Stanford News

2/11/97

CONTACT: Stanford University News Service (650) 723-2558


Economists discuss relevance of their profession

In the business world, many economists don't call themselves economists anymore. They prefer the term "consultants." On Wall Street, investors are interested in what economists have to say, but they also are cautious. After all, five years ago no economists accurately predicted what the American economy would look like today.

On college campuses, economists sense that other academics often view them as people who, in the words of David Kreps of the Graduate School of Business, "deal in high mathematics making ridiculous assumptions."

In spite of this skepticism, economic concepts and language continue to be widely used in many professions to analyze everything from environmental pollution to political turmoil. Furthermore, the recent stability of the American economy and the financial discipline now being exhibited by other countries' governments may be partly attributable to widespread acceptance of macroeconomic principles.

These competing views of the relevance of economics were expressed by some of the United States' most prominent economists, who met recently at Stanford to discuss the status of their profession. Prompted by a Dec. 2 article in the New Yorker titled "The Decline of Economics," John Taylor, director of Stanford's Center for Economic Policy Research, invited Stanford economists and visiting experts to discuss the field's relevance during a recent meeting of the center's advisory board.

The mission of the center is to provide relevant economic insights to policymakers, and Taylor is one of dozens of center-affiliated economists who either have worked in Washington as policymakers or served as part-time advisers to various local, state, national and international organizations and governments.

In his article, New Yorker author John Cassidy argued that economics has become too theoretical, mathematical and irrelevant to the real world. He suggested eliminating the Nobel Prize in economics, and quoted one of last year's winners, the late William Vickrey, as saying that the 1961 research paper for which he won the prize was "of minor significance in terms of human welfare."

Cassidy's lengthy article apparently touched nerves within the economics profession nationally. At the annual meeting of the American Economics Association in New Orleans last month, Taylor said, the impact of the article wouldn't go away. "It was discussed at every session I attended."

At the Stanford discussion on Jan. 17, which drew about five dozen people, mostly economists, the consensus of those who spoke was that the New Yorker article not only was wrong in many details but also behind the times in suggesting that theory dominated the field today. Theory development has gone into decline, perhaps temporarily, the participants said, while empirical work related to real-world problems and including laboratory experiments is on the rise.

"The hottest people in the job market, and I would say an increasing fraction of submissions [to economic journals], are what we think of as being relevant to a certain problem," said Orley Ashenfelter, a Princeton economist and editor of the American Economic Review, the largest economics journal.

Most economics dissertations at Stanford are in applied areas today ­ for example, speakers said that even theoreticians, when they meet at international meetings, often discuss real-world problems such as trade policy and how to restructure a socialist economy.

Yet "the image of economists as dealing in high mathematics making ridiculous assumptions is fairly widely held" within the academy, according to Kreps. He said he came to that conclusion after discussions with professors in other disciplines on national academy committees.

Kreps and others said that empirical studies, in which economists form hypotheses and test them on real-world data, have increased, partly because computers have made more data available and more subject to sophisticated analysis. The development of game theory and information economics also has allowed microeconomists to study "face-to-face" economic activity, including behavior in controlled laboratory experiments.

"It isn't to the point that I can claim we are as good as particle physicists, but I think [experiments] have contributed a lot of realism to what we do," Kreps said.

Stanford Nobelist Kenneth Arrow was among those who agreed that theory has declined.

"I think we discovered some flaws in theory partly through practice. We know now that we know less than we thought we did. Obviously economics is a very imperfect science."

Arrow argued that economics should promote "eclecticism, diversification of the portfolio. You never know what's going to turn out."

Macroeconomists who study the overall economy have received "the brunt of the criticism," according to Taylor, a macroeconomist.

The New Yorker article alleged that "economists don't even have an agreed-upon story about how the economy as a whole works" since the ideas of John Maynard Keynes fell into disrepute in the early 1970s.

But Keynesian economics "has been replaced by a body of knowledge which is used actively in policy by people doing applied work," Taylor said. "If it is not apparent so much in our graduate schools, it is because we are on to new and more interesting things."

Newer macroeconomic models used by the Federal Reserve and other government agencies, Taylor said, have built in more recent economic theory on "rational expectations and equilibrium ways of thinking about the world, and by my way of thinking, they have made a hell of a difference."

Consider the following," Taylor said: "This expansion which we are in now is the second longest peacetime expansion in our history. [The first and the second longest] were separated by one of the mildest contractions in our history, so in a sense these two expansions . . . are unprecedented in their length and their degree of macroeconomic stability. It seems to me that reflects the received wisdom that the policymakers are using ­ keeping inflation low and not trying boom-bust policies."

Keynesian economics was not so much disproved, said Stanford economist Tibor Scitovsky, as changed by world circumstances.

"Keynesian economics was designed for a closed economy and in the meantime, economies have become much more open and financial markets more integrated," Scitovsky said.

Several economists also suggested that the public held unrealistic expectations for economics after World War II, and that it was not until the early 1970s that problems were presented that economists could not explain.

Macroeconomics is attracting fewer graduate students today, several economists said, because large mathematical models capable of guiding government planning have not been produced and because government research funding has shifted to studying more applied problems. The macroeconomists who are admired now, Stanford's Ronald McKinnon said after the meeting, are not theorists but "people who formulate simple hypotheses that can be tested, such as Taylor's rule," referring to John Taylor's guidelines for when the Fed should hike interest rates to keep the economy from overheating.

Graduate training

"It's not a question of whether the highest peaks visible are beautiful, but whether the whole mountain range is well designed," Henry Aaron, an economist from the Brookings Institution, said of the profession.

Aaron, a visiting fellow at the Center for Advanced Study in the Behavioral Sciences on campus, asked whether enough resources are directed to studies of real-world problems.

Criticisms about the relevance of economics research began within the profession at least a decade ago and prompted the American Economics Association to appoint a commission of 12 eminent economists to study doctoral programs. Members included Stanford's Arrow and Joseph Stiglitz, who currently is on leave while he serves as chief of President Clinton's Council of Economic Advisors.

The commission, chaired by Stanford's Anne Krueger, who was then at Duke University, reported in 1991 that it had found substantial discontent among economics professors, graduate students and non-academic employers. All groups agreed, to varying degrees, that graduate education, particularly in the first two years, had become too removed from real economic problems and focused instead on teaching students to use a growing number of analytical tools.

Lower-ranking departments emulated the top-ranked departments, the commission said, so that if the top departments did not teach their students how to apply economic tools, the professors they produced were not likely to teach their students to apply them, either. The commission, noting the economic principle of "comparative advantage," suggested that economic departments diversify and that they re-design core courses to provide students with more and earlier opportunities to use the technical tools on important problems.

Mathematics itself was not seen as a problem, Arrow said of the commission report, but some members felt it was used too often for its own sake.

"We really don't teach that much math if you look at our graduate syllabi," he said after the meeting. In fact, he added, the investment banking industry recently has been hiring Stanford-trained physicists because of their training in solving partial differential equations that are used in pricing derivatives.

Krueger was quoted in the New Yorker as saying that "basically, if only the report and a pin had dropped at the same time, the pin would have been noisy."

She told Stanford Report that she had not expected a strong reaction, however, because "I don't think things work that way. We always are tossing things out, and people are influenced by various things in their discussions within their departments."

Discussions about the graduate curriculum are ongoing at Stanford and faculty have differing views, department chair David Starrett said in an interview.

"It's certainly true that our program has become more technical over time, but I think most people would agree that there was a misleading suggestion in the New Yorker article that by handling things verbally and not doing more technical analysis, we could do better."

One trend, Starrett said, is that research support has shifted to applied areas. "Consequently, most of our dissertations are written in applied areas, but that's on top of a trend that has made the field more technical, more mathematical than it was 10 to 15 years ago."

Empirical research is fine, Starrett said, but "it's very costly to collect data and it is not as high a quality as you would like. In some sense, the technique has outrun the ability to test. That's a fair criticism but I am not sure what to do about it."

He added that "to do good experiments on human subjects is virtually impossible, compared to the physical sciences."

Business economics

In the business world, economists are less visible on payrolls than they used to be, said Walter Hoadley, senior research fellow at the Hoover Institution and retired chief economist of Bank of America.

"Virtually every CEO in every corner office in America has got an MBA degree, taught in many cases by you people," Hoadley told the academic economists. "They all feel they know it or that they don't need it anymore."

Second, he said, business is primarily interested in the ability of economics to forecast the future, but "the impression one gets from the outside, I'm told, is that we agree on very little ­ that you can get whatever view you want."

Young economists have "great opportunities in business, but they've got to be broad, peripheral-vision people in order to sell what they've got to sell more effectively," Hoadley said. The profession has not focused as much on how it can contribute to the bottom line as corporations would like.

James Sweeney, an economist in Stanford's School of Engineering, agreed that a broad-based education is valued.

"I see a very large growth in the pervasiveness of economics, but it is not controlled by economists," he said. "Economic concepts have been very widely dispersed into the basic thinking of people" in other professions, such as engineering.

Amateur economists now include investors, McKinnon said. Agencies like the International Monetary Fund first pressured national government leaders to follow sound economic advice, but their so-called "Washington consensus" on how to run an economy, he said, now influences investors' market behavior.

"Most countries are open to international financial flows, and they are vulnerable now to a run on their foreign exchange rate if they don't follow the Washington consensus," McKinnon said.

Currently, there is a strong job market in the financial sector for economists who can "provide added insight and perspective on developing countries," because of the growth in cross-border financial transactions, said John Lipsky, chief economist for Chase Manhattan Bank and formerly of the International Monetary Fund. But business in general is cautious about the value of anything but short-term predictions from economists, Lipsky said, because "the performance of the U.S. economy in the last five years has defied almost everybody's predictions."

"The most popular title for economists in business today is consultant," Hoadley noted. "They are not using economists for reasons that have been indicated here."

Public attitudes

Some economists also indicated they are troubled by a public perception of their profession as politically partisan.

Economists have become a "lightning rod" to the general public, creating a market for a journalism that plays "fast and loose" with the facts about economic research, said Princeton's Ashenfelter, who sits on the advisory board of Stanford's Center for Economic Policy Research. "Whatever you do in economics, some interest group is not going to like it."

As an example, Ashenfelter contrasted comments published in the New Yorker and Business Week on research by his Princeton colleagues David Card and Alan Krueger. The economists polled fast-food restaurants in two eastern states and concluded that an increase in the minimum wage in one state had not increased unemployment ­ empirical research results that contradicted conventional theory on the relationship between wages and employment.

When the American Economics Association was about to give Card its John Bates Clark Medal in January last year, a Business Week columnist, under the headline "Slipshod Research About to be Honored," suggested that colleagues of Card's had "rushed" the research into print without the usual review process. The New Yorker, in contrast, praised the Princeton scholars for challenging a "neoclassical shibboleth" and said their work was "worth a hundred theoretical models [published] in the American Economic Review."

"But it was published in the American Economic Review!" said editor Ashenfelter. Although he wasn't editor when Card's research was published, Ashenfelter said he checked Business Week's contention and found that "rushing here means 19 months from the time it was submitted for publication, just like every other article in the American Economic Review."

The impact of a politicized view of economics, he said, may be for "an inevitable number of people to move into sponsored research in areas that are really not driven by what we would most like to think of as science."

Stanford's Victor Fuchs said his recent research suggests that economists are already more politicized than they like to admit.

"When economists talk about economic policy and make recommendations about economic policy, it appears from my data that they are very much influenced by values, how they feel about some basic things such as the extent of redistribution of income in the society or the extent to which one should emphasize efficiency or one should emphasize equity, whether one should emphasize individual responsibility or social responsibility. You tell me how the economist feels about those things, and I can make a pretty good prediction about what kind of a position they will take on a variety of controversial issues in economic policy."

In surveys of economists, Fuchs said he and colleagues are finding "really big ranges of estimates among the leading people" on economic factors that should be established empirically. They also detect too much confidence by individual economists in their own estimates.

As an example, he cited policy discussions about whether the government could encourage Americans to save more by increasing the amount of income they can put into individual retirement accounts on a tax-deferred basis. There should be more agreement than there is, he said, on what proportion of the money going into individual retirement accounts represents additional saving, rather than merely a transfer from one type of savings to another.

"We are not sufficiently aware of the extent to which we don't know the magnitudes of the things that are absolutely essential to know if you are going to make good policy recommendations," Fuchs said.

Fuchs' research in the overconfidence of economists in their estimates did not surprise Arrow, who said studies have shown that scientists in general overestimate the probability that their research results can be replicated by others.

"There are all sorts of assumptions that individuals make that aren't specified in the model," Arrow said. "And in economics there are more of those."

-30-

By Kathleen 'Toole