The Next Stage in the Digital Economy: Nano-Transactions and Nano-Regulation

 

From the Economics of Property Rights to the Economics of Transactions

 

©Eli M. Noam

Columbia University

 

New Orleans

December 2000

 

This essay explores the future of the information environment. It will argue that we are taking an antiquated perspective to telecommunications; that we focus on property rights as the key to the future, instead of on the economics of transactions; and that have not developed a technological vision as a foundation for a policy vision. The essay will provide such a vision -- nano-transactions in information, the integration of information with money—their reasons, and their implications for digital business and policy.

 

Information used to be a scarce good. It is now an abundant good. From this, many consequences flow, technological, operational, business, and policy. For a long time information was controlled by the state. It was produced by state-affiliated institutions like monasteries and universities. It was stored in state libraries. It was disseminated by state schools. Later, it was distributed by state-run or regulated telecom and television networks. This phase of information control lasted until recently. Its underlying organizational logic was that information was scarce and hence valuable, and had to be produced, distributed, and shared under some public control. A body of theory provided the intellectual underpinnings, such as those of natural monopoly, public goods, industrial policy, and economic development planning, not to mention Marxist economics.

 

The state control model broke down for various reasons, a major one being the industrialization of information production that made the restricted state system inefficient. The production of information in most fields of science has been growing at an exponential rate of 6-8% per year. It has been said that 90% of all scientists who ever lived today. For entertainment type production, the growth has been even faster. In New York City, the growth of video programs available over cable TV has grown at a compound annual rate of 11% (Noam, 1998).

 

With information becoming plentiful, its production and distribution grew in volume and complexity. Rival providers offered alternatives. Users required more advanced services. The system became too complicated for any single organization, whether a school system, a monopoly telecom provider, a broadcaster, or a cable TV firm, to run and run well, and for government to supervise. This called for a different treatment.

 

The new model was one of openness, competition, and property rights.  The basic idea was that anybody could enter the information sector, markets would provide the control functions, and property rights the incentives (Posner, 1998). Potentially, that is true. But property rights are only the starting point, and not the only one, for efficient markets. Other institutions are also required, among them exchange mechanisms and payment that enable the transactions that make markets happen.

 

 

The property rights approach worked best in the information sector when it dealt with “old economy” physical assets, such as wire line networks competing with each other. Its adherents, having slayed the state model and become the reigning orthodoxy, could not envision a future beyond, once competition had been established. It was the end of history.  But of course, it never is. In the digital economy, we are not at an end of development; we are just at the beginning.

 

The property rights approach had only spotty explanatory power when it came to the new digital environment. Its thinking could not help with the most interesting new developments in communications, except in a labored way. Network externalities and communities do not fit into the property rights analysis, just as economics in general had a difficult time with externalities. The whole Internet must be a property rights advocate’s intellectual nightmare. Try to answer the question who owns a packet (not just the information contained in it) as it traverses the Internet. Given the trillions of such packets moving about at any instance, that would seem to be a fairly fundamental question.  But what is the answer? Whatever it may be, the system seems to work quite well without it.

 

The reason why the property rights approach has hit its limitations is that just as the state approach before, it, too, cannot deal with the new levels of complexity that the digital environment is rapidly creating. The key to thinking about the next stage of the information economy is not the sanctity of property but the rapidity of transactions.

 

This transaction approach to firms and the economy goes back to Oliver Williamson (1995), with some credit to Ronald Coase (1960). Williamson explained firms’ size and structure by their desire to minimize transaction cost. Firms’ hierarchical control of internal transactions reduced their costs below those of market coordination. This led to large firms.  Today, for information, external transaction and internal coordination costs are radically changing, and their relative magnitude should greatly affect the size and structure of firms and industries.

 

The key fact about information transactions is that they have increased in volume and applications, and in consequence there is so much of information moving around and being processed and stored and monitored and bought and sold, that it becomes too burdensome to control directly.  Network firms and service providers – not to mention individuals -- are increasingly unable to monitor and guide it effectively.  Initially, they delegated some control to IT machinery such as computer processors and software that is under their direct control. But that is not enough. The environment in which information exists and operates is becoming increasingly complex and decentralized.

 

Storage networks – Users share and optimize storage resources by linking hard drives and other storage devices. In network relations such as Napster, Gnutella, and Freenet, kids link their hard drives to share music. Do not think of them as pirates’ cooperatives, but rather as resource sharing outside or regular economic relations. And consider the fact that even if they wanted to be legal, there is no convenient payment mechanism for them to do so.

 

Processing Networks – People have similarly discovered that much of networks processing capacity is underutilized, and are creating collaborative processing. The first such applications were used to analyze data from outer space for signs of intelligent life, to crack encryption, or to analyze the products of large prime numbers. But the same principles of distant and decentralized processing are inevitable for commercial applications. This role is taken by applications service providers (ASPs) who offer encryption access to software programs, forwarding, protocol conversions, etc. In some cases, the hardware or transmission devices of one provider acts as a backup for the other. 

 

Web Interaction – Websites are beginning to talk to each other directly. Microsoft has been developing its net, which permits web interaction without human intervention of clicking etc.

 

Processor and sensor interaction – Transactions are moving from direct human control to those of delegated machinery. It’s not quite Arthur Clark’s HAL, but the day is near when our automobiles will be communicating directly with highways, our packages with shippers, our suitcases with airlines, and our light bulbs with utilities.

 

Transmission Interaction – The transmission networks connecting these devices utilize a wide variety of transmission media, provided by numerous network firms.  The information itself is a huge flow that includes entertainment, transactions, and interaction. It flows over wires and fibers, over the air, in space, wherever there is capacity at a good price, avoiding the various and shifting technical and economic bottlenecks.

 

Information Trade – At any moment there is a vast economic activity in buying and selling information itself. Some information is sought by the user, who would be willing to pay for it. Entertainment and financial information are examples. In other cases, the information makes demands on the user’s attention, such as for advertising, and needs to compensate users for their attention.

 

How can such a system function in operational and economic terms? Not by human control, except on the macro level; humans cannot handle such granularity of decisions. Not by giant firms dealing with each other to account for trillions of transactions. That is the past. Not by centralized machines. Not even by networked machines talking to each other. Too much of transmission and processing would be used up by each piece of information having to be controlled from the distance, report back, receive instructions, account for itself, etc.

 

50 years ago, Friedrich von Hayek, in The Road to Serfdom (1944), the classic text of classic liberal economics, argued for laissez faire markets because of the great complexity of the economy. He saw the decentralized markets as huge information processing machines, which could deal with the myriad of transactions of society in a way that centralized decision-makers such as governments or monopolies could not.

 

The same insight applies to information. Information processes are becoming too complex and varied to be run in any other way as through decentralized decision processes. At an earlier stage, this was accomplished by pushing the decision mechanisms down the hierarchy, from the state level to that of various carriers and service providers. But now, with the increasing complexity of the environment, it becomes necessary to push them down again, down to the level of the information itself. Information needs to be engaged in direct transactions that involve it.

 

 What does this mean, concretely, and how could it be accomplished?

 

The key here is to understand that information and its transmission networks are moving from continuous streams of analog or digital signals, to discrete continuous packets transmitted discontinuously by packet-switched networks.  These packets are identifiable by information in the “header ".  These so-called "overhead” bits surround the actual information (the "payload") and permit the packet to be routed to an address. The packets are labeled by sender, by location within a document, and by hop limits. If a transmission path is congested (or destroyed: this possibility was one factor for the Pentagon to sponsor packet-switched communications), the packet finds alternative and possibly circuitous ways to the destination (Baran, 1995).

 

This principle of identifiable information being surrounded by operational data, just like the contents of a letter that are surrounded by an envelope, is an enormously powerful concept, and there is no reason why it should not be taken much further than it is today.  Though its origin was the desire to optimize scarce transmission capacity, it can do much beyond that, as the transmission becomes plentiful. Already, the next version of the Internet protocol, IPv6, includes data for authentication and integrity to prevent various forms of mischief such as “IP spoofing” and “host masquerading.”  It also provides for the assignment of priority. The sender can choose fifteen levels of priority.  But without an economic dimension to the priority (Mackie-Mason and Varian, 1994), the system will predictably fail.

 

The problem is that we have divorced information from payment. We are throwing information into the big traffic and business system called the network, the Internet, but naked, without money to pay along the way or ability to make decisions beyond the routing. It’s like sending a kid into the New York subway system without any money. Instead, we have been charging in very indirect ways, whether for transmission or for storage or access. We charge for some physical measure, such as transmission channel bandwidth or for storage capacity. But we do not take the next step, which is to link the information flows directly to transaction and payment mechanisms. The way that automobile drivers, for example, move around across highways and bridges, and pay for passing various gates. Or buy gasoline. Or park on the street.

 

Electronic Postage Stamp – The first conceptual and technical step would be to add payment tokens to the packet, just like postage stamps on letters. How would one do this for information? One way would be by adding pre-paid tokens to the information. Tokens like access codes. Without the access code token information would be blocked from traveling over network segments, or be stored at servers, etc. These tokens would be, in effect, electronic coins. They would be sold to the users by a variety of institutions, and also redeemed by them for credit or money. The users would add the tokens to the packets, and they would be stripped from the packets by the various transmission and storage providers along the way. This can be distinguished from earlier proposals along this line by the author for the common and unlicensed use of spectrum (Noam 1995), and by Hardy and Tribble (1995) for a counting device system (an “accumulator”), to count passing packets, read this counter and settle by producing real money. 

 

Multipurpose tokens with packet controllers – But why stop at this first step described above, with electronic postage stamps?  Suppose we provide packets with an electronic wallet of multiple tokens usable for multiple purposes and with some intelligence and instructions on how these tokens can be used.  If priority of service is required, for example, payment might be authorized that could be higher than otherwise.    Transmission networks, servers, storage devices and processors require toll payment from packets using their facilities.  If the price charged would be too high, for example due to congestion, the information would refuse to pay and turn around to look for a cheaper path or wait for a lower off peak price. Or they would offer the transmission provider a lower price, and engage in negotiation. Once a transaction is accomplished, tokens would be stripped off the packet and transferred to the other party to the transaction.  Once its mission is accomplished, the envelope would return to the sender with any leftover tokens, or store them in other ways as credits. 

 

 

Thus, packets would seek the economically most efficient operation subject to their mission definition.  They would almost instantaneously optimize the utilization of existing facilities.  And they would do so in a decentralized fashion, without continuous control and guidance by their owners.  In this scenario, information would travel with its own money, and engage in nano-transactions. Or, just as much, money would travel with its own information (that is, with its own software instructions). We would integrate information with money. We will have “smart” money and “rich” information.

 

Access to Information – But why stop here, with operational transactions?   Suppose, for example, that one sends out a software agent, a "bot", to search for certain data.  At present, the economic models to obtain information are those of (a) subscription, (b) advertising, or community creation to facilitate (a) or (b). Subscriptions, however, require an advance relationship, purchase decision, and payment—all of which have met resistance. Advertising, on its part, makes sense only when humans view it, not software packets (though one could imagine promotions aimed at machines, too). Now, armed with its own wallet, the bot software could pay directly for the information it collects, and for admission to others’ databases. At the same time, it would also be possible for sites, or for interested third parties, to pay for visits, in order to generate traffic.  And in many cases, information would be offered for free.

 

Selling Information - And why stop here, either?  So far, the packets we described were on the buy-side.  They obtain various services from other parties and pay for them. But suppose that they are on the sell-side. A music publisher sends out, “push”-style, the latest tracks, at prices that could vary by time, date and user. It pays tokens for transmission and collects tokens for the use of this music. If the user then re-transmits the music to third parties, the packets would be would be programmed to require additional tokens for their re-usage. 

 

In other cases, the recipients of unsolicited information could be paid for the demand on their attention, based on how much of compensation would be needed to satisfy them. You want to bother me, fine, as long as you pay me. Tiny slices of time would be compensated in tiny amounts. This would also take care of many privacy problems. If people want to intrude let them pay the admissions fee I set. If people want information about my online shopping habits, fine, as long they must compensate me at the price I set.

 

Such a system creates an integrated and continuous economic system of nano-transactions.  It could be used for access, for transmission, for storage, for processing, and for the information itself. It could be used to gain priority in service.  It could maintain flat rate pricing but would not need to. And it would decentralize the numerous transactions that information and its users and suppliers encounter.

 

The usefulness of micro-transactions has been well understood and discussed (Kytpoki, 2000). A variety of approaches for electronic cash have been introduced, such as Cybercash with its CyberCoin; Netbill; Digicash; Mondex; and others. All aimed at large transactions, were kludgy to use, overprotective against misuse, and suffered from an absence of a critical mass of users (Muller and Schmidt, 1997, Riessner, 2000). A second generation of electronic money includes DEC/Compaq’s MilliCent, MicroMint by the noted cryptology pioneers Revest and Shamir, IBM’s MiniPay, and HP’s system by Wenbo. (Chi, 1997)

 

These micro-transaction systems, while steps in the right direction, operate on the level of the entire transaction, and usually involve humans. In contrast, nano-transactions operate on the packet level, and are conducted by the packets themselves. (In practice, the line between the two might not always be bright.)

 

There are some technical problems associated with this nano-transaction system, in particular, the length of the packet and its security.

 

The size of the packet overhead will have to grow if it is to include money tokens, search bots, etc.  It might be argued that this would be inefficient because it would reduce the available payload. It’s as if one adds 5 pilots to a plane, with less space left for passengers. That’s true, unless one builds a bigger airplane. The packets will have to be much larger than at present. And why not? The present packet size is based on scarcity of transmission. Packet protocols aimed to squeeze as much information as possible into transmission paths, the latter being scarce and dear, and this could be best accomplished by keeping the packets short. This is not the primary concern in the future. Transmission will not be in short supply; the focus is not of maximizing transmission but in minimizing transaction costs. (Another approach, that of ATM, aimed to standardize packets into short cells of 53 bytes. That approach is being squeezed in the technology market place by the less constraining Gigabit Ethernet.) The Internet’s new IPv6 specifications can handle packets expanded to a Jumbo Payload option of 65,535 bytes (about 60 pages of text), though with more limited capabilities. There is no reason why the packet size will not grow further, such as to a few minutes of video. True, present switching and routing is based on brief storage of the packets, and are designed for such short lengths. Engineers will handle this issue, as well as the associated issues of latency, scalability, interoperability, and robustness. As a result, switching costs are likely to rise, while transaction and coordination costs will decline. Processing requirements will rise, but the utilization of a given processing capacity will be more efficient.

 

Furthermore, intermediate arrangements are likely. Not every packet would need to transact on its own. There could be “pilot packets” which would set up the transactions for subsequent and identified packets, much in the way that locomotive routes and moves subsequent boxcars. In transmission, this would resemble present “virtual circuits”, and it would carry this concept to non-transmission uses as well.

 

Another technical question is how to maintain the integrity of the overhead information against tampering, stealing, copying, and counterfeiting. One would need elements of encryption and authentication. That is already done in the IPv6 protocol. One needs to differentiate between several elements of a package. The internal control instructions need to be encrypted and inaccessible to the partner in a transaction. The information itself may or may not be encrypted, just as with payload today. The money tokens would be encrypted to protect against tampering, but would not need to be decrypted by parties to the transaction, as long as they can pass them on, just as in the case of regular cash, which is hard to forge but easy to transfer.

 

True, any encryption can be defeated. But keep in mind that we are talking here about nano transactions, and stealing fractions of pennies may not be worth the trouble. Most likely, intermediate institutions for authentication would emerge. They could also provide escrow services to assure that a service that is being paid for is indeed performed. Technical fixes are possible, such as token value being activated only after a certain period, to prevent fly-by-night tampering. On top of that, one could envision the networks patrolled by packets and software bots whose function it is to check on what’s going on and ferret out industrial-style forgeries. It would be interesting to analyze the legal and civil liberties implications. But that is a topic for another day.

 

One implication of nano-transactions and identifiable packets is that information can be treated in a highly differentiated fashion. Counter to the claim of information becoming a “commodity”—a staple of the left-- or that it is technologically –“undifferentiated”— part of the techno-mantra that “a bit is a bit”(Negroponte, 1995) -- it is actually quite the opposite. Each packet has an address and sender and a recipient, and soon much more. Which means that packets can be treated quite differently. Thus, partners in a transaction can charge different prices to some senders, e.g. dot coms, who might be viewed as less price elastic than dot orgs.  Bill Gates might be charged higher prices for watching Internet TV than I am. Or, more likely, lower ones, since his credit worthiness was better.

 

Different usages might be priced differently. For example, a video entertainment packet would receive big volume, quality, and priority discounts relative to a real-time voice call packet. This would help resolve the quandary of uniform bit pricing, whether to make voice calls free, or charge for a movie at an unattainable price (Mackie-Mason and Varian 1994). The point is, identification makes possible price discrimination, which has positive as well as negative aspects.

 

The flexibility in pricing and transactions also means that the existing flat rate pricing, or zero pricing, arrangements for certain services could be maintained. Nothing requires a pay-by-the-packet, priority priding, etc. The system can accommodate any form of transactions and pricing, but it does not exclude any for reasons of technical impracticality. If a backbone provider wants to charge some traffic passing its node on a zero (bill-and-keep), peer basis, it can do so, while at the same time charging other traffic from non-peers a different price. Other payment modalities are possible, for example barter, in which storage gets traded for processing, or for music. (“I got it for a song”). Money, of course, is a much finer instrument than barter.

 

Nano-regulation – The ability to identify also has significant implications for the future policy and regulation. Many of the advocates of competition sincerely believe that by creating a competitive network structure, behavior and performance follow, and regulation becomes unnecessary. But they have no vision of the future beyond. Others, mostly the libertarian majority of Internet enthusiasts, believe regulation impossible due to the ability of any 14-year-old cyber-punk to run electronic circles around flat-footed government regulators.

 

For better or worse, both these perspectives are quite wrong. The future is more complex than it used to be. Identification of packets opens entirety new avenues for government regulation on the level of nano-transactions. We can call this nano-regulation.  And the presence of new tools, on top of new (as well as old) problems, will be irresistible.

 

Legacy Regulation – Notice that under a system of self-transacting packets, many wholesale arrangements among carriers and other macro-providers become much less important, and hence much less requiring regulatory intervention. For example, interconnection charges, in which carriers or service providers pay other carriers for their services, would be replaced by the actual information paying for transit services. This is the same as today for highways. The New Jersey Turnpike does not pay the New York Thruway for interconnection; cars moving from one highway to the next do the paying. As long as there is price competition among transmission carriers, the information will find the optimal routing, and transit pricing will be economically efficient.

 

Similarly, the concept of common carriage, under which carriers must equally serve all customers, becomes meaningless when information transacts directly for transmission.

 

But just because some issues of the traditional regulatory agenda will disappear does not mean that regulation as such will wither away. Let’s look at some of the potential regulatory uses of the potential to identify packets:

 

Redistribution – Regulated price discrimination in favor of certain uses or users, such as education of rural users, and a corresponding disfavoring of certain uses and users, such as games. Subsidy mechanisms, in which some users would receive a stack of tokens for free or at reduced prices. Most likely, they could use these tokens only for certain meritorious applications, such as education, rather than to obtain information on sex, horse racing, or financial derivatives.

 

Preferential Treatment – Differential treatment of packets provided by some types of firms, such as broadcasters, relative to the treatment of packets from others. Thus, a version of today’s “must carry” rules of TV signals over cable is quite possible in technical terms.

 

Privacy – Privacy issues will emerge insofar as packets, their senders and recipients can be identified. This could lead to private and public snooping and interception In consequence, various privacy and access protections are required, and wiretapping laws might need modification.

 

Content regulation becomes possible – For example, there could be a requirement those packets containing materials harmful to minors would have to be labeled, so that blocking becomes possible, by parents or by other institutions.

 

 

Trade – In some countries, political labeling, depending on the sender, would be mandated.

 

Other countries might require a labeling of the origin of the information, and use it to favor domestically produced information, such as in the case of movies.  Poor countries might find the system draining their resources, and could set tariffs and protectionist measures. Conversely, they could become the beneficiaries of price differentiation by richer countries.

 

Arbitrage – The objects of such regulatory actions will be to undermine it by loading packets of one category onto packets of another category.  In consequence one essential regulation would be to require protocols that identify such piggybacking.

 

Thus, the notion that information and its transmission will be free and unregulated is wishful thinking. Information will be identifiable, and therefore targetable, and therefore regulable. The notion that “ you cannot regulate the Internet” will turn out to be nonsense. To the contrary, one can regulate packetized information and its conveyance much more effectively than undifferentiated waves and bits. The question is legal, constitutional, and political, not technical or practical. (Of course, there will be resourceful people who keep a step ahead of enforcement capabilities; but that does not negate the basic point).

 

MacroEconomics of Nano-Transactions – Other important questions emerge. Who can coin those nano tokens? Isn’t it creating money? Yes. The tokens would have to be redeemable in other value, i.e. money. They could be backed up by governments, but more likely by a variety of intermediate financial institutions, like banks, credit card companies, and various brokers, some of whom could also be short-term lenders to replenish tokens to a customer who runs short.  Playing such a role would also be natural for present phone companies as an extension of their phone bills. (The Click AT&T service, introduced in 2000, already permits web transactions over 25 cents to be charged in that way.) There would emerge private moneys, of global acceptability and convertibility, like American Express today. Government would not be out of the picture, not as issuer of money tokens but as regulator. But it would have a more difficult time maintaining monetary policy by controlling the money supply, when electronic coins, issued around the world, abound. A related -- and unpopular -- question is where the power to tax these transactions would reside, and how to go about the taxing.  One way would be a surcharge on access tokens. Another important question is the stability of the entire system, both in its macro-economic and its transaction aspects.  Could it be subject to over-compensation, oscillations and gridlock congestion that would prevent equilibrium to be reached?

 

Conclusion

The next level of convergence, that much-anticipated merging of various electronic delivery systems will bring together information and money – which, after all, is a form of information. (Next, we can anticipate the mind-blowing convergence of information technology, money and microbiology.).

 

The convergence of information and money and transactions will not occur tomorrow, but soon thereafter. Nor will it replace more traditional arrangements. Hybrid arrangements are quite likely, through various intermediary-wholesaling institutions like the emerging bandwidth capacity brokers whose emergence the author predicted a decade ago (Noam, 1991). Similarly, some transfer arrangements could be on the wholesale level, by various resource providers charging each other without involving their customer directly. But I have no doubt that information will eventually travel with its own money, that money and information will become integrated, that it will pay along the way for transport, storage, and especially access. That in other cases it will be compensated for access. And that the invisible hand will be furthered by invisible packets. But getting there requires technology that is not quite here yet. We need to create those tokens, account for them, protect them from forgery, create transaction nodes, and add the inevitable legal and regulatory layers.

 

One of the many clichés around the Internet is that “ information wants to be free”. Whatever that means, I do not think it’s true. What information really wants is to have its own wallet and credit card, and to go out on its own. Information does not want to be free. It wants to be emancipated.

W:Eli\Elipaper\Nano\nopnano13

 

Acknowledgements

Helpful comments for their comments:

Robert Atkinson, Kenneth Carter, John Kasdan, Yushiko Kurisaki, Michael Noll and

Hal Varian are gratefully acknowledged

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Bibliography

           

            Alchian, Armen A. “Some Economics of Property Rights.” Economic Forces at Work.” Il Polanco Vol. 30. Indianapolis Liberty Fund. 1977. pp. 816-829.

           

            Baran, Paul, “Is the UHF Frequency Shortage a self Made Problem?,” Unpublished paper, presented at the Marconi Centennial Symposium, Bologna, Italy, 1995.

Braun, Hans-Werner, Kimberly C. Claffy, and George C. Polyzos, “A Framework for Flow-based Accounting on the Internet,” Proc. Singapore International Conference on Networks (SICON'93), Singapore, September 1993. ftp://ftp.sdsc.edu/pub/sdsc/anr/papers/accting.sg.ps.

 

Chi, Ellis. Evaluation of Micropayment Schemes. http://www.hpl.hp.com/techreports/97/HPL-97-14.html. January 1997.

 

Coase, Ronald. “The Problem of Social Cost.” Journal of Law and Economics, 1960

           

Crocker, Steve, The Siren Song of Internet Micropayments. http://www.cisp.org/imp/april_99/04_99crocker.htm. April 1999.

 

Demsetz, Harold. “Toward a Theory of Property Rights.” American Economic Review 47.1967. pp. 347-358

 

Dukach, Semyon. “SNPP: A Simple Network Payment Protocol.” Proceedings of the Eighth Annual Computer Security Applications Conference ftp:snpp-paper.ps/snpp-paper.ps. December 1992.

 

Dyson, Esther. A Design for Living in the Digital Age. New York: Broadway Books. 1998

 

Ferguson, Niels, “Single Term Off-Line Coins,” Proceedings of EuroCrypt'93, ftp.cwi.nl/pub/CWIreports/AA/CS-R9318.ps.Z, 1993

 

Graff, Jamison. An Introduction to the Work of Oliver Eaton Williamson. http://users.iems.nwu.edu. June 1995.

 

Gunning, Patrick J. “The Property System, Public Goods and Market Failure.The Failure of the New Subjectivist Revolution: A Critical Essay on the Economic Theory of Ludwig von Mises's, Nomad Press, 1994. See also http://www.fortunecity.com/meltingpot/barclay/212/demo/workpape/pbchpbgd.htm.

 

Hardy, Norman and Tribble, Eric DeanThe Digital Silk Road.” Tulloh, William, Mark Miller, and Don Lavoie ed. Agoric Systems: Market Based Computation. ftp://ftp.webcom.com/pub1/agorics/postscript/DSR.ps, September 1995.

 

Hayek, Friedrich A. Von.  The Road to Serfdom. University of Chicago Press. 1944.

 

Hirsch, Werner Z. “Law and Economics: An Introductory Analysis.” San Diego: Academic Press. 1999

 

Kytpoki, Jari and Kapijoki. Vesa. Micropayments - Requirements and Solutions. http://www.hut.fi/~vkarpijo/netsec99/, January 2000.

 

            MacKie-Mason,Jeffrey K. and Hal R. Varian, "Some FAQs about Usage-Based Pricing," Computer Networks and ISDN Systems, Vol. 28, 1995. pp. 257-65. Also in The Journal of Electronic Publishing. See also http://www.press.umich.edu/jep/works/mackiemason.usage.html.

.

MacKie-Mason, Jeffrey K. and Hal R. Varian, "Pricing Congestible Resources," IEEE Journal of Selected Areas in Communications, Vol. 13, no. 7, September 1995. pp. 1141-49.

 

Schmidt Carsten and Rudolf Müller, A Framework for Micropayment Evaluation,

Netnomics 1 No. 2, 1999. pp.187-200, See also http://ns.baltzer.nl/netnomics/netnomics.asp

           

Negroponte, Nicholas. Being Digital. Knopf, New York, 1995.

           

Noll, Michael A. “Internet Pricing vs. Reality.” Communications of the ACM. Vol. 40, No.8. August 1997. pp. 118-121.

 

Noam, Eli. “Beyond Liberalization: From the Network of Networks to the System of Systems," 1991. Also in Telecommunications Policy, Vol. 18, No.4, 1994. pp. 286-294. Working Papers, CIT, Columbia University

 

Noam, Eli. Taking the Next Step Beyond Spectrum Auctions: Open Spectrum Access.” IEEE Communications Magazine, December 1995, pp. 66-73.

 

Noam, Eli. “Spectrum Auctions: Yesterdays Heresy, Today’s Orthodoxy, Tomorrow’s Anachronism.” Journal of Law and Economics, December 1998. pp. 765-790.

 

            Posner, Richard A. Economic Analysis of Law. Aspen Publishers, 5th ed. January 1998

 

Riessner, David, Micropayments-The enabling tool for doing (micro-) commerce in the “Internet Economy.” www.ines.fr/internet/contrribution/riessner_david.htm, spring 2000.

 

Umbeck, John “Might Makes Rights: A Theory of the Formation and Initial Distribution of Property Rights, Economic Inquiry” Vol. 19. 1981, pp. 38-39.