Revealed Preferences

In response to my misplaced Jan. 27 speculation about the likely outcome of the search to select a successor to Martin Feldstein as head of the National Bureau of Economic Research, N. Gregory Mankiw, 50, of Harvard University, last week explained on his blog that, while the search committee had talked to him, he had at some point taken himself out of consideration.

Why did I decide not to pursue the job? As in many such decisions in life, various factors were at play, both personal and professional. But what really pushed me over the edge was pending tax policy. With the Bush tax cuts set to expire in a couple years, I am looking for ways to reduce my taxable income.

His announcement raises some interesting questions:


First of all, how much does the NBER presidency pay?  Feldstein, of Harvard University, has done a remarkable job of building the organization over thirty years. Last year he earned $600,000 in salary, with another $151,000 in benefits. (That is, of course, in addition to his university paycheck.)


Second, how much is Mankiw already taking home? The answer must be: Plenty. For it was in December 1992 that Mankiw startled the textbook world by leaving the publisher of his intermediate macroeconomics text to write in introductory text for a rival firm, for a $1.4 million advance, or roughly three times the previous record for a college text. The events were reported by Sylvia Nasar in a memorable story in The New York Times in 1995.


Mankiw, of Harvard University, had written his highly-successful intermediate book, the traditional warm-up to a new introductory text, for Worth Publishing. Scott Stratford, then an editor at McGraw-Hill, offered a princely $1 million to defect. Then, John Greenman, economics editor at HarperCollins, told Nasar, “That little money-grubber started shopping the deal.” Greenman declined to bid.


Worth Publishing matched the McGraw-Hill offer, but Harcourt Brace suddenly entered the bidding with a $400,000 signing bonus, plus a 22-percent royalty share. Mankiw signed the contract with alacrity in December 1992 – the year before president-elect Bill Clinton’s proposed  tax hikes would be legislated and take effect. A young Harvard star, Mankiw was known at the time as Lawrence Summers’ protégé.


(In those days, Harcourt was a recently-acquired subsidiary of General Cinema, which had been built by Richard Smith, one of seven members of the self-selecting corporation that governed Harvard University. Was the highly participatory Smith, who had just stepped down as CEO, influential in approving the offer – the biggest advance in the history of college publishing? Probably — Harvard didn’t get to be Harvard through a series of happy accidents.  But Harcourt and McGraw-Hill had traded top executives as well that year, and Harcourt intended that the $1.4 million contract should be a signal to the industry that Harcourt, recently returned from bankruptcy incurred in the course of fighting off a takeover attempt, was back and a player in college publishing again.)


Four years later, Mankiw delivered the introductory text. It was an immediate success. Heralded by a cover story in The Economist (“writing more in the style of a magazine than a stodgy textbook”), its exposition was indeed graceful and clear, but its real innovations lay elsewhere: as much as a third shorter than most competing texts, it contained hardly any equations.  It was adopted for introductory courses around the country, especially in junior colleges.


Fifteen years on, Mankiw’s intermediate macro text is still the best-selling book in its niche, ahead of competitors Andrew Abel/Ben Bernanke, Olivier Blanchard and Stephen Williamson.  His introductory text, too, has gradually worked its way to the very top of its market, outselling Robert Frank/Ben Bernanke, Paul Krugman/Robin Wells, Alan Blinder/William Baumol, Karl Case/Ray Fair, doing as well as the all-time champ, Campbell McConnell and Stanley Brue’s time-honored, easy-to-teach knock-off of Paul Samuelson’s original text. Today, in its fourth edition, it is published by Cengage, Harcourt’s college publishing division having been sold twice more.


List price, $175.95.


That adds up to a formidable income stream – with editions in 20 languages, perhaps as much as $6 million per edition of the principles text alone, or more than $2 million a year.


True, this is probably about to change. It is in intermediate macroeconomic textbooks that the current dogma in university economics is encoded. Mankiw’s 1992 book, like the one produced by Fed Chairman Ben Bernanke, formerly of Princeton University, and Andrew Abel, of the University of Pennsylvania, reflect the “new Keynesian” understanding that prevailed in Cambridge, Mass., in the late 1980s, when both books were conceived and written. Much has changed since then, especially as the result of a flurry of research on the sources of growth, which has greatly expanded the scope of traditional macroeconomics (these are the events described in my Knowledge and the Wealth of Nations: A Story of Economic Discovery) and in the conduct of monetary policy.


The first new intermediate textbook to take account of those developments appeared last month. Macroeconomics, by Charles I. (Chad) Jones, of the University of California at Berkeley, seems likely to precipitate a genuine collision of views in the college marketplace and to spawn many imitations. “Old School versus New School in highly anticipated economics textbook battle,” trumpets its publisher, W.W. Norton.  Louis Johnston, of St. John’s University, says in the accompanying literature, “All I can say is, Wow! This is exactly how I teach macro but I’ve had to cut, paste and mangle what’s in the other textbooks in order to do it.”


(It adds charm that Jones is a 1989 graduate of Harvard College, where Mankiw began teaching in 1985. “I have great respect for Ben and his coauthors and for Greg, but their books were originally conceived before the New Growth Theory had become standardized,” says Jones in a news release prepared by Norton.)


Thus Mankiw’s and Bernanke’s intermediate texts probably are headed towards their inevitable retirement, casualties of the continued unfolding of economics. But their introductory books will remain gushers for many years to come.


So how do marginal tax rates interact with money wages and non-pecuniary considerations in decisions of this sort?  Is it plausible that prospective changes in the tax law triggered Mankiw’s decision not to seek the presidency of the NBER?


Surely the real issue here is opportunity cost – the various activities that Mankiw would forgo in order to do the NBER job. He has a wife and three children. He seems to greatly enjoy teaching the introductory economics course at Harvard, which has the college’s second biggest enrollment (814 students), after Michael Sandel’s Moral Reasoning 22, “Justice” (1115 students). He has become a counselor to politicians (George W. Bush, whom he served as chairman of the Council of Economic Advisers, and Mitt Romney) and an economics personality (his daily blog is among the most popular such sites on the Web). He is an indifferent producer of new economic knowledge, but only six months or so of hard work are now required every three years to earn the next $6 million. And recently he was named – along with Summers and Brookings Scholar Douglas Elmendorf, late of the Federal Reserve Board – an editor of the influential Brookings Papers on Economic Activity. There’s no way he could do the NBER job without a substantial commitment of time.


So isn’t Mankiw stretching his credibility considerably when he says that that “what really pushed him over the edge” was the prospect of having to pay a somewhat higher tax rate on additional earnings? Or is his decision an example of emotional reaction overriding rational calculation that is the subject of so much research in today’s “behavior economics?” (It is, of course, always possible that he withdrew his name only after it became clear he wouldn’t get the job. That happens sometimes, too. )


Finally, how well does Harvard College understand these matters? That the professor of its introductory economics course should not only practice but preach tax mitigation is so far from that Puritan college’s traditional self-image (liberi liberaliter educati, or “gentlemen educated as gentlemen,” in the phrase of a seventeenth century president) that you have to wonder. To be sure, a university contains multitudes: John Kenneth Galbraith, who for decades was the most famous Harvard professor of them all, also expounded a politically-charged gospel on the Left. But JKG never taught introductory economics.


And, unlike Feldstein, who taught Harvard’s principles course for twenty years before him, Mankiw’s posture as a leader in his field has come at the cost of a certain amount of obfuscation. It is not clear that his students (or even many of his colleagues) understand how completely left behind he has been by the events of the past twenty years. For example, he is still describing the lighthouse as a public good, even though the various mechanisms governing the potential excludability of its warning have turned out to be the key to this most familiar of all examples of a nonrival good. And his most famous paper, “A Contribution to the Empirics of Economic Growth” of 1992, with David Romer, of the University of California at Berkeley, and David Weil, of Brown University, is generally considered to have failed in its defense of the “augmented Solow model,” meaning one in which human capital as well as capital and labor are sufficient to explain variations in growth.


As for the NBER appointment itself, we’ll have to wait to see how it plays out.  The job is the single most prestigious position in American economics. A search committee, led by Michael Moskow, former president of the Federal Reserve Bank of Chicago, is completing its assignment. The NBER, with its 51 directors drawn from virtually every sector of expert economic opinion – business, labor, government, professional societies and universities – is well-governed and tightly-run. Today James Poterba, 49, of the Massachusetts Institute of Technology, is a better-than-ever bet to succeed Feldstein.




I should explain, incidentally, that my assumption all along has been that a key issue underlying the NBER succession has had to do with control of office space. While the NBER serves as a meritocratic virtual forum for around a thousand policy-oriented economists from universities all over the country – that is the source of its strength — the organization’s  several floors of an office building are a physical space as well, a state-of-the-art combination of  carrels, offices and seminar rooms, located in Putnam Square, in Cambridge, Mass., between Harvard and the Massachusetts Institute of Technology, though much closer to Harvard. For twenty-five years, those cubicles and coffee pots have served as a hothouse nursery for many of the brightest young policy economists in the world, working as graduate assistants and visiting on fellowships.


Harvard has benefited considerably over the years from Feldstein’s stewardship of “the Bureau,” and from that of econometrician John Meyer before him. (It was Meyer who, in 1973, successfully moved the venerable institution’s headquarters to Cambridge from New York.) But the ability of its economics department to carry out a long-planned renovation of its offices in Littauer Hall was stymied last spring by Harvard’s Dean of Arts and Sciences, who ordered the department to share its space with the Fine Arts library for a period lasting up to a decade. Issues of office space thus loom larger than ever.


With its age-appropriate Clark Medal winners disqualified from contention for one reason or another — Lawrence Summers, Andrei  Shleifer and Susan Athey — Mankiw became the department’s logical candidate. With his withdrawal, the possibility that the Harvard and MIT would engage in an unseemly struggle behind the scenes for office space is moot.




February 4, 2004


A friend writes to ask if it is fair to characterize Mankiw as an “indifferent” producer of economic research. It is a severe judgment, but I think, one justified by the facts. Certainly he was a stellar contributor to the literature in the 1980s and early 1990s, when was one of the leaders of the “New Keynesian” movement. He has remained a dedicated participant in graduate education ever since.


Between 1984, when he completed his dissertation at the Massachusetts Institute of Technology, and 1992, when he began to write his introductory text, Mankiw wrote or co-authored 37 NBER working papers. Since then, he has written or coauthored 25 more, mostly on macroeconomic management, but only those co-authored with his student, Ricardo Reis, now of Princeton University, seem to have been of fundamental research interest.


Even allowing for the time he spent at the Council of Economic Advisers (2003-05), Mankiw’s research productivity is not what you would expect from a star economist in his forties. But the blog, the teaching and the textbooks take a lot of time. He’ll be a very good editor for the Brookings Papers.