How is the Internet like printing, the steam engine, the factory system, railways, electricity, and the computer? According to economists Richard Lipsey, Kenneth Carlaw and Clifford Bekar, the Internet has joined these and a handful of other technologies on a select list: the twenty-four “transforming general purpose technologies” developed over the entire span of human history. The Internet is in good company. These are all technologies with a broad range and variety of uses, extensive spillovers, susceptibility to improvement through the development of complementary technologies, and the ability to transform economic, political, and social structures.
In light of the Internet’s importance, its custodians – our generation in general and organizations like the FCC in particular – have a responsibility to preserve its dynamic character and maximize its potential for improving our lives. As I think about the FCC’s role, I see four key economic issues. One is the implications of possible market failures that might limit innovation in applications complementary to the network. These might include the transaction costs of contracting, spillover benefits of applications innovation, and perhaps the market power of Internet service providers or applications providers. The second is evaluating the social welfare effects of price discrimination on both sides of two-sided platforms. How do these effects depend on the extent to which mobile Internet service substitutes for fixed service for different types of customers? Or on the number of fixed and wireless providers available in a region? The third issue is ensuring adequate incentives for investment and innovation in the network – or the platforms, if wireless and wireline service are viewed as imperfect substitutes. The fourth is providing incentives for efficient network operation, perhaps through congestion pricing or network management.
Economists: this is your opportunity to educate me and the Commission. Am I thinking about our task in a productive way? What other economic issues should be on my list?
TTY: 1-888-TELL FCC (1-888-835-5322)
Before posting I just completed a “Find” operation in this page for the word “competition”. It failed to find any instance of that word here. How can you have a post on the Economics of a new ground-breaking technology, developed in a capitalist country (yet not necessarily by capitalist means — ARPANET and MILNET came way before AOL) and not comment at all on the need for greater competition in that arena? The closest I see to any mention here of the general lack of telecommunications competition, wired or wireless, in America is James V. DeLong’s mention of “regulatory capture.”
I know it’s not on your list of “transforming general purpose technologies”, but the national Interstate system, and other smaller public road systems, deserve a mention here. Just imagine what driving would be like now if we let the road systems go the way of the train rail system. I fear the Internet connection provision market is going more the way of the rail line than that of the open road.
Another way to say it is that the system of the Internet is mired in tolls, all taxed against the primary productive nodes, on either end of a connection that is seeking rents for every interaction. These rents are consistently exploited from the productive nodes that rely on it, despite the fact that the installation of that connection is largely a one-time cost, with logarithmically decreasing maintenance costs. It would cost less to have every adjacent node split the finance and installation and maintenance of their own inter-connections — even expensive installation with high loan repayment interest is consistently more cost effective than monopoly rents.
You can’t merely homestead a wire by putting a router on it, can you? You can’t homestead a portion of electromagnetic frequency spectrum either, however small, whether it’s visible or not. There are more colors than our eyes can see.
You asked two questions:
Am I thinking about our task in a productive way? You referred to possible market failures; market power; investment and innovation incentives; price discrimination; two sided platforms; efficient network operation; and, social welfare effects. These are interrelated and each suggests productive, and necessary, areas of economic inquiry. The challenge to you and other Commission economists will be a) to provide a consistent analytical framework for considering these and related economic issues and b) to insist relentlessly that maximizing consumer welfare be a central element of the Commission’s inquiries and objective function. That perspective too often gets lost in “public interest” rhetoric. In that context I am delighted by your mention of two sided platforms and price discrimination, since the Commission’s recognition of the role of each as sources of consumer welfare will be critical. Relatedly, perhaps the most productive task for Commission economists is to bring to bear the enormous amount of intelligence available in the economics literature as means of offsetting economically baseless conjectures by non practitioners.
What other economic issues should be on my list? Regulatory debates continue to center on market imperfections and the extent/type of government action needed to offset them. There is, however, an enormous economic and political science literature addressing imperfections in government processes and actions. Joe Stiglitz, Nobel prize-winner and former CEA Chairman of the President’s Council of Economic Advisors recently wrote: “Anyone who has watched the U.S. government in the last seven years is well aware not only of the possibility of government failure but also of its reality.” An important economic and policy issue going forward will be the need for, and the extent to which, the Commission’s policy decisions reflect a) clear and consistent assessments of different kinds of institutional failure and b) balancing at the consumer welfare implications of the infirmities of both private and public action.
Finally, I cannot emphasize enough the need for more facts and analysis respecting “innovation.” The record in several proceedings is full of assertions about innovations – their relative importance, their source, their motivators, the impact of regulation thereon, etc. Most of these are neither supported nor supportable by reference to the vast economic literature on the matter. Given the dynamic nature of information markets, the Commission cannot responsibly decide many of the issues before it on the basis of speculation and conjecture.
You raise lots of interesting issues. Here are a few thoughts. (For elaboration, see “Avoiding a Tech Train Wreck,” THE AMERICAN, May/June 2008. http://www.american.com/archive/2008/may-june-magazine-contents/avoiding-a-tech-train-wreck/?searchterm=delong.)
1) The comparison to the railroads seems more apt than the reference to other transformative technologies because the railroad was/is a platform on which other applications ride. Other such platforms are computer operating systems, search engines, or (to stretch it) WalMart. In any of these situations, the crucial factor is going to be the relations between the platform provider and the complementary applications. Personally, I think too much attention is paid to consumers and not enough to the intermediate producers who become dependent on the platforms.
2) Two of the major issues are price discrimination and sunk capital. Economists love price discrimination, and they do indeed make a great case for it, including the idea of Ramsey Pricing. Everyone else hates it, especially businesses who must depend on the platform. (I confess to occasional guilty sympathy for Robinson Patman.)
Sunk capital is important because many of the problems that arise with platforms are not really market imperfections – they come from the reality that in bargaining situations the loser is usually the party who has committed his capital first. In free market theory, this is handled by contract, but that is often not very realistic, given complexities of contingent contracts and uncertainties. One of the world’s oldest business models is “find some sunk capital and take it,” and there are many who approach the Internet in that spirit.
3) To a large degree, the advocates of the open internet and their critics are talking past each other. The open internet crowd worries about creativity and generativity and the track record of large corporations in stifling these things. They don’t worry much about investment, sclerotic government, or regulatory capture. (You mention investment, but that is not common in the open movement generally.) Critics worry about sclerotic government, regulatory capture, and the investment issue. They do not think much about nurturing generativity at the edges. There should be a synthesis here.
btw – I am on record as believing that Microsoft did a good job at seeking a middle ground in its Windows Principles, which are directed precisely at the problem of the relationship between a platform company and the producers of apps. See http://www.microsoft.com/about/corporatecitizenship/en-us/our-commitments/our-principles/windows-principles.aspx
Nice post, and a good reading of the lessons from the literature on the deployment and diffusion general purpose technologies. Perhaps a fifth lesson should be on the importance of policies that permit “co-invention” — namely, the additional inventive activity undertaken by users and firms to customize the new technology to the unique problems each face. As examples, think about the innovative ways suppliers have deployed wi-fi in the last decade, or the innovative ways enterprises have adapted the Internet to their data needs, or the innovative way users have adapted the web to sharing their videos. This type of activity happens coincident with first deployment and continues as technologies diffuse. Policy can play a nurturing role at the outset, but its role does not end there. Policy continues to matter at various stages in the diffusion of a new service.
In short, Information Economics.
If we take the idea of dynamic character to improve our lives, then let’s deconstruct a future improved Internet that is ubiquitous and connects everyone in the world. This would make the supply of information be defined as the creation of information for consumption and demand defined as the number of people that consume it immedietly and in the future. We need an exchange to track demand and consumption on a personel and business level. Information is as much a “goods” to be bought or sold for an amount. This “currency” will be a type of Information Credit which will allow for a new economy to form after we get tired of the current one. In transition the establishment of this system now will allow for many of the current concerns about the Internet to be addressed. Either the government or a private company will take the primary role as the Contact Manager, which will provide to the people and entities a primary contact reference, your Virtual Personel Settings, between you and other people. This will allow people to clarify the type of virtual connection the citizen wants to have toward another. Remember that this is a future where the Internet is on every screen you can see or touch and people will have programs that can move from their watch to their phone to their tablet to their tv and to local Internet walls, because the devices themselves exchange information with each other based on IPv6 MAC address. I call this the Rule of 5s, the concept is simple, 5 levels of personel settings (how YOU the citizen want to classify your connection with your real and virtual self to other people and businesses). The technology is already here, it’s only a question as to when will it be the majority. Strike now so it will already be in place when the time comes for the Home Agent Server to be the only real technology purchase that matters because all the others have been reduced almost to simple comodities, you can see this starting now as phones and computers and TVs get integrated together with touch and eventually AI interfaces. Your AI (being run on the Home Agent Server) can follow you around the world on any screen as it can be linked, on the chip level, through a SSL encrypted connection back to your HAS which could be at home or provided for by any govt or business that starts now. The citizen will primarily a simple interface of 5 levels per setting for every contact they interact with(the devices can know who’s around you because of their autoconfiguration and routing systems), an example of this is you are walking down the street and someone is coming up behind you, your HAS compares your Privacy levels against his HAS and you two have never met, so it logs it and either sends you a notification or not based on your notifications. We need one contact system that only keeps fundamental info about each person or business you interact with either virtually or in real life. The system will he fairly simple, we just have to hammer out the definitions of the 5 different levels (Trust, Public, Private, Security, Size, Publish). Size is more for the device than the user and should be sent, per device that the citizen uses. The middle, Level 3, is based around the idea of self,Security three means you trust unclassified contact enough to have a face to face or Trust level 3 for people that have been classified, while Security-1 is means you want only people you have classified up to Trust-1 and are invisible to the web. Simple slider based systems that can be turned to standard icons, for the Universal OS, so anyone can have an equivilant experence reguardless of the screen being used.
Great, what does this mean? What needs to be done now?
We can talk more about this, I am creating a proposal now in the form of written, audio, and video demonstrations of how the deconstructed future is only a few terms away. We are on the cusp, and all who wait to see where it is will have already fallen over. With enough general knowledge in technology standards, manufacturing costs and product development, it can be shown easily that in the next few years, every new tv will have a computer with wifi, ipv6, wimax connectivity and the wireless carriers will roll out 4G, an optical connection will be the standard for all data and power needs to mobile devices (let’s hope it’s a modified microUSB standard to comply with the EU moble connector standard), and other size screens, the ubiquitous connector is only a stopgap while the WiFi Alliance finish their WiFi Direct adjustments and Wireless charging (like Powermat) becomes more ubiquotious, and ip providers start offering IPv6 addresses, as it’s standard on DOCSIS 3.0, LTE, WiMax, and others. Can you see that in 2 years cheap devices with unique addresses will be able to create a pivate Link back to a HA router. What is next is that the people that link the Home Agent to the contact service will have the metrics to show the rise and fall of contacts in and out of importance to the citizen, ie they will control the value of a contact based on the amount of links the person has, not to others, but back to himself. This means that I could have a Trust-3 score of 200, which would be 200 people have me currently at a Trust-3 level and a Trust-5 score of millions, as there would be millions of people who know nothing about me. These levels, on the aggregate should show who is a creator and who is a consumer. Let’s hammer out the metrics.
Embrace the idea of future technology and help find out how we will get there.
Robert L Sawyer III
SawyerIII on Twitter
210-593-8723 Google Voice
Oh, were but that I had economic training, I’m sure I could come up with something.