comparing them with case studies from over 16 jurisdictions we find that unless the governance structure is appropriate and specific incentives are provided, grid expansion proves elusive.

After a brief examination of how physical limits and operational practices jointly contribute to transmission capacity, Léautier summarizes the results of theoretical and empirical work he did with Véronique Thelen:

The empirical evidence collected in this study appears consistent with theoretical prescriptions: creation of an independent transmission company subject to strong grid expansion incentives lead to congestion alleviation (e.g., England and Wales).

On the other hand, the Independent System Operator (ISO) model pursued in the United States, where operation but not ownership of the grid is vertically unbundled, does not appear to have performed well. This is consistent with “theoretical” predictions: transmission asset owners in the ISO model face mixed incentives for expansion resulting from limited vertical unbundling of the assets.

Michael Giberson

Over 200 companies have qualified to participate in the upcoming FCC spectrum auction. While that sounds like enough to create a lot of competition, as the auction approaches FCC chairman Kevin Martin is beginning to sound a little nervous. Yesterday he expressed concern that problems in the credit markets and general economic conditions may discourage some bidders or limit their ability to participate, Reuters reported. However, "Martin said the auction must go forward since Congress has ordered the FCC to begin the auction by January 28."

Contributing to pre-auction jitters at the FCC may be the recent demise of Frontline LLC. Frontline had been expected to be one of the primary bidders for the "D Block" spectrum, the band of airwaves that is being offered with a bunch of strings attached in terms of priorities for public safety usage, build-out requirements, and other limits. The Frontline folks were heavily engaged in the regulatory processes at the FCC creating the D Block requirements. Last week the company concluded it didn't have enough financing to go into the auction, so it closed up shop. (Grant Eskelsen at the Progress and Freedom Foundation says the failure of Frontline should be no surprise, and hopes it will be the "end of what has been a tragically flawed experiment in the D-block from the outset.")

What happens if the auction is a bust? The reserve prices adopted for each block are: Block A - $1.8 billion; Block B - $1.37 billion; Block C - $4.6 billion; Block D - $1.33 billion; and Block E - $900 million. If the reserve prices for blocks A, B, C, and/or E are not met in the auction, the FCC will close the auction for that block and set a new date to auction the unsold spectrum. According to this summary of the FCC auction rules, the new auction would use the same reserve prices as before, but the FCC may alter performance or access requirements imposed on the winners. The new auction would have to begin within a few weeks of the conclusion of the first auction.

According to the FCC rules, "If the reserve price established for the D Block of the 700 MHz Band is not satisfied by the results of Auction 73, the Commission may decide to re-offer that license subject to the same service rules or reconsider the rules applicable to that block." The WSJ reported yesterday that the FCC could give the spectrum to the highest bidder even if the bidder didn't reach the $1.33 billion reserve. Silicon Alley Insider suggests that some interesting gaming may result.

Michael Giberson

The Federal Aviation Administration is proposing to change its policy toward landing fees to "provide greater flexibility to operators of congested airports to use landing fees to provide incentives to air carriers to use the airport at less congested times or to use alternate airports to meet regional air service needs." As explained in the notice issued by the FAA, the proposed policy change will not allow true congestion pricing, because they will not allow airport authorities to charge prices sufficient to balance demand will capacity without regard to "allowable costs of airfield facilities and services." Instead, the FAA is proposing to allow airports to re-shuffle currently allowed costs in ways which reflect congestion at airports.

However, the proposal would clarify existing policy that allows airports to charge "dual element" landing fees. Most airports rely on a single element weight-based landing fee. The notice explains that an airport's fees could be revised to incorporate both a per-operation component and a aircraft weight component "so long as the two-part fee reasonably allocates costs to the appropriate users on a rational and economically justified basis." Such a shift would at the margin tend to promote use of larger aircraft into the airport without any other changes in rates. The FAA said the presence of congestion would be one factor that could be taken into account in revising rates. In particular, the per-operation component of the landing fee should vary according to the times congestion is present, said the FAA, if congestion is used to justify the change in fees.

Any airport can switch to a two-part landing fee. The FAA is proposing two other changes that only congested airports would be permitted to use. One change is a proposed ability to add the costs of facilities under construction into current rates (at present airports are allowed only to charge for facilities in use). The second change would allow airport authorities operating multiple airports to shift some costs from uncongested regional airports into the fees charged by the authorities congestion airports. This second proposed change is subject to several limits generally intended to insure that users of the congested airport can benefit from the shift in traffic expected to follow a shifting in costs.

As the Washington Post reports, not all in the air travel business are happy with the proposal, but that reaction is not a surprise. Here at Knowledge Problem we have long favored the economically-sensible approach of airport congestion pricing. While the proposal may not be pure congestion pricing, it would appear to allow airports to make significant steps in that direction.

The FAA is accepting public comments on the proposal, see the FAA notice for details.

Extra credit topic: Research in networks suggests that congestion rents can in some cases flow to non-congested network elements. Think, for example, of a generator connected to a high demand area by a congested line. Rather than the transmission line capturing all of economic rents, the generator may find it can profitably raise its rates and capture some congestion rents itself. The idea suggest the possibility that airports with departures headed into congested airports might find a way to extract some of the possible rents. Of course, that would require a little strategic sophistication on the part of airport rate authorities, and given the reactions reported in the Post's story it seems that airports are not exactly excited about using the modest amount of rate flexibility they already have.

Michael Giberson

The New Jersey experience illustrates one other key insight into regulation of a network. In a network industry such as electricity, where the network extends across state lines, it is possible for one state to exert significance externalities on others. The siting of transmission lines or an electricity generator in one state can alter supply conditions in other states. Moreover, if transmission is natural monopoly, one has to decide how to pay for it. Attempts by one state to pay less could increase the burden on other states. It is not obvious that the optimal geographic scope for unified regulation necessarily follows state or country boundaries.

That's from "Mergers in Regulated Industries: Electricity," by Dennis Carlton of the University of Chicago, currently serving as the Deputy Assistant Attorney General for Economic Analysis. As the title of the paper indicates, the subject is primarily merger analysis in electric power markets. Carlton uses the abandoned merger between Exelon and PSEG for illustrative purposes. The "New Jersey experience" he refers to is the inability of the merging companies to sufficiently satisfy regulatory demands made by the state.

While a now-abandoned merger proposal may in itself be of little interest, the more general issue of mergers in network industries remains important. What's more, the question of appropriate policy for networks crossing political borders will only become more important as the efficient size of networks grows larger and larger.

(HT to the Antitrust & Competition Policy Blog.)

Michael Giberson

The NCAA basketball season is well underway, and soon enough March Madness will be here. Do you have your brackets worked out yet?

Last year while working out a few thoughts on arbitrage opportunities in basketball tournament prediction markets at Inkling, it occurred to me that the Inkling pricing mechanism was just a little bit off for such applications. The question is whether something better can be done. An answer comes from the folks at Yahoo Research: yes.

This is a long post likely only of interest to prediction market geeks, a few computational theorists and some mechanism design people. You have been notified.

Michael Giberson

A few recent news articles on congestion has Peter Klein at Organization and Markets asking, "Why the Resistance to Pricing?"

When the quantity demanded exceeds the quantity supplied — causing shortages, delays, congestion, misallocation — the solution is to raise the price. Every freshman economics student knows this. Why, then, are regulators, industry groups, and consumer representatives so often opposed to rationing by the price mechanism?

Klein offers two examples, the airport congestion issue (citing a Reason Foundation commentary and my blog post of last week) and a Wall Street Journal story on internet congestion. In the case of pricing proposed to help clear airport congestion, Klein draws upon the quote I used from the Air Transport Association, "We are unalterably, adamantly opposed to it." From the Wall Street Journal article on internet congestion, Klein notes the extensive interest in technological solutions -- bigger, better, faster, more. "All purely technological remedies. No mention of pricing," he says.

Scott Wallsten riffed on the same topic in a commentary in response to news report that internet service provider Comcast was sometimes limiting capacity usage by customers who use file sharing services:

While this story immediately degenerated into a fight over net neutrality, economists' ears should have perked up. If network traffic needs to be "managed," then something is probably wrong with prices. Getting prices right--by charging heavy users for the costs they impose on everyone else, for example--would go a long way towards reducing the need to manage the network.

Wallsten draws on analogies to road use, electric power, and water utilities, to suggest that rates for high-speed internet use could be priced by some mechanism other than a flat access fee. "Consider highways. ... Policymakers have generally tried to deal with congestion by building more roads" -- the technological solution rather than pricing. Pricing proposals to address congestion, like HOT lanes, often face opposition even from parties would be benefit from faster travel.

Klein asks, "Is [the opposition to pricing] simply Bryan Caplan’s anti-market bias? Is it interest-group politics? Or is there something specific people don’t like, or don’t understand, about prices?"

It is possible that any proposed price will make some people worse off, possibly almost unavoidable, so some part of any opposition may be simple self-interest. But I think there may be a deeper phenomena at work, in Klein's words, "something specific people ... don't understand about prices."

People naturally understand the allocative function of prices -- the paying you X so I can get Y -- but have a harder time understand the coordination function provided by prices -- the paying you enough X so that other people don't take all of the Y first. Particularly when the value of the coordination function varies dramatically over time (rush hours, peak loads, high season at vacation spots), it seems harder to grasp the abstract service of coordination provided by prices.

Maybe there is work in the behavioral economics literature about congestion pricing, or maybe some sophisticated economics experiments have teased out these differences. I don't know, but if not it seems like a promising research topic: Why the Resistance to Prices?

Michael Giberson

A New York Times article reports on negotiations initiated by the U.S. Department of Transportation seeking to get airlines to voluntarily give up landing slots at Kennedy Airport, one of the nation’s most congested airports. (Many other stories on this topic are available.)

After a pep talk by the secretary of transportation, Mary E. Peters, and the acting administrator of the Federal Aviation Administration, Bobby Sturgell, the airline executives were taken to separate rooms and brought back one by one to talk to government officials about their schedules.

At some hours, Kennedy has more than 100 scheduled arrivals and departures. The F.A.A. said the airport actually handled 80 or 81 per hour this summer, which is the maximum the Transportation Department wants the airlines to schedule.

The airlines said Kennedy could handle more with better equipment and procedures, and have complained that the department’s target number is too strict. Another problem is that some traffic may migrate to Newark, adding to delays there.

The government is hopeful it can get “voluntary” reductions, which would then be codified into a regulation. If the airlines do not “volunteer,” the government has said it could set quotas and assign slots. But, Ms. Peters said, “We have high hopes for market-based incentives.”

The D.O.T. has said it may order landing fees that vary by the hour as an incentive to move flights to off-peak periods. But Ms. Peters said, “We may very well need scheduling reductions to help solve congestion in the near term.”

… Years-old federal controls on how many planes can use Kennedy ended on Jan. 1. Since then, traffic jumped 20 percent, according to the F.A.A., to 1,200 flights a day from 1,000. In August, it was 1,300 flights a day.

According to the F.A.A., one result is that there are 77.4 delays per 1,000 landings or takeoffs so far this year, continuing a steady rise — there were 20.9 in 2003, 27.5 in 2004, 39.6 in 2005 and 60.4 in 2006.

The meetings did not run smoothly. USA Today: “The government's effort to cut record flight delays at New York got off to a bitter start Tuesday when the airlines' trade organization threatened to challenge new controls in court.” The Cox News Service: “It appears that the federal regulators ‘intend to impose cuts,’ said James May, president of the Air Transport Association, which represents the major carriers. ‘We are unalterably, adamantly opposed to it.’ ”

Economists at the U.S. Department of Justice recently produced a paper advocating the auction of airport take-off and landing slots. Tom Whalen, Dennis W. Carlton, Ken Heyer, and Oliver Richard explain the source of some of the problems and some suggested approaches:

Airlines’ private incentives to schedule flights to serve more destinations and offer passengers more choice in departure times do not take into account the delays that their own flights impose upon other airlines because airlines do not face the proper price incentives to use scarce airport capacity. Consequently, airlines schedule too many flights, generating delays that ripple across the highly integrated airline network and adversely affect all passengers. One approach to solving this problem might be to get the airlines together and have them collectively hammer out a solution. … Such collective decision-making would not necessarily benefit consumers. Indeed, collective decision-making by actual and potential rivals raises serious risks to competition.

The current approach is only slightly more problematic than industry-based collusion – as described above the Transportation Department employs, in essence, a combination of administrative fiat, bureaucratic saber rattling, and moral suasion to extract reluctant corporation compromise. You don’t need a Nobel Prize for mechanism design theory to spot the flaws in this system.

Whalen, et al., write, “Our preferred method to allocate scarce airport capacity is to auction slots for landings and takeoffs by time of day and to convey upon their purchasers well-defined property rights.” The slots would be resalable, so airlines have the flexibility to reorganize their flight schedules to changing demand, and could come with cancellation priorities that would come into play if weather or other conditions temporarily reduced capacity at an airport.

While the airlines oppose administrative cuts, most proposed market-based mechanisms really seem to make them crazy. The Washington Post story this morning quoted airline executives as saying that “congestion pricing and caps will curtail flights to towns and cities served by smaller planes,” and complaining that the government hasn’t done enough to expand capacity through New York airspace and at the Kennedy airport.

Of course the point is that flights are already being curtailed, via congestion that spills over through the air transportation system. Only, in the present system much of the risk and cost is hidden in unrealistic flight schedules that leave travelers guessing rather that made transparent through prices that adequately coordinate consumer preferences and air transportation costs.

The proposal to auction airline takeoff and landing slots has been around for at least since 1982, when “A Combinatorial Auction Mechanism for Airport Time Slot Allocation,” by Stephen Rassenti, Vernon Smith, and R. Bulfin, was published in the Bell Journal of Economics.

Strictly speaking, auction of airline slots could be done airport by airport, and done only for highly congested airports. While such an approach would be simple to implement, it presents bidding challenges to the airline. If the auction for Airport A happens before the auction for Airport B, and the airline wants to fly a route from A to B, it needs to know the price for landing at B before it can submit an efficient bid for a takeoff slot at A. The Rassenti proposal, the first published description of a combinatorial auction, provides a mechanism for addressing these and other complications.

The Washington Post noted, “operations are complicated by the fact that an unusual range of aircraft types use JFK's runways, a mix of small regional jets, medium-size planes and wide-body jets. Smaller planes need more space to take off safely behind larger jets. Properly sequencing those flights during busy periods can be a challenge…” A combinatorial auction is designed to manage these kinds of interacting constraints on the system. It may sound complicated, but remember that airlines have a great deal of experience in using dynamic pricing systems to allocate scarce resources when the object is maximizing ticket revenue paid by consumers. Slot auctions just employ similar tools to induce airlines to better coordinate their use of scarce air travel resources. Yes, consumers traveling through popular airports at popular times will end up paying higher ticket prices, but at the same time they will be more likely to arrive on schedule.

Various proposals for extending and improving upon the ideas developed in the Rassenti article have appeared over the last 25 years – but really, improvement is beside the point, the point is to get started. While airport-by-airport auctions may be the technological equivalent of a barnstormer’s bi-plane, the current administrative jawboning is no more than a hot air balloon.

[HT to the Antitrust & Competition Policy Blog for the link to the Whalen et al. paper.]

Michael Giberson

Duquesne Light has announced it wants to drop out of PJM and join the neighboring Midwest ISO, citing the high costs emerging from PJM's capacity market as their motivation. The capacity market is called the "RPM" market after the "reliability pricing model" which serves as the underlying pricing mechanism. Duquesne has filed a request with FERC seeking a Commission order directing PJM to exclude the utility from the next running of the RPM auction, scheduled to begin next week. The Energy Legal Blog presents a few more details, and notes a few other cases in which utilities are eying the exits.

The Pittsburgh Tribune-Review made brief mention of the request in a column of energy-related notes:

The Downtown-based utility joined PJM in 2005, but said the organization's price changes since then would increase its costs in coming years. "We don't feel that is what is best for customers," spokesman Joe Vallarian said. ... The company said it may join another grid operator based in Indiana; it wants to avoid costs related to the PJM auctions.

The Electric Power Supply Association (EPSA) responded in the filing at FERC requesting the agency to deny Duquesne's request. In a news release, EPSA stated:

"FERC should deny Duquesne's complaint on the grounds that Duquesne remains a signatory to PJM's reliability agreement and, contrary to what Duquesne has claimed, there is no credible information that the utility will not be in PJM in 2009," said John E. Shelk, President and CEO of EPSA.

EPSA is the trade association representing the "competitive power supply industry," i.e. non-utility generators, and its members will be net recipients of payments from the PJM's RPM markets. Duquesne Light, on the other hand, as predominantly a wires-and-retail power services company will be net payers into the RPM markets.

Michael Giberson

Quoting Grimmelmann:

First off, the Times pleads its inability to re-report every challenged story. Fair enough....

It’s one thing not to revisit stories as new information becomes available. (The Times isn’t Wikipedia, after all, and we shouldn’t hold it to the same higher standards of timeliness.) But it’s something else not to append corrections to articles whose reporting was no good. The Times pleads lack of resources: We’re too busy making mistakes to fix them.

The Times isn't Wikipedia? Correct. Just so.

Michael Giberson

I stumbled across the Home of the Groove blog some months back. It turned out to be a great place to delve deeper into New Orleans funk and R&B and related musicality. And recently, the Home of the Groove has sprouted its own internet radio station running a 10-hours and growing playlist of New Orleans lost hits, arcana, and other rarities.

Internet radio is not dead yet. Let's hope it survives the continuing loving embrace of SoundExchange.

Michael Giberson

Yesterday, PJM Interconnection announced the sudden retirement of CEO Phillip Harris. In a press release, the PJM board stated:

We are announcing that President Phillip G. Harris has elected to retire from PJM. We want to acknowledge Mr. Harris’ considerable contributions to PJM and our industry over his many years of service. Without his guidance and vision, PJM would not enjoy the reputation for technological and operational excellence that has made it the benchmark for regional transmission organizations the world over.

Restructuring Today reported today that Harris was asked to leave the position after (allegedly) telling PJM staff at a PJM event that the organization never wanted a market monitor in the first place. PJM remains embroiled in a dispute with its market monitor over the degree to which the organization interfered with the ability of the market monitoring unit to do its job. As Harris discovered, it is a very sensitive subject.

Background - previous posts on market monitoring and on PJM's market monitor troubles:

U.S. Senator Weighs in on PJM Dispute with Market Monitor
PJM Market Monitor Reports Interference from Management
The Kluge That We Need: Local Market Power Mitigation Measures
Ex Post Market Monitoring in Electric Power Markets
Why do we have Electric Power Market Monitors?
Market Monitors in Electric Power Markets, II
The Role of Market Monitors in Electric Power Networks

Michael Giberson

A Swiss software security research company, WabiSabiLabi, is establishing an online auction site to allow security researchers to auction off discoveries of software vulnerabilities. In their press release, they said:

Recently it was reported that although researchers had analyzed a little more than 7,000 publicly disclosed vulnerabilities last year, the number of new vulnerabilities found in code could be as high as 139,362 per year. Our intention is that the marketplace facility on WSLabi will enable security researchers to get a fair price for their findings and ensure that they will no longer be forced to give them away for free or sell them to cyber-criminals.

Yes, they will screen the bidders in the effort to determine that they aren't "cyber-criminals," and they will test reported vulnerabilities before allowing an item to be put up for auction. The Washington Post described vulnerability researcher Dino Dai Zovi as excited about the vulnerability auction service:

"I can see this service creating much more incentives for researchers to find flaws," Dai Zovi said. "Not everyone is willing to spend 20 to 40 hours looking for vulnerabilities in software just to receive a little thank-you note in Microsoft's security advisories."

The discovery of software vulnerabilities provides something of the nature of a network or club good. Presumably the software vendor - the provider of the initial good or service around which the network grows - would have an incentive to pay for acquisition of this information so as to put out a better product. But if the standard offer of payment is "a little thank-you note," perhaps the existing market for such intellectual property is not yielding competitive prices.

Michael Giberson

From The Star-Ledger:

U.S. Sen. Robert Menendez today urged a federal agency to hold hearings into whether the company that runs the regional power grid is undermining the work of a monitor whose job is to guard against manipulation of electricity prices.

In a letter to the Federal Energy Regulatory Commission, Menendez (D-N.J.) said allegations by the monitor, Joseph Bowring, against grid operator PJM Interconnection are troubling. The senator said there is mounting evidence that PJM management is undermining the monitor's work by stripping him of resources, denying him vital data and censoring his reports.

Michael Giberson

The tech trade press seems abuzz with speculation over Google's interest in the FCC’s spectrum auction rules. Google phone? Google wireless? Google the spectrum monopolist? Google turning your cell phone into a mobile AdWords delivery device?

All of this wild-eyed speculation seems a bit over the top to me, perhaps because the essential idea proposed by Google seems so sensible. In the letter (link below), Google wrote: "In particular, the Commission should clarify that the service rules governing the 700 MHz bands already allow the use of dynamic auction techniques, such as real-time auctions and per-device registration fees."

Here’s the big idea: dynamic auction techniques, such as real-time auctions. Rather than having a telecom company buy a large chunk of spectrum in the hopes of eventually being able to fill it with subscribers, a spectrum wholesaler would own and auction off spectrum usage rights in small slices. Telecom companies could buy what they needed as they needed it. Or, eventually, wireless devices would buy the spectrum bandwidth needed, when they need it.

Prices in this resell auction would naturally be dynamic -- going up and down with the demands placed on the available capacity -- and dynamic pricing seems to make some people nervous. But really it is just about, as Google says in the letter, devising a system under which a "particular slice of spectrum ends up in the hands of the user who values it most at any particular time and place." If your wireless fax isn’t time sensitive, just say so and your device will bid low. Need to talk to someone right now? Go ahead and pay the current price, and less time-sensitive uses of the spectrum will get out of your way. Most retail consumers would likely continue to sign up for wireless phone services like they do now – paying fees related to the level of service demanded – and the service provider taking the price risk in the secondary spectrum market. Over time, wireless devices, and their owners, will get smarter about buying and using the kinds of service they need.

Notice that two separate auctions are involved here. The first would involve the FCC auctioning off large amounts of spectrum to interested buyers, using one of the FCC's existing auction designs or some improvement thereof. Google is proposing that the winners in that first auction be allowed to resell some or all of the spectrum purchased in a secondary auction, using dynamic auction processes and systems they hope to manage.

Adam Kovacevich, a spokesman for Google, said, "In general, it's the belief of a lot of people in the company that spectrum is allocated in an inefficient manner." The letter cites one report indicating that use in “any given geographic area averages some 5 percent of total available spectrum.” They think they can do better, and in the letter to the FCC they ask whether the rules will prevent them from trying.

NOTES: The Google letter to the FCC can be found at this link. (Thanks to Computerworld.) Another somewhat related news story, concerning a lawsuit over FCC rule changes in an earlier auction, is up at MarketWatch. Yesterday I posted Auctioning the Airwaves, Google Style.

Michael Giberson

Google filed a proposal on Monday with the Federal Communications Commission calling on the agency to let companies allocate radio spectrum using the same kind of real-time auction that the search engine company now uses to sell advertisements.

The NYTimes article is a little short on the technical details of the proposal (I guess not everyone is a market design geek). Usually these filings turn up on the FCC website, but in a few brief attempts I haven't turned up anything yet. I'll let you know when I find it.

In the meantime, some additional details and commentary can be found at O'Reilly's Radar, The Next Net, GigaOM, and bloggingstocks.

If it turns out that spectrum will be available via a continuous, dynamic auction, similar to how Google sells space for advertisements, it will raise the significance of auction theory work on the properties of that market design. For economists and other technical types, a good place to start is Hal Varian's paper, "Position Auctions." If you are new to auction theory, begin with Ken Steiglitz's Snipers, Shills, and Sharks, a fine introduction. While the book focuses on eBay-style online auctions, it also discusses and provides citations on position auctions.

Lynne Kiesling

OK, this is one the better April Fool's Day techie pranks I've seen in a while:

MOUNTAIN VIEW, Calif., April 1, 2007 - Google Inc. (NASDAQ: GOOG) today announced the launch of Google TiSP (BETA)™, a free in-home wireless broadband service that delivers online connectivity via users' plumbing systems. The Toilet Internet Service Provider (TiSP) project is a self-installed, ad-supported online service that will be offered entirely free to any consumer with a WiFi-capable PC and a toilet connected to a local municipal sewage system. ...

"I couldn't be more excited about, and am only slightly grossed out by, this remarkable new product," said Marissa Mayer, Google's Vice President of Search Products and User Experience. "I firmly believe TiSP will be a breakthrough product, particularly for those users who, like Larry himself, do much of their best thinking in the bathroom."

Hat tip to Slashdot.

Tee hee!

Michael Giberson

Not much analytical work has be published looking at the role of market monitors in regional power markets. One of the few pieces to be published in the trade press was an article by Prof. Robert Michaels in Public Utilities Fortnightly, "Watching the Watchers: Can RTO market monitors really be independent?"

Writing in 2003, Michaels calls market monitoring institutions (MMIs) a "a novel regulatory tool never before contemplated in legislation," and says "no regulatory institution has ever achieved so much centrality with so little forethought."

Market monitoring is generally presented as a pro-consumer institution that complements government regulatory oversight. Michaels said in the case of the California Independent System Operator and the now-defunct Calfornia Power Exchange, history suggests an alternative interpretation.

Michael Giberson

Economists have devoted a great deal of attention to market power in electric power markets in the somewhat general, structural sense, but very little of that work focuses much attention on networked-nature of the industry. The network matters a lot – that’s one of the views well articulated in the Thomas, et al., paper mentioned here last week – and it is the networked nature of the industry that accounts for the emergence of the institution of market monitoring in the integrated power markets.

The Federal Energy Regulatory Commission is hosting a technical conference April 5 to “explore the effectiveness of market monitoring both in performing market oversight and in serving a variety of interested stakeholders.” In a recent notice, FERC listed several questions it wanted panelists to address. The fourth question is where things really get interesting….

Michael Giberson

There were several bits of competition-related news items in Friday's Washington Post. Catching my interest were Steven Pearlstein’s column advocating against the XM-Sirius merger (See also Howard Kurtz article from Wednesday and Rob Pegoraro's tech column on Thursday) and a story on the passage of a bill by the Virginia state legislature that would terminate the state's botched attempt at restructuring retail electric power.

But the news that really caught my eye was an article suggesting that, in a small way, water utility competition was breaking out in parts of Fairfax County, Virginia. I live in the city of Falls Church, a small independent city of about 10,000 people surrounded on three sides by Fairfax County, with a population of about 1,000,000. The city of Falls Church runs its own water utility, and for a variety of historic reasons the city's water utility also serves a significant portion of the population in parts of the county surrounding the city, including much of the fast growing area around Tyson's Corner and Merrifield. As the article explains:

Governments at all levels often scuffle over water, but the disputes usually involve who gets to use it or who has to clean it up. Yesterday's action by the city is unusual because it centers on who has the right to sell it.

"Yesterday's action" was the filing of a lawsuit by the city against the Fairfax County water authority. Apparently the formal agreement between the city and the county, by which they had historically parceled out monopoly water utility territories, expired in 1989. Until recently the county continued to respect the historic boundaries, but now the county is engaging in "a deliberate and concerted action" (in the words of the lawsuit) to interfere with the city's customer base.

And of course by "interfere with the city's customer base" what the article means is that the county water authority was offering a developer a much better price than city water utility customers get.

While my short-term economic interests fall in with the city in this case - the city water utility generates a little money that supports the broader city budget and in theory keeps my city taxes lower than they otherwise would be - I couldn't be happier to see utility competition breaking out in Virginia. Let's hope it brings benefits to all the local water consumers!

UPDATE: Here's the Falls Church News-Press article on the lawsuit.

Michael Giberson

XM and Sirius, two satellite radio networks, announced plans to merge yesterday. Amusingly, in the New York Times the story begins with "The nation’s two satellite radio services, Sirius and XM, announced ...", while in the Washington Post leads with ''XM and Sirius, the two satellite radio companies ...." In each case the hometown company goes first.

Both stories highlight the apparently high antitrust standard the two companies must overcome to gain approval for the merger. The Post:

The FCC bars a single company from controlling the satellite radio market, but FCC Chairman Kevin J. Martin recently noted that such rules can be changed. Martin said yesterday that the hurdle "would be high. . . . The companies would need to demonstrate that consumers would clearly be better off with both more choice and affordable prices."

As the Times explains:

An army of merger and antitrust lawyers for both sides worked several marathon weeks of conference calls and trips to Washington to gauge the political climate for the transaction before opining that the deal should pass regulatory muster. Simpson Thacher & Bartlett and Wiley Rein are representing Sirius; XM is being advised by Skadden, Arps, Slate, Meagher & Flom; Jones Day; and Latham & Watkins.

Doesn't it seem a little silly that federal regulators are suggesting somehow the world would be worse off with one satellite radio company (for the time being) where as of a few years ago there were none? Do consumers have a right to two money-losing national music, talk and news services? Doesn't the FCC know that by raising barriers to exit, they create barriers to entry for some future satellite radio rival?

I'm not a subscriber to either service -- in fact I just barely had a CD player put in to my car a few months back when I started commuting to an office. The CD player also plays mp3 files and has an audio imput so I can plug in an iPod. I don't really need more options for in car entertainment, but I've been tempted to go satellite after conversations with a few passionate fans.

You know, just maybe if they called up Texas Fred, the Zydeco Cowboy and put him on coast-to-coast, I might have to do it.

Lynne Kiesling

Today's Wall Street Journal has an article on Stockholm's road congestion pricing pilot experiment (subscription required). Stockholm is a city of islands, so the road network is subtantially a set of bridges. Not surprisingly, congestion often ensues.

From January through July, Stockholm tested one of the world's most sophisticated traffic-management systems as part of a plan to reduce gridlock, lower smog levels and improve quality of life in the city. Unlike most other traffic-control plans in place in cities such as London and Rome, Stockholm used a dynamic-pricing system in which drivers were charged different amounts depending on the time of day. If Mr. Astrom, for example, left the city center at the busiest time of the afternoon rush, from 4 to 5:29, he would have paid the equivalent of $2.76. But by waiting until 6:30 p.m., he traveled toll-free. "People changed their habits," he said.

Traffic volume fell from 9% to 26% at the major tollgates.

The engineering to make this system work is substantial. IBM is one of the technology partners that developed the ability to identify a car within milliseconds at one of the 23 tolling gates around the city.

Mr. Westman [of IBM] worried that the system wouldn't be able to identify cars in the harsh Stockholm winters because of all the salt, snow and dirt on the roads and vehicles. But the trial period went smoothly and the cameras functioned well in the winter months. IBM's customer-service center, which anticipated 30,000 calls a day, fielded just 2,000 a day; and few appeals of charges were filed to the tax authorities.

Stockholm officials also found evidence of a basic tenet of economics: people change their behavior in the face of changes in relative prices. Take the case of public transportation, which is extensive in Stockholm.

The Stockholm trial produced another insight into a vexing traffic-reduction programs: getting people to use public transportation. Before the trial began, Stockholm spent about $180 million on improvements to public transportation. It bought about 200 new buses, and added rush-hour trains, express bus routes and more park-and-ride lots. But the changes had little impact on the number of people who left their private cars at home. In spring 2006, however, during the trial, use of all forms of public transportation jumped 6% and ridership on inner-city bus routes rose 9%, compared with a year earlier.

If the cost of driving to the consumer more transparently reflects the actual cost of an individual driving, at the margin that cost will change the behavior of those consumers who don't value being able to drive at that time of day; they either time shift or take public transportation. People respond to incentives.

Stockholm is also using this experiment as an experiment in democracy: they are putting the congestion pricing to referendum. The six-month trial period has ended, and residents will soon vote in a referendum on whether or not to keep the system. Officials say that if a majority votes against the system they will dismantle all of the equipment in which they have invested, which was extremely expensive. But with the reduction in the peak/offpeak commute time ratio from 3/1 to 2/1, many Stockholm commuters are in favor of the system, as are cyclists, who enjoy a less stressful commute, even though the toll has induced more people to ride their bikes to work instead of driving.

Michael Giberson

The NYTimes has a story on municipal WiFi, focused on Tapei, Tiawan. You get a sense of the story line from the title: "What if They Built an Urban Wireless Network and Hardly Anyone Used It?"

That such a vast and reasonably priced wireless network has attracted so few users in an otherwise tech-hungry metropolis should give pause to civic leaders in Chicago, Philadelphia and dozens of other American cities that are building wireless networks of their own.

Like Taipei, these cities hope to use their new networks to help less affluent people get online and to make their cities more business-friendly. Yet as Taipei has found out, just building a citywide network does not guarantee that people will use it...

"There is a lot of hype about public access," said Craig J. Settles, a technology consultant in Oakland, Calif., and author of "Fighting the Good Fight for Municipal Wireless." "What's missing from a lot of these discussions is what people are willing to pay for."

The problem isn't technology, so much as business model, the story reports.

But even if Q-Ware meets its target this year, the company will need 500,000 users in a given month to break even, a target it is not expected to hit for several more years, according to Chou Yun-tsai, the chairwoman of Taipei's Research, Development and Evaluation Commission, which oversees the WiFly project.

"It's a huge task," Ms. Chou said.

Lynne Kiesling

Today is a dynamism day here at KP, whether it's marketing or technology or business models. Business Week has an interview with Frank Moss, the new head of MIT's MediaLab. He points out the importance of several things that I think are crucial for dynamic innovation: entrepreneurship, new models of collaboration and invention, and focusing on disruptive change instead of incremental tweaks:

How do you view the nexus between technology and entrepreneurialism?
It is hugely important. In fact, entrepreneurs are really the primary vehicle for innovation in our society. They've played an incredible role. Thirty years ago, the primary source for innovation was large corporate labs. That is where all of the money went. Then, 20 to 25 years ago, the source of ideas and creativity shifted to venture funds and startups.

Over the past 20 years, we've seen the economy and society change due to innovation from small independent efforts outside of corporate labs. Technology has enabled startups to have a big influence, and consequently they have had a tremendous effect in the technology scene today.

What role will startups play in the future?

I see tremendous economic growth from startups from 10 years ago. Entrepreneurs will go from the 1,000 startup ventures funded in the last 10 to 20 years to ideas coming from people working together in network-based environments, using computers to dream up innovations in a way they never did before. It could be people in developing countries with low-cost computers.

The Media Lab has given a start to many entrepreneurs. What would be your advice to would-be entrepreneurs in today's environment?

Resist the current temptation to make incremental changes to attract funding. It might get you off the ground, but I don't think it will get you very far. Today, the funding climate has changed. The successful (entrepreneurs) will look for fundamental disruptive change. I encourage them to take risks, rather than just polish the faucets. There will always be an appetite for game-changing technology.

Note in particular his focus on distributed, decentralized, networked innovation. This way of thinking about innovation taps in to the idea that creativity resides in many distributed individuals, and that a distributed network may do a better job of generating beneficial disruptive change than a centralized, coordinated, top-down lab approach.

I think it makes for an interesting challenge for an organization like MediaLab; at some level, it will be harder to direct and target resources on projects that are likely to be transformational, because so much more innovation will be distributed and not concentrated in big labs. At another level, though, MediaLab can seed particular projects or individuals who are working in a distributed network, and that bottom-up approach might be more likely to lead to successful disruptive technologies than the coordinated large-lab approach.

Thanks to Glenn Reynolds for the link.

UPDATE: Paul at Truck and Barter has a link-filled post on the same question, inspired by an article in the Economist on a recent AEA session on entrepreneurship. Note his link to Chris Coyne's post at Austrian Economists, where he laments the absence of attention to Kirzner in the Economist article, a lament that I second.

Lynne Kiesling

This week marks the Electronic Frontier Foundation's 15th anniversary. EFF is a tireless warrior for preserving and enhancing freedom as communication and information technology evolves.

Cory Doctorow has a Boing Boing post discussing EFF's activities over the past 15 years:

This week marks the Electronic Frontier Foundation's 15th anniversary -- a decade and a half of changing bad laws, creating good court decisions, and building a technological civil liberties movement that now comprises dozens of organizations, activity all over the world, and millions of geeks with a burgeoning consciousness that the Internet isn't free because of its nature: it's been kept free by the struggles of activists and users who have fought back the forces of repression who would have tamed it and crimped it and rendered it little more than an AOL-1.0-style toy.

EFF is holding a commemorative blog-a-thon. Contribute your tales!

I have no activism story to tell, other than our continuing financial support of EFF. But as communication technology has evolved and become a more pervasive and valued dimension of my daily life, I treasure the freedom of online privacy, to blog, and to be free from government surveillance. These freedoms have brought about a flourishing of content, of substance, of culture, of community (and yeah, a flourishing of dreck, but that's what filters are for, and it's a small cost if it gets us all the other stuff).

For me, the freedoms that EFF fights to protect are a crucial set of rights in the fundamental property rights that allow free and responsible people to live together in civil society.

Blog-a-thon tag:
EFF15

Lynne Kiesling

Catching up on my reading ... Ray Gifford has a nice post on the role broadband over power lines can play in creating the distributed, interactive, active-demand-enabling, smart and resilient electric power network.

BPL potentially has a large role to play in enabling grid-friendly appliances and the customization of electric power service to heterogeneous customers. Say, for example, we want six-sigma power quality in the room in our house where our main computers and wireless router live, but are willing to pay less to get less reliable service throughout the rest of the house? BPL can help make that value proposition a reality. It also raises the opportunity to maintain sufficient flow to those appliances that need to have their clocks reset so that the clocks don't go off, but you cut the remainder of the power.

Oh, by the way, it's electricity goddess, not electricity geek. See if you can keep that straight [wink].

Lynne Kiesling

I don't have much to say about Grokster because I'm of many minds about it. Turns out I'm not alone; Tyler says economists should feel conflicted about peer-to-peer file sharing.

One aspect of Tyler's argument is what I call the "relevant and irrelevant externality" argument. Will file sharing reduce economic efficiency, leading to less-than-optimal amounts of music being produced? Depends on whether the externality created by the file sharing is relevant at the margin, and that's where a lot of the discussion has been taking place. I am also glad that he points out that a shrinking music industry is likely to be consistent with increased efficiency, due to the effects of technological change and disintermediation.

One area where I'm not sure I agree with Tyler is the extent to which consumers are oriented toward new music and not toward older music. I have two caveats, one systemic and one short-run. I think there is always likely to be some subset of the population that goes back and "rediscovers" the music of earlier years, which is likely to create some demand for the backward "long tail" of older music. And in the short run, the transition from vinyl to CD to digital will keep the demand up for older music for consumers like me, who had lots of stuff on vinyl and never bought it on CD, but are now buying it digitally. I did this recently with Jesus & Mary Chain, Love & Rockets, and an early Love Tractor album.

This Information Week article mentions using micropayment-based peer-to-peer file sharing as an alternative market institution. The fertile innovation and plummeting costs of new technolgies continue to reduce the transaction costs in such a system, so it becomes more feasible all the time. So ways to innovate Grokster to be legal during this transition to a more disintermediated industry are feasible.

Speaking of disintermediation ... it's been a while since I've put in a plug for Magnatune, an online record company with a great portfolio of artists, good royalties to artists, and good prices to consumers.

Lynne Kiesling

MIT Media Lab is performing a research survey on weblog author behavior. Please help them out and become a datum.

Thanks.

Lynne Kiesling

Everyone's talking about eBay's 10th anniversary. NPR is running a series of stories on Morning Edition. The Economist's print edition this week has a cover article on eBay, with clever art:

Much of the analysis focuses on the lessons to learn from eBay's success. The Economist article put it succinctly:

To succeed, firms need agility, an open mind and the ability to reinvent themselves repeatedly. Most of all, they need to listen carefully to their customers, paying close attention to what they do and don't want. ...

It is not impossible that some kind of monopoly might emerge, wielding true pricing power, but right now it looks unlikely. eBay's managers, for example, admit that customers are shaping its business more than they are, and seem acutely aware that groups of customers could easily depart together to set up their own specialist auction or sales sites if eBay charges too much for its services or lets them down.

When I give talks about dynamism and competitive electricity markets, I draw inspiration from eBay frequently, on precisely these two points. Market processes reward the agile, the flexible, the adaptable suppliers, creating value for them and for their customers beyond those imaginable in any alternative way of structuring transactions. eBay illustrates that potential in spades.

It also illustrates what happens when you empower consumers, even when you empower them in transactions that you aren't sure are that valuable to them. Given the platform, customers tell you which transactions are valuable to them and which aren't. 30% of eBay's volume is now from used car sales, which eBay employees didn't envision. Customer input and choice through their decisions provided the valuable information to potential suppliers that has led to the increase in used cars brought to market on the eBay platform.

But the second pull quote illustrates the value of platform contestability, the Schumpeterian dynamic on which we thrive. eBay can't rest on its laurels or start extracting rents, because these buyers and sellers could split off and start a speciality car auction site. That discipline keeps eBay innovating. And it illustrates why, in the same way that government-granted monopoly firms are a bad idea, government-granted monopoly market platforms are also a bad idea. This is my biggest criticism of the RTO/ISO regionalization in electric power -- RTOs/ISOs run the danger of becoming monopoly market platforms, because of the bundling of their system operation coordination function and their market platform function.

John Berlau made similar