New Economy

Created: 2000-04-26
Last Modified: 2000-5-04
Go to
Brad De Long's Home Page

Teaching | Writing | Career | Politics | Book Reviews | Information Economy | Economists | Multimedia | Students | Fine Print | Other | My Jobs

Brad DeLong's mailing list | Comments on this page | Questions and Answers

New Economy Forum Briefing

J. Bradford DeLong

May 2000


The Current Macroeconomic Boom

Micro and Macro Studies Agree: Computer Investments Are Driving Our Current Economic Boom

Three Factors That Make Our Age the Age of the Computer Revolution

Consequences of the Computer Revolution for the Business Cycle

The New Economy and Antitrust I: Microsoft

The New Economy and Antitrust II: Principles

Intellectual Property

Public Funding

Standards and Markets

Open Source

The Technology Tsunami

Moore's Law

The Relative Speed of the Information-Processing Revolution

The Price of Computers Has Fallen More than Ten Thousand-Fold in a Single Generation

Volume of Semiconductor Production

Productivity in Information Technology in the 1990s


The Build-Out of the Internet

Standard Modem Technology Proved Capable of Extraordinary Improvement

The U.S. in 1997 Had One Internet Computer for Every Twenty-Three People

More than 60 Million Computers Are Now on the Internet


The Internet and American Economic Growth

The Direct Contribution of Computers to GDP Has Become Large

The Size of the Internet Economy (as Seen From U. Texas)

Size of the Internet Economy

Trade in Information Technology



Federal R&D Spending


The Need for Inclusion

The American Distribution of Income Is Now More Unequal

Combined E-conomy Sidebars...

Laura D'A. Tyson


This is meant to be an annual event.

Forum John Doerr's idea...

No money, no deals; dialogue


Walter Shorenstein

Fifteen years ago Walter established the Joan Shorenstein center at Harvard's Kennedy School--media governance in the twenty-first century.

Profound technological revolution now underway...



National finance Chair of the Democratic Party: Joel Hyatt; idealab


Roger Ferguson

Recognize people.... Janet.... Alan

Views are his own...

Extraordinary macroeconomic performance: better than even the most favorable forecasters would have imagined...

Long run benefits of deregulation, a movement started under Ford and accelerated under Carter...

Cyclical factors have played some role...

More important: capital spending boom and new technologies

Federal Reserve tries to keep output growing at potential; needs to know what potential is...

Better information within firms?

Oliner and Sichel...

Ferguson asks eight questions:

  • Can we count on such rapid productivity growth in the future?
  • Is the computer really unique? How about the telephone--the transatlantic cable?
  • Technology other than the internet: what's so special?
  • Continuous improvement or once-and-for-all upward jump?
  • What are manufacturers doing in the way of using computers?
  • What new technologies are in the pipeline?
  • How are .coms valued? How should they be valued?
  • What will happen to venture capital in an environment of equity-price uncertainty?

Job of the Federal Reserve...



Hal Varian: are LANs the key innovation behind the productivity boom?...

Cisco story and the mfg. supply train--95% of their orders now coming in online...

Moore's law until 2012

VC an old phenomenon

Internet as about the consumption of media--scarce channels. Channels are now free and abundant...

Laura Tyson: look at Morgan Stanley research...


Joel Klein

Rare to be in a room with so many friends, especially when no one wants to acknowledge that fact... but I know who you are and I thank you...

"Plus ca change, plus c'est meme chose"

Core principles of antitrust reflected in the Sherman Act should not and will not change

Core principles have served us well...

Two caveats: technology effects, network effects

Traditional focus on price will be reoriented

Core principles: market competition

But markets won't always remain free and competitive without appropriate government intervention

"Skill, foresight, and industry" as the permissable ways of acquiring market power.

Big is not bad. Success as the result of skill, foresight, and industry is to be applauded

Quotes from _WSJ_ on AT&T case...

Microsoft: the reason the browser was so important was that it was bound to be ubiquitous...

Office as the second most ubiquitous product

Natural monopolies...

We have been won over to the view that in the absence of illegal activity, natural monopolies are rare and transitory...

Network externalities, positive feedback, nothing-succeeds-like-success

Network externalities... (mandate interconnection?)

Economies of scale...

No reason to attack companies that benefit from positive feedback rules

The future of antitrust disciplines not the creation of monopoly, but the use of market power to discouraged innovation and crush competitors.

"There are just so many ways to illegally hurt your competitor..."

Cutting off competitors' access through intimidation or exclusionary contracts; tying; predation to rise rivals' costs--these are the standard ways of misusing market power.

"Tying" vs. "integration."

How would a small software company compete on products Microsoft plans to fold into ints operating system? Steve Ballmer's view:

fight a losing battle...

sell out to Microsoft...

not going into business to begin with...

"Predatory" cut off air supply--cashectomy performed on Borland...

In antitrust, the new new thing ain't so new after all...

"Plus ca change, plus c'est meme chose"

Re: Time-Warner/ABC--the key problem is the fact that there is only one pipe to the house. Hope for a technological fix...

Microsoft's attempts at predation had little to do with its eventual victory in the browser or the Java wars, which was due not to predation but to "skill, foresight, and industry"--that Explorer 4 was a better piece of software than Netscape 4, that Explorer 5... well, there never was a Netscape 5... Java had the problem of being an order of magnitude further away from the bare metal than C... Only Oracle has managed to produce

I don't want to defend Microsoft...

The possibility of Microsoft's version of Kerberos working with NT only, the fact that Microsoft FrontPage server extensions are extremely difficult to make work under Apache, the fact that I am humiliated every week by my continued inability to keep Explorer from launching at inappropriate moments.... All these make me fear Microsoft...

But this discussion seems a little too friendly: so let me try to channel the spirit of some executive two states to the north...

You said that Antitrust Division policy in the new economy was unchanged: holders of monopoly power should tread carefully--take care not to take any steps to abuse their market power. For then they will be broken up, for the social losses in terms of reduced economies of scale are small, and the benefits to the consumer in terms of increased competition are large.

But in the new economy we have not just the old economies of scale that gave you 20% lower unit costs at twice the volume, we have information goods-based economies of scale where twice the volume gives you 50% lower unit costs. We have economies of standardization and interconnection as well. There is reason to fear that break-ups will sacrifice more in the ways of economy of scale. There is reason to fear that the benefits to the consumer in terms of increased competition will be transitory--for the same economy-of-scale forces that led to the initial near-monopoly will swing into action once again.

Shouldn't the balance of antitrust change if the relative costs and benefits of breakup have changed?

In the old days, the response to a true "natural monopoly" was regulation. We no longer trust regulation. So what do we do when it seems as if a much larger chunk of the new economy than the old is made up of natural monopolies, disciplined not by other competitors but by their own installed base of previous versions and by possible piracy?

Or is the role of the Antitrust Division that of King Canute, commanding and winning only temporary victories against the tide of information-age economies of scale?

A question: why didn't Bill Joy or Bob Frankston build a spreadsheet for Berkeley Unix? We economists tend to say that information goods are public goods--should be distributed essentially for free, and produced by organizations like universities. I think that it says something about the defects of that organization model and the advantages of moving R&D into the private sector that Visicalc emerged not as part of freely-distributed BSD 4.3 for IT division-managed minicomputers but as a for-profit program for individually-owned microcomputers.


John Doerr, Roger Altman, Roger Ferguson, Alan Blinder...

A brief thought on "leveraging the top line" the business impact of the telephone...

If you look at pictures of the original American investment bank that took companies public--the original investment bank of J.P. Morgan, whose IPOs included such high-tech firms for their day as U.S. Steel, International Harvester, General Electric, Pullman, and the catastrophic IMM shipping company whose most famous ship was the Titanic--you see everyone, the partners and their assistants, sitting behind their desks, in one big room.

They had no telephones. So they had to stay in one big room in order to talk to each other. They could not keep each other up-to-date and up-to-speed without being literally cheek-by-jowl day after day. And they had to spend a lot of time in the office listening just to realize what the firm was doing.

This meant that the span of control of the original J.P. Morgan was very, very short: couldn't expand beyond forty professionals, lost its edge when it tried to monitor and assess more than eighty companies, couldn't do due diligence in any reasonable way, ...


Even after the end of the current cycle, even if the current cycle ends in a full-blown stock market crash, much of research and development will still be handled in a new way: the business of commercializing new technologies will still be rapidly spun-off into new companies, engineers will still be motivated by stock options, venture capitalists will still trawl for good ideas, and new technologies will still be launched into the marketplace in a small fraction of the time that product development cycles used to take. As Venture Law Group principal Tae Hea Nahm has argued, the organizational capabilities and modes of thought that have been created will still be vibrant and useful even if Wall Street ceases to be so eager to snap up IPOs.

We often tend to forget how different this new business ecology is from the way things used to be. But the executives and innovators who grew up under the old system remember. Michael Hiltzik's book _Dealers of Lightning_ reports on Adobe Systems' cofounders John Warnock's and Charles Geschke's days at Xerox's Palo Alto Research Center. As Warnock remembers it: "Chuck Geschke and I had a conversation in his office and said, 'You know, we need to go do something else, because we've spent two years of our lives trying to sell this thing [internally] and they're going to put it under a black shroud for another five.'" But at the time the bureaucratic system of Xerox had its rationality. With long product development cycles, little in the way of channels for testing new products, and high costs of ramping up to full-scale production, even a Xerox could not afford to introduce many new technologies. So it made sense to pit ideas against each other internally in a bureaucratic tournament, hoping that the fittest would survive the gauntlet.

Moreover, at the time the bureaucratic system at Xerox had--and this was not clear to me until I read Clayton Christensen's brilliant _The Innovation Dilemma_--a necessary irrationality as well. A strong, competent, successful organization It was also a place where the launch of a disruptive technology was likely to be hobbled. Such a technology means that old divisions shrink and old customers are told to buy new products. But the more competent are the engineers and managers running the old divisions, the greater is the number of reasons--good reasons--they will think of that the new technology won't work. The more the company listens to its customers, the less will it be open to engineers proposing radical shifts. The things that had made Xerox successful made it hard for it to adopt disruptive technologies.

Warnock's and Geschke's company, Adobe Systems, came in the early days of our new business ecology. They were in perhaps the second generation of development of what we now see as the Silicon Valley System: rapid prototyping, short product-development cycles, early test marketing, options-based compensation, venture funding, early corporate independence, and so forth. It seems to be a highly successful system for launching new technologies into the marketplace.


Diminishing Returns?

Capital Spending Boom?

Magnitude of Inventory/Sales Ratio Decline?

How Well Do We Measure Output?

Who Pays for the Internet Boom?


Larry Summers:

Five points on the new economy:

  1. In the information age, big markets are the best markets
  2. In the information age, our best resource is our people
    1. Resources for education
  3. Needed: a proper system for innovation
    1. As important as IP protection is...
    2. As important as VCs are...
    3. ...knowledge is a sequential process of discovery
    4. can't build new knowledge on top of old if old knowledge is closely held and guarded
    5. ...hence the critical importance of the public role in research and development
  4. Markets need to be adequately regulated so that they serve customers
    1. Competition policy
    2. New technologies bring new challenges
      1. Privacy
        1. I want people to know I'm a tennis player so they can try to sell me tennis stuff
        2. I don't want people to construct a detailed personality profile of me based on all the checks I've written over the past decade
    3. Need to make sure there's no "race to the bottom" in the use of new technology
  5. Include everyone
    1. No excuse for not including everyone today, when in this economy we don't have people looking for jobs, we have jobs looking for people.
    2. Inclusion not just necessary for the nation, but good for the nation's businesses
    3. 25% of Black American families don't have bank accounts, and spend $15,000 over their lifetimes in check-cashing fees

Sign up for Brad Delong's (general) mailing list

Add a comment on this page...

There seems a widespread agreement among economists like yourself both that the internet (and related) booms resulted from open standards, and that, in the larger scheme of things, the twentieth century has grown wealthy as a result of science. The common feature of both of these is publishing of knowledge which can then be critiqued by others.

As a corollary to this, the computer software industry is largely driven by trade secrets. While economists have criticized Microsoft on the grounds that it has abused its position as a monopoly, and while many zealots have simply claimed that Linux is the future, I have seen no reputable academic studies looking at this issue. I contend that rather than yet another study arguing whether computers are or are not driving the US economy, what is really needed is a study discussing the extent to which closed software (with all the attendent evils---lack of peer review, lack of interoperability, low standards of quality compared to other engineering disciplines, problems when the person who wrote the code leaves the company etc) are consequences of - an immature industry that is changing constantly or - inevitable given the nature of software or - a consequence of the fact that the software industry is governed by trade secrets rather than open exchange of information, implying that a change in the legal environment might drastically improve software quality (and thus overall GDP growth) although a consequence would, of course, be the loss of monopoly rents by Microsoft and friends.

Contributed from by Maynard Handley ( on July 25, 2000.

Well done, a lot of food for thoughts

Thank you, Hans Kaufmann, Member of Swiss Parliament

Contributed from by Hans Kaufmann ( on May 12, 2000.

Go to related links...

Professor of Economics J. Bradford DeLong, 601 Evans Hall, #3880
University of California at Berkeley
Berkeley, CA 94720-3880
(510) 643-4027 phone (510) 642-6615 fax

This document:

Search This Website