The productive career of Robert Solow
Last summer, as he turned 95, the economist Robert M. Solow sat at home poring over a draft outline of “The Work of the Future,” an MIT report about technology, jobs, and economic growth. Solow has been studying these topics since he returned from fighting in World War II—and won a Nobel Prize in 1987 for demonstrating that technological innovation generates a huge portion of economic growth.
True, Solow’s eyes bother him these days, and he reads less than he once did. His wife, Barbara, herself an economic historian, died in 2014, after nearly 70 years of marriage. And the economics colleagues he worked alongside for decades at MIT—and with whom he built a powerhouse department from scratch—are no longer around either.
“I’m the only one who’s still inhaling and exhaling,” Solow says wryly, sitting on his living-room couch.
But Solow, an Institute Professor emeritus, is doing quite a bit more than that. He reads academic literature, including papers about productivity, and follows economic trends, world events, and policy debates. His “wickedly devious sense of humor,” as his Institute Professor colleague Daron Acemoglu puts it, remains intact. Having joined MIT in 1949, Solow is a macroeconomist whose career almost predates the word “macroeconomics.” Yet here he is seven decades later, rigorously examining the draft materials of the new work report.
Solow serves on the Work of the Future Task Force’s advisory board, and near the project’s start in late 2017, he and MIT sociologist Susan Silbey wrote a memo offering guidance for the report’s authors—MIT economist David Autor, MIT engineer and historian David A. Mindell, PhD ’96, and Elisabeth Reynolds, PhD ’10, director of MIT’s Industrial Performance Center. They pointed out that despite ongoing speculation about what robots, AI, and automation will do to work, the more pressing job issues in the United States right now are the loss of middle-class careers and the rise of inequality. While the task force’s brief was broad, and the report does examine technological developments, Solow and Silbey stressed the importance of policy decisions in shaping these workplace trends.
Solow and Silbey were the first to emphasize to the group that “you can’t have this discussion without thinking about the backdrop of all the social changes of recent decades,” says Autor. After reading the task force draft materials, Solow met with Autor and Reynolds and reiterated the message: our policies, not just our technologies, are dramatically affecting work, careers, and income inequality.
Although Autor and others on the task force have long studied job issues, a little encouragement from Solow can go a long way, like Sandy Koufax telling a pitcher to trust his fastball. The interim report, published in September, emphasizes economic and social realities. Technology has polarized the US workforce, generally aiding white-collar workers while hurting others. The future impact of high-tech advances on work is unclear, but what is certain, the report contends, is that new policies are needed to rebuild a healthy middle class, including better workers’ representation in firms and a tax code that benefits labor.
That conclusion may not be what people expect from MIT, or from an economist famed for quantifying the impact of technological innovation on society. But Solow thinks today’s situation demands it.
“I was afraid that at a place like MIT, at which I’ve spent my life and which I love, the tendency was going to be to think of the future of work as primarily a technological problem, and primarily a problem to which if there’s any solution, it would be a technological solution,” Solow says. “I thought that was dead wrong, and I was afraid it would lead nowhere. But I’m very glad to say the interim report that the commission has put out struck me as pretty damn good.”
That view reflects Solow’s intellectual flexibility, Autor says: “It’s especially interesting coming from Bob, since his Nobel-winning work is about the benefits of technological progress. Bob is just not dogmatic.”
Inequality is one area where his thinking has changed. In 1962, when Solow was a staff economist for the White House’s Council of Economic Advisers, the top 1% of earners in the US accounted for 12.6% of national income, while the bottom 50% of earners had 19.5%. That has reversed: in 2014, the bottom 50% accounted for 12.6% of income while the top 1% had 20.2%—the highest percentage since the crash of 1929, when Solow was five.
The disparity in net assets is even greater. In 2014, the top 1% of households had 38.6% of the nation’s wealth, compared with −0.1% (that’s right) for the bottom 50%. Talk to Solow about the economy today, and it’s likely inequality will come up sooner rather than later.
“Now it’s just a topic you can’t ignore,” he says, adding, “Society has changed. And one of the ways in which the society has changed is, as everyone now knows, [we have] vastly more inequality than we used to have.”
Solow’s willingness to talk bluntly about the topic has changed too. The late philosopher Ronald Dworkin, a friend, once asked what he thought about inequality. “I cautiously said, ‘Well, I think it’s un-aesthetic,’” Solow recounts. “And Ronnie said, ‘Do you mean you think it’s immoral?’ And I gulped three times and I said, ‘Yeah, I think it’s immoral.’ Economists are taught not to think in those terms. But just the sheer fact of vast inequality is important for our politics.”
Indeed, Solow emphasizes, inequality has created a self-reinforcing loop in which money shapes government policies that favor the rich, and thus begets more inequality.
“This great inequality of wealth should concern everybody, because it can’t help but influence or control politics,” he says. “One of the things you can buy with vast wealth is votes in Congress.” He adds: “It’s going to affect legislation that affects the labor market and property, not to the advantage of wage earners.”
Solow’s interest in the topic dates back to the 1940s: his doctoral thesis at Harvard focused on income inequality, though his perspective then was markedly different.
“I started writing a PhD thesis about [income] distribution not because it seemed like an urgent question of the time, but because I thought I had found tools that would enable me to study the subject more efficiently than anyone had before,” he says. “It’s not that in 1949 or 1950 I looked around and said, ‘Oh, the real problem our society faces is inequality.’ I didn’t think that. I thought the real problem our society faced was not to have wars, since I’d just been soldiering for three years, and not to have depressions, which may have been part of the cause of the war.”
Now Solow is more concerned with, say, the reduction of labor union power in the US, a lesser consideration in 1949.
“Undoubtedly one of the forces, and this is in the [MIT] commission’s interim report … is the decay of trade unions,” he says. “One of the things [the task force] started to recommend, bravely and I think rightly, is that one way to improve the situation is somehow to give wage earners more power within firms.”
But why have trade unions declined? Clearly the rightward drift of politics around 1980 meshed with the offshoring of work, which reduced labor’s leverage, but Solow thinks it’s hard to construct a definitive account. “Getting at causes is even less like science than the rest of economics,” he quips.
Solow thinks we do not fully appreciate how many people now work in service industries. And, he adds, “it’s probably harder to organize service workers than factory workers.” Six employees scattered in a CVS do not have the same type of experience as 600 on a factory floor. The nature of work today reduces labor solidarity.
“Workplaces over time in the economy as a whole probably have become more isolated,” he says. “In fact, the image of anybody working today is one person, staring at a computer screen.” Yes, technology does matter, but sometimes in unexpected ways.
A small program big on camaraderie
Growing up in Brooklyn, where his father was a fur merchant, Solow skipped two grades in school. He earned a scholarship to attend Harvard at 16, in 1940, and enlisted in the Army two years later, at a time when the draft age was 21. After a stint in Algeria, he went to Italy, in a company intercepting German radio signals just behind the front. Having picked up German in college, he often hunkered down in specially equipped trucks, trying to translate messages. If the trucks had been spotted, he says, “we would have been dead meat.”
When Solow talks about the war, though, it’s mostly about people. He notes that the Army exposed a college student like him to Americans of many social backgrounds from around the country, and his commander, John Faison, became a lifelong friend.
Solow returned to Harvard in 1945 and married Barbara (“Bobby”) Lewis, a Radcliffe student who helped interest him in economics, which he started studying in earnest only after the war. He finished his undergraduate degree and soon zipped through Harvard’s PhD program. By 25, he had a wife, a doctorate, and three years of war service to his credit. What he needed was a job. In 1940, Harvard had let a newly minted economics PhD, Paul Samuelson, slip away to MIT. In 1949, Solow ventured down Massachusetts Avenue as well.
At MIT, he found a small program big on camaraderie: the professors would eat lunch together every day and keep their office doors open. “I think it was because we were a low-prestige department,” Solow says; there were no spoils to fight over. And yet Samuelson gave MIT a huge comparative advantage as it built the department.
“Paul was the best economist in the world, and there was nothing stuffy about him. He interacted with everybody,” Solow says. “Just us kids together—we had that kind of atmosphere.” For about six decades, he shared an office suite with Samuelson, which also generated about six decades of conversation (see “The Office Next Door,” MIT News, November/December 2011).
“I miss him,” Solow says, simply.
A whole group of these economists would become MIT lifers: Charles Kindleberger, Harold Freeman ’31, Cary Brown, and Robert Bishop, among others. (George Shultz, PhD ’49, the future US secretary of state, was an early colleague but left MIT in 1957.) Together, they built a department producing innovative research—using sophisticated models grounded in empirical data—while taking great pride in teaching.
Every year, Solow would rip up his undergraduate lecture outlines, force himself to rethink his course material, and write new lectures, which he often delivered without notes. “I found that if I wanted to teach something difficult, even for an undergraduate course, the first time I taught it, I thought I did a lousy job,” he says. “The second time I taught it, I was better at it. The third time was probably the best.”
All that collegiality paid off intellectually, too. “More than being pleasurable, I think it makes for good work,” he says. “Talking to your colleagues—or, in my case, standing at a blackboard with them and talking and scribbling—improves the product.”
Zeroing in on technological progress
In the mid-1950s, Solow also started publishing his career-defining work. His paper “A Contribution to the Theory of Economic Growth,” from the Quarterly Journal of Economics in 1956, presented his famous model describing how under certain basic conditions, even a growing population and increased capital investment will not sustain economic growth. Instead, it is technological progress that will power such growth over time.
The next year, in a paper called “Technical Change and the Aggregate Production Function,” Solow provided the numbers to back this up, based on data from the US economy in the first half of the 20th century. “Total factor productivity,” as he called it—the set of technological, cultural, and economic factors that included everything other than population growth and simple capital investment—accounted for 80% of growth. Then, in 1960, he extended his analysis in a third paper modeling a scenario in which capital investment becomes more technologically sophisticated over time.
When Solow won the Nobel Prize in 1987, all three papers were cited for providing a framework for understanding growth. For decades that work has illuminated the importance of innovation and technological progress to society’s increasing wealth.
“The work Bob did on growth was really addressing one of the most fundamental questions in all of economics,” MIT economist James Poterba says. “Bob is the person who basically laid out the framework that is now the standard part of the economics toolkit for thinking about … how labor and capital together, along with the advance of technology, contribute to changes in the standard of living, and changes in overall economic output.” As such, Poterba says, “Bob has often been the giant on whose shoulders other researchers stood.”
Economists have tried to measure the potential components of total factor productivity in greater detail, and commentators still point out that it is, as Solow has always maintained, a catch-all category that includes things like the education necessary to produce innovations. He acknowledges the general nature of the concept—he says it’s been called “a measure of our ignorance”—and advises thinking broadly about innovation. For instance, “we probably overvalue the inventor of something brand new,” he says, “and undervalue continuous improvement … making things a little more efficient and better without anything you would identify as a discovery or a patent.”
Solow’s greatest misgiving about the concept is that it remains difficult to isolate the economic impact of specific innovations.
“I was always faced then with a puzzle that I don’t think has been resolved yet, by anyone,” he says today. “What I devised was a way of looking at fairly aggregative data, and extracting from those data a record of what you could think of as a general year-by-year rate of technological change. On the ground, so to speak, all that is the result of individual changes in individual technologies. And I always wondered, would it ever be possible to link the aggregative series with things that were happening at the level of either individual industries or firms or technologies? And I don’t think that has happened.”
So how would he study that?
“I’m 95,” Solow says. “If I were 70 years younger and I wanted to go at that question, I would try to do something like choose a fairly narrowly defined industry, and get a technological history of that industry over some years, so that what you were looking at were discrete, in a way, changes in productivity.” Even that would be tricky, he says. For one thing, new technologies are adopted over time, something that would have to be accounted for carefully.
Solow’s evaluation of his own work is consistent with his view of the discipline as a whole: good research starts with meaningful unanswered questions.
“I think the way people do economics today is too much governed by the availability of data,” he says. “A lot of the articles that I see written in the journals seem to exist not because there is a problem here that needs to be solved, or a puzzle that needs to be explained, but because I have come upon this enormous bunch of data, [and figure] these data have to include the answer to some question.” But, he adds, that’s “natural,” given the sheer amount of data on hand and the pressure to publish.
The “Solow Decade”
In 1961, when Solow joined the Council of Economic Advisers, he found himself giving tutorials to a new pupil. The council’s chair, Walter Heller, would list the author on each memo he sent the president. Occasionally the phone would ring, Solow recalls, and John F. Kennedy would be on the line, wanting to talk economics.
Kennedy “had curiosity,” he says. “If Heller sent him a memo, he read it, and he wanted to make sure he understood it. It’s the beginning of wisdom.” Moreover, Solow was working alongside notables such as James Tobin and Kenneth Arrow. “We thought we had one of the best economics departments in the country, sitting there in the Old Executive Office Building,” he says.
But the best department by then was MIT’s, which had added more star professors, including Franco Modigliani. And Solow missed having full-time students. He returned to Cambridge in 1962, and resumed teaching, advising a long series of superb graduate students.
“I like teaching, and I adored working with graduate students,” he says. “I had the most amazing collection of PhD students you could imagine.”
In one 10-year period, for instance, Solow served as the principal PhD advisor to future economics luminaries Peter Diamond, PhD ’63; George Akerlof, PhD ’66; Robert Hall, PhD ’67; William Nordhaus, PhD ’67; Joseph Stiglitz, PhD ’67; Martin Weitzman, PhD ’67; Avinash Dixit, PhD ’68; and Alan Blinder, PhD ’71. Dixit has called this the “Solow Decade” at MIT, because Solow oversaw so many of the department’s dissertations. “Paul Samuelson was royalty, but Bob was very much the prime minister running the country,” Dixit recalled, in remarks for a 90th-birthday gathering for Solow.
All told, Solow served as principal advisor to more than 70 PhD students, dozens of whom became noted economists and four of whom would win Nobels in economics themselves. (See “Solow’s laureate legacy,” below.)
“There’s no question that if one looks at the hall of fame for economics advising during the 20th century, Bob is one of the charter members,” Poterba says. Solow’s explanation for this is simple. “You only have really good ideas once in a while,” he says. “I would rather teach a really bright student than write a mildly interesting paper.”
With Solow as their advisor, graduate students regularly received copious notes on their work, full of constructive criticism, encouragement, and suggestions. “If my wife were still alive, she could tell you that’s how many evenings were spent: me sitting in the living room, grading some student’s thesis draft and scribbling comments all over it,” he says. Barbara Solow, who also got her PhD at Harvard, had her own work to do. An expert in Irish economic history and the Caribbean slave economy, she taught at Brandeis and Boston University.
Today, Solow’s former students eagerly attest to his brilliance as a teacher and mentor. “He was extremely generous with both his time and his ideas,” says Akerlof. “Bob believes in taking care of people.” Blinder, who calls him “the best of the best,” says that when he arrived at MIT himself, the biggest piece of advice he got was “Take anything Solow teaches.” What made him a great teacher, he adds, was his “ability to get at the essence of something without spurious complexity.”
Dixit has also called Solow “the best classroom teacher I ever had,” adding that as an advisor, he was “amazingly quick to grasp one’s vague thoughts and questions and offer suggestions.”
When Akerlof was a graduate student, “the faculty were there to greet students on the first day we arrived at MIT, and to give us advice,” he recalls. “I mean, you would not expect the most distinguished professor at the best economics department to be there, waiting for you. That was extraordinary.” Beyond the classroom, departmental activities included picnics and a student skit night, to take the edge off the academic pressures.
“There’s an old line that the MIT department was the only kibbutz in economics,” Autor says. “Bob always understood that students don’t usually do their best work when they’re feeling miserable and intimidated.”
That applied to undergraduates as well. MIT biologist H. Robert Horvitz ’68, a Nobel laureate in physiology or medicine, wrote a senior economics thesis for Solow, and recalls him as a “wonderfully warm and supportive” advisor. “I would go to him completely stuck, and within a few minutes he would resolve my issue and send me on my way,” Horvitz says. “I would then work on the thesis for two or three weeks, get stuck again, and return to Bob for the next round of his advice.”
That sense of admiration lives on. Acemoglu says Solow has been “a maestro with mathematics and technical arguments, while never losing his sense of the big picture and the social and ethical aspects of economics.” Silbey says she is amazed at how he has remained “as sharp and witty and knowledgeable as ever” well into his 90s. Today, a conversation with Solow—marked by boundless affability, crisp observations, and a slew of humorous asides—produces a feeling his listeners want to bottle.
“In a sense, it has been bottled,” Autor says. “The culture of the MIT economics department was built by Bob’s priorities.”
If nudged slightly, Solow will say how important this is to him. “One of the things I’m proudest of in my whole life is we built this department, or helped to build this department, so it would pay attention to students,” he says. “The department still retains that spirit.”
“We didn’t sit down and plan how we were going to convert this service department into—well, I think it was the best department in the country or in the world,” he adds. “We just had fun and we taught hard.”
Solow’s laureate legacy
The list of 70-plus alumni who can count Nobel Prize winner Robert Solow as their advisor reads like a who’s who of notable economists. Four won Nobels themselves—and an undergraduate advisee won a Nobel for biology research.
Joseph Stiglitz, PhD ’67
Nobel Prize in Economic Sciences, 2001
Powerhouse researcher who also built a high-profile policy career. Stiglitz’s scholarship has often illuminated the inefficiencies of markets, including incomplete information, monopolies, and suboptimal wage levels. He has written: “The particular style of MIT economics suited me well—simple and concrete models, directed at answering important and relevant questions.”
George Akerlof, PhD ’66
Nobel Prize in Economic Sciences, 2001
Scholar known for work on market imperfections and other social phenomena. His 1970 paper “The Market for Lemons” described how the limited information available for buyers (“asymmetrical information”) influences prices, with the existence of low-quality products undercutting prices for higherquality goods. Of Solow, Akerlof says, “He wasn’t just your professor at the time. He became your professor for a lifetime.”
Peter Diamond, PhD ’63
Nobel Prize in Economic Sciences, 2010
Made major contributions in multiple areas, including his model of public debt, and studies about taxation levels and economic efficiency; won the Nobel for his work on “search theory,” which applies to the friction in job markets. Has called Solow “an outstanding mentor in all the dimensions that mentorship should have.” An Institute Professor emeritus, he joined MIT’s faculty in 1966.
William Nordhaus, PhD ’67
Nobel Prize in Economic Sciences, 2018
Longtime Yale professor noted for his “integrated assessment model” estimating the economic effects of climate change, which has often been used to make the case for a carbon tax. The Nobel citation for Nordhaus and co-winner Paul Romer notes that their insights “highlight the strengths of Solow’s original framework.”
H. Robert Horvitz ’68
Nobel Prize in Physiology or Medicine, 2002
A distinguished biologist and an MIT faculty member since 1978, Horvitz received the Nobel for research that helped identify genetic pathways implicated in human disease, including programmed cell death. He majored in economics and mathematics as an undergrad. Solow served as his advisor for his senior thesis, which examined business incentives to deplete the environment.
Geoffrey Hinton tells us why he’s now scared of the tech he helped build
“I have suddenly switched my views on whether these things are going to be more intelligent than us.”
Deep learning pioneer Geoffrey Hinton has quit Google
Hinton will be speaking at EmTech Digital on Wednesday.
Video: Geoffrey Hinton talks about the “existential threat” of AI
Watch Hinton speak with Will Douglas Heaven, MIT Technology Review’s senior editor for AI, at EmTech Digital.
Doctors have performed brain surgery on a fetus in one of the first operations of its kind
A baby girl who developed a life-threatening brain condition was successfully treated before she was born—and is now a healthy seven-week-old.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.