Kennedy School Conference: American Economic Policy in the 1990s: Presentations and Comments

J. Bradford DeLong

June 2001

Another copy, without the comments...

International Economic Policy: Kennedy Forum

Back when I started out as an economist there were several years during which it seemed that most of the articles I wanted to write had, I discovered, already been written in the previous decade by Barry Eichengreen. He reports that when he started out he found himself subject to the same phenomenon--only with respect to Charlie Kindleberger. So I spent part of last weekend rereading Kindleberger's (1978) _Manias, Panics, and Crashes_, looking for places where Kindleberger had already said what I think before I thought it, and had expressed it better than I can.

I found a number of such places: Kindleberger's declaration that in the last analysis the making of international economic policy under such circumstances "is an art" and that that "says nothing--and everything." And there was Kindleberger's summary that the rescuer of the system, the "lender of last resort", "should exist... but his presence should be doubted.... This is a neat trick: always come to the rescue in order to prevent needless deflation, but always leave it uncertain whether rescue will arrive in time or at all, so as to instill caution in other speculators, banks, cities, or countries.... some sleight of hand, some trick with mirrors... because monetarist fundamentalism has such unhappy consequences for the economic system" when expectations converge on the "unfavorable" equilibrium.

Kindleberger's declaration that he does not wish to "contravert" the claim that the "presence of a lender of last resort weakens the self-reliance of the banking system and increases its likelihood of falling into excesses of overtrading, revulsion, and discredit," even though this argument "has overtones... that there is no use providing the poor with housing since they will only keep coal in the bathtub" and the possibility that the known existence of a lender of last resort causes expectations to converge on the favorable equilibrium: that it does not "increase speculation and overtrading" but "calms anxieties when overtrading occurs."

That finding the point of balance for all these conflicting issues and concerns is very difficult, and keeping the point of balance is almost impossible, is very clear in Kindleberger's "on the one hand... on the other hand" argument. That finding the point of balance is very difficult was also my thought on listening to Allan Meltzer this afternoon. This afternoon we have heard the Meltzer Commission Meltzer: the Meltzer who fears the growth of moral hazard, who thinks that large-scale lenders of last resort create by their very existence the crises they then are forced to handle, the one who believes that a lean IMF is a good IMF and that it is important that speculators fear mightily that they might get burned.

But if you read your _Financial Times_ last May 10, you would have heard a different Allan Meltzer, one who seeks a much larger and stronger IMF with enhanced powers. He proposed that the IMF commit in a crisis to buy any and all government bonds that private investors wish to sell, albeit at a discount to the IMF's estimate of their post-crisis market value. Sooner or later, however, in some crisis or other, the IMF's discounted estimate is going to turn out to be higher than the market's estimate. The IMF is going to need the resources to make good its commitment. And not even Stanley Fischer has asked for the IMF's funding to be topped up to the point where it could buy the entire national and provincial government debt of Argentina and Turkey, plus Brazil, Indonesia, and Korea--even at fire sale prices.

Meltzer's hope is that the commitment to buy will be enough, and so the last resort will never come. But history teaches us that when economies wish they wander off the subgame-perfect equilibrium paths that we economists love so dearly. The major financial institutions of New York had every economic incentive to carefully monitor and supervise what Long Term Capital Management, geared at 25-1, was doing with its portfolio in 1998: yet they did not do so. Those who bailed out of Mexican Tesebonos at the end of 1994 abandoned what turned out to be very attractive returns. And money managers who can explain today why they were selling Asian assets in January 1998 are hard to find.

So on the one hand we have Allan Meltzer of the Meltzer Commission, which believes that too many resources are devoted to crisis intervention and too much micromanagement of economies in the form of conditionality is imposed. On the other hand we have Allan Meltzer of the _Financial Times_, who wants many more resources at the disposal of the IMF so it can credibly commit to the mother of all rescue packages, and who calls for an IMF powerful enough to impose conditions on whom a government can repay and limit the contractual obligations a government can assume vis-a-vis its creditors. The _Financial Times_ IMF is a different organization than the current IMF, using different procedures to achieve the same goals, but it is also a much stronger IMF exercising more control over national governments that borrow from it than the IMF we have now. There is some intellectual schizophrenia here.

Now let me be clear: I am not criticizing this intellectual schizophrenia. I understand it. I share it. It is a necessary and inevitable consequence of any serious thought about this issue.

You are asking for trouble if investors in emerging markets begin talking about the "moral hazard play" or if investors in the S&P begin talking about the "Greenspan put." But it is desperately important to have a lender of last resort to put the situation right when the reasons that the economy has settled at the "unfavorable" equilibrium have less to do with economic fundamentals than with panic, revulsion, and discredit.

And the point of balance? It is an art, and that says nothing--and everything.

Things I Said:


Let me join Rob Stavins and Jon Gruber in the favorite sport of piling on to Kip Viscusi.

Because individual utilities are interdependent, I don't think that this distinction between internal costs--which are the business of the individual making decisions--and external costs--the neutralization of the externalities from which are the proper object of public policy--can be fully sustained. The premature death of my paternal grandmother from lung cancer saved my father and his brother some $20,000 a year--a present value of perhaps $140,000 given non-smoking related death rates. Yet if you asked my father and his brother what their willingness to pay to cure their mother's lung cancer would have been the day before her death, they would not have said that their willingness to pay was -$140,000; they would have said it was +$400,000 or so. An economist would argue that they should have made a transfer of $500,000 or so to bribe my grandmother to stop smoking, but she was at an internal equilibrium: she would not have spent the money, but have given it back, so there was no way they could boost her consumption to compensate her for not smoking.

Because of within-the-household transfers, interdependence of interpersonal utilities, et cetera, et cetera, doing the social welfare function evaluation seems to me extremely difficult conceptually. But just because it is hard is no excuse for not doing it. This half a million dollar money-metric social welfare effect is real, and it appears nowhere in Kip Viscusi's numbers. I am puzzled that there is no attempt to do the full social welfare analysis. So I don't understand what Vicusi's numbers are supposed to be measures of. Why do you think they are interesting numbers to present?

Health Policy

It seems to me that since Chris Jennings is not here, someone should try to channel him, so let me take up that task. I'm not a health economist, but I played one in the first two years of the Clinton administration, carrying spears for Lloyd Bentsen's deputy for health policy, Marina Weiss. And at the start of the Clinton administration fundamental reform to improve America's health care system did not seem politically impossible. Large businesses were terrified of the risk of accelerating health care costs. Small businesses were angry at the fact that they were shafted by insurers in terms of the prices they had to pay. Insured consumers were nervous that their health care coverage would vanish if they lost their jobs. A well-designed program that provided universal coverage seemed likely to be very attractive to all three of these groups. Given the extraordinary administrative inefficiency of American health care, as Uwe Reinhardt would point out were he here, there was also the possibility of achieving major administrative savings from administrative simplification over the decades subsequent to reform.

So it seemed highly probable in 1993 that some welfare-enhancing fundamental health care reform would pass. The executive branch would develop a policy by forming a large internal coalition, all of whom agree that the proposal will make the world a better place, and that it is close to the best that can be attained at the current moment. You form this coalition by hammering out an internal compromise within the administration. Then you have a large group of people who are enthusiastic about the proposal: they will go out and make your arguments for you. The compromises and concessions that had to be made within the policy-planning group in order to form the coalition will then perform a very important exterior purpose: just as they brought people within the process onboard, so they will bring other people outside the process who think in a similar fashion onboard as well.

But there was a problem: Ira Magaziner a management consultant. A management consultant's principal goal is to win a debate in front of his employer, the senior decision maker, the CEO. You win a debate by making intellectual arguments, controlling the flow of information to the senior decision maker, walling-off potential adversaries from the process, and winning the confidence of the CEO by telling him things that he likes to hear: that he is smart, that his goals can be achieved, that the nay-sayers just don't grasp the issues. Then, if you are a management consultant, the CEO likes you, hands you a large check, and you leave.

However, that's not the way it works in politics. Even if the President thinks you are a genius after you use your access to override its objections, the Treasury's objections are then repeated in stronger form by the CBO analyst of the Health Security Act, and then picked up by key senators like Moynihan, Breaux, and Chaffee. Because they were not answered in the internal executive branch policy process, there was no one who could persuade the center of the senate to support this reform. And because the initial proposal had been so sweeping, there was no one who could persuade the left to accept a smaller program.


I was hoping to hear something in this session that I have not heard. I was hoping to hear an answer to the questions: (i) how large should medical spending on the elderly be forty years from now? and (ii) what share of this spending should be paid for by Medicare? I don't understand how I can even start thinking about what the future of the Medicare program should be without answers to those questions, yet no one seems to have set out a view.


Yesterday it seemed to me that the health policy session needed someone to channel the views of some of the health policy High Politicians--in that case, Chris Jennings. It seems to me today that this session needs someone to channel the views of some of the Clinton Administration's environmental policy High Politicians. But I am loathe to do so: I fear permanent and severe mental injury should I succeed in channeling Carol Browner or Al Gore.

But here goes...

The Clinton Administration's environmental policy was a very odd combination of bold and aggressive environmental rhetoric, a fair measure of command-and-control, and an extraordinary aversion--except in a very narrow range--to the use of market mechanisms to give people incentives to undertake more environmentally-friendly production processes.

At least when I stood in the back of the Roosevelt Room, carrying spears for Treasury principals, whenever they would suggest the consideration of incentive-based environmental policies they would meet an immediate and visceral response from the High Politicians: "We tried that. We tried that by including the BTU tax in the 1993 budget. We got creamed. We got annihilated. There is substantial political support in the U.S. today for a strong EPA that commands companies to adopt clean and green policies. There is no political support in the U.S. today for a broad-based incentive-based environmental policy..."

I once asked some of the lobbyists for the American Petroleum Industry whether they had in fact shot themselves in the head by opposing the BTU tax so strongly and successfully. Hadn't the demonstration that "efficient" incentive-based policies were politically unwise meant that in the long run they would suffer much heavier and burdensome command-and-control environmental regulation? "Nah," they said. They did not believe in such linkage, and did not believe that what they would have lost in the BTU tax they would have more than recouped by better and more effective regulations in the rest of the Clinton Administration.

So I am actually more depressed about the future of environmental regulation than even Robert Hahn...

Things I Did Not Say (Because I Was Not There) But That I Would Have Said If I Had Stayed for the Rest of the Conference:


John Orszag, Peter Orszag, and Laura Tyson, "The Process of Economic Policy Making During the Clinton Administration"

As Paul Samuelson once wrote, “The leaders of the world
may seem to be led around through the nose by their economic advisers. But who is
pulling and who is pushing? And note that he who picks his own doctor from an array of
competing doctors is in a real sense his own doctor. The Prince often gets to hear what
he wants to hear.”

George Shultz and Kenneth Dam, Economic Policy Beyond the Headlines (University of Chicago Press: Chicago, 1998), page 176.

Reich asserts that Darman told him late in 1992, “Forget the Cabinet…They’re out of the
loop. This is where the loop begins. This is the loop. Right here. OMB. This is where
all the centers of power meet up. It’s power central.”35 Reporter Bob Woodward wrote a
devastating article on the lack of coordination among Bush’s economic advisers in late
1992.36 The article, for example, revealed that CEA Chair Michael Boskin had difficulty
obtaining a meeting with the President; Boskin finally had to threaten to resign before
being granted the meeting.

“Bush has never established a permanent, effective structure for
handling economic decisions…According to two sources, Bush has
complained that it is hard at times to sort out or make sense of the
advice he receives from his economic team of Treasury Secretary
Nicholas F. Brady, budget director Richard G. Darman and
Council of Economic Advisers Chairman Michael J. Boskin. ‘This
is not like the NSC [National Security Council],’ Bush told one
adviser this spring, according to one source. ‘It doesn't come to
me [in a decision memo] saying Brady thinks this, Darman that,
Boskin disagrees.’”

In the Woodward article mentioned above, for example, OMB Director Darman was quoting as calling
Secretary of the Treasury Brady a “dolt” and “probably the weakest treasury secretary in the history of the
country.” Such leaks could not have improved the working relationship between the two officials.

Enforcement of the President’s message can also, however, raise
tensions with agencies: The CEA, for example, resented the pre-publication
review of the Economic Report of the President that was undertaken each year by
the NEC; many on the CEA believed that the NEC staff were too concerned about
spin and not enough about substance.

In his best-selling book, The Agenda, Bob Woodward emphasizes the
differences of opinion within the economics team on the tradeoffs between deficit
reduction and investment spending, maintaining that those who favored the former and
represented the bond market prevailed over those who favored the latter and represented
traditional Democratic values. Compared to the size of the economy, however, the policy
differences separating these groups -- on the order of only $50 billion to $100 billion over
five years -- were insignificant. In fact, the economics team was remarkably united in the
view that a substantial amount of deficit reduction was essential to restoring the
economy’s long-run health. Deficit reduction came first -- not spending increases or tax

The failure of the health care reform process highlighted the importance of a
policy development process that would both reflect the opinions of all the relevant
agencies and develop proposals that anticipated the objections of opponents. The NEC
and its interagency process were undoubtedly strengthened during the first term by the
contrast between the budget plan’s success and the health care plan’s failure.

The President emphasized the apparently low-cost emission reduction
possibilities. As he noted during the October 1997 conference he hosted on climate
change, “I’m plagued by the example of the light bulb I have in my living room of the
White House that I read under at night, and I ask myself: ‘Why isn’t every light bulb in
the White House like this?’” The light that plagued his energy conscience was a compact
fluorescent bulb that is more expensive on the shelf, but less energy intensive, than
traditional bulbs.103 The President believed that wider dissemination of energy-efficiency
technologies like the light bulb were feasible, and could play a role in reducing
greenhouse gas emissions at relatively low cost. He once referred to some of the
economists who doubted such claims as “lemon suckers.”
The denouement of all these cross-cutting pressures was a policy through which
the United States agreed to the Kyoto Protocol, but with an emphasis on flexibility
mechanisms (such as international trading) and a pledge that the protocol would not be
submitted for ratification to the Senate without “meaningful participation of key
developing countries.” As part of the Kyoto Protocol, the United States agreed to a
mandatory emissions target of seven percent below 1990 levels by 2008-2012. Some
observers believe that the U.S. commitment was extreme, and it is almost impossible to
attain at this point. (In March 2001, President George W. Bush’s Administration
effectively renounced the U.S. commitment to the Kyoto Protocol.)

way of reconciling the different costs estimates -- a system of tradeable emission permits with a cap on the
permit price -- was briefly considered but rejected by the interagency team after intense pressure from the
environmental community. Under the proposed system, the U.S. Government would stand ready to sell
permits at some given price. Sales of such additional permits would ensure that the market price of the
permits never exceeded the price cap, thereby providing an upper bound on the costs involved. The cost of
such insurance is uncertainty in the degree of emission reductions achieved.

If an honest broker is important, where should the honest broker be located? We
argue that the presumption should be the White House itself. Only a White House entity
typically has the requisite access (in both appearance and reality) to the President and
other senior officials (including the White House Chief of Staff) to enforce the rules of
the game. Furthermore, only a White House entity can represent the interests of the
Administration as a whole, rather than the parochial interests of a specific agency.106 In
the corporate responsibility case study, for example, the Treasury Department was one of
the parties involved in an interagency dispute over policy. How could the Treasury
Department have simultaneously brokered a deal and represented its own position?

Pamela Samuelson and Hal Varian, "Information Policy in the Clinton Years"

All in all, the development and subsequent privatization of the Internet is a textbook case of
technology transfer. The Internet, with its open protocols and highly extensible nature, would have
likely never been developed, much less deployed, by the private sector. Open standards of the sort
that made the Internet possible may well require some sort of industry-wide, or publicly sponsored,
research and development. But once the Internet reached critical mass, the government wisely
decided to privatize the technology. Though, in retrospect, there were a few things that could have
been done differently, overall the process was hugely successful.

Page 6 from 43
The Web put a friendly face on the Internet, providing an interface that a 10-year old, or even a 50-
year old, could easily understand. Perhaps more importantly, the back-end protocols for the Web
were easy to understand and use as well, facilitating the rapid deployment of Web servers. The first
Web sites were in universities and research centers, but other organizations soon followed.
The next step in the development of the Web was to bring in the for-profit sector. In March 1994,
Jim Clarke, founder of Silicon Graphics, initiated talks with Mark Andreeson, the graduate student
at the University of Illinois who led the Mosaic project, about forming a company. The company,
initially named Electric Media, was formed in April 1994 with the goal of developing browser-like
technology for interactive TV.
Within a few weeks, the founders realized that interactive TV was less attractive a market than the
Internet itself, and they decided to try to develop a commercial version of the NCSA browser.
Accordingly they changed the company's name to Mosaic Communications and started
programming. After a dispute with the University of Illinois, the name was changed once more in
November 1994 to Netscape Communications.11
By 1995 everything was in place to create fertile ground for widespread Internet deployment. The
critical components were:
Personal computers.
Universities, major corporations, many small businesses, and a significant number of home
users had access to personal computers.
Local area networks.
Most businesses had networked personal computers together to make them easier to manage.
Standardized technologies like ethernet had become commoditized, and were widely available
at low prices.
Wide area networks.
Universities and research organizations had embraced the Internet as the primary network for
wide area communications. The technology for home access through dialup modems was
widely available, and there was a thriving culture of dialup access to computer bulletin
The Web.
Mosaic had provided a standardized interface for both users and information providers that
allowed for easy deployment of, and access to, online information.

Even during the height of Internet mania,
many sober observers pointed out that the valuations for "dot coms'' were simply not sustainable.
Here is but one example, published in July 1999.
Yahoo's revenue per page view consistently has been 4/10 of one cent ...If it takes
Yahoo 250 page views to get $1 in revenues, then to reach $35 billion in revenues [a
level which would justify its stock price] will require 8.75 trillion page views per year,
or about 24 billion page views per day. If the world population is 6 billion, then on
average everyone in the world, regardless of age, language, or access to the Internet,
will have to view 4 pages on Yahoo per day.14
Even proponents of Internet technology recognized the problem. Anthony Perkins and
Michael Perkins, founding editors of The Red Herring, a magazine devoted to the New
Economy, published a book in November 1999 called The Internet Bubble : Inside the
Overvalued World of High-Tech Stocks. Traditional value investors like Warren Buffet
warned that a crash was coming, but few investors listened.
Should the Administration have done something to dampen the swings of the stock market?
Was this, in fact, the motivation behind the interest rate increases of 2000? The Fed was
worried about the stock market bubble, but decided not to respond in any dramatic way. In
part, this was due to Greenspan's belief that the real economy was fundamentally sound. The
Fed continued to maintain a low interest rate policy during the late 1990s, apparently
believing that there was little danger of inflation. Chairman Greenspan argued in several
speeches that the widespread adoption of information technology had shifted the aggregate
supply function outward, so that the economy could produce more output without
experiencing upward pressure on prices. It was widely believed that IT had made the
economy more competitive, more responsive, and more productive.

Brynjolfsson and Hitt extend David’s argument by asserting that investment in computers only
achieves its full impact on productivity when work processes and practices change as well.17
Changing organization structures is quite time consuming and prone to failure. Hence, the delayed
response of measured productivity to computerization is due to the slow pace of organizational
A second explanation for the slow response of productivity to computerization is measurement
error. Many of the benefits from computers show up on the services side of the economy, and it is
notoriously difficult to measure improvement in the quality of services. As Brynjolfsson and Hitt
put it: “According to official government statistics, a bank today is only about 80 percent as
productive as a bank in 1977; a health care facility is only 70 percent as productive, and a lawyer
only 65 percent as productive as they were in 1977. These statistics seem out of touch with reality.”
Services have become a larger and larger component of the economy, even in traditional industries,
and the measurement problem is clearly significant.

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