The Council of Economic Advisers on Inequality and Structural Unemployment
On December 12, MDRC hosted a colloquium in Cambridge, MA, to celebrate our 40th anniversary and the contributions of former Board Chair Robert Solow, professor emeritus in economics at MIT. In one panel, former members of the president’s Council of Economic Advisers recounted the Council’s role in economic policy over the years. The panel was chaired by Rebecca M. Blank, Chancellor of the University of Wisconsin-Madison and a former member of the Council. Robert Solow himself spoke about the actions of the Council when he was its senior economist in 1961 and 1962. Joseph Stiglitz, now a University Professor at Columbia University, spoke about his time on the Council from 1993 to 1997 (he became its Chairman in 1995), and Alan Krueger, since 1987 a professor in economics and public affairs at Princeton University, spoke about his tenure as Chairman of the Council from 2011 to 2013. Their presentations were wide-ranging; these paragraphs summarize their remarks on inequality and structural unemployment, topics all three of them touched upon.
Robert Solow: In 1963, Mollie Orshansky developed a method of estimating the poverty line: what income was required for a family to be described as “poor.” It was done in a very simple way. She discovered that low-income families spent about a quarter of their income on food, so she suggested calculating the cost of a minimal, adequate diet and multiplying it by four, and describing that as the line that separated poor people from nonpoor people. This was extremely important, because now it was possible to count poor people. One only needed family income distribution data.
Without intending to, Orshansky cut the connection between poverty and inequality by defining the poverty line in absolute dollars. In other parts of the world, like Europe, the notion of a poor family is different. The poverty line in any European country is a fraction of median income. That is a different concept from a poverty line measured in dollars, and it makes an enormous difference.
Joseph Stiglitz: The inequality issue became a very big debate in the Clinton administration, partly because one of the members of the Cabinet was Robert Reich, and he felt passionately about that. He was concerned about the middle class, but people like Mary Jo Bane and others were worrying about poverty at the bottom as well. We had other people worrying about inequality at the top. There was a lot of presidential accounting as always, and we pointed out that during the Reagan and Bush administrations inequality had been growing. We hoped that the growth in inequality would stop and be reversed. That did turn out to happen, though I am not sure it had much to do with anything we did. Then, not surprisingly, once Bush got elected in 2001, inequality really started to grow again.
A couple of things we did made a difference. One was the Earned Income Tax Credit (EITC). There had been a very small EITC in existence to offset the Social Security payment of low-income workers, but our big expansion I think has proven to be the most effective of all antipoverty programs.
One of the things we proposed that turned out to be very counterproductive was to suggest a tax on wages over $1 million, because we were concerned with inequality at the top. An exception was inserted for performance pay, which turned out to be unintentionally destructive. It induced and exacerbated the use of stock options, and it created incentives to misreport income so you could show performance based on fake accounting, which led to the scandals of the early 2000s. It really contributed to creative accounting, but not to good economic performance. That was one of the real mistakes we made.
We also had four or five other policy issues on our agenda to alleviate or address inequality at the bottom. The most important was welfare reform, moving people from welfare to work. That was also a heated debate. The administration did not divide along lines that you might have thought. One of the people who opposed welfare reform was the Secretary of the Treasury Robert Rubin, because I think he was really worried it would not work. In terms of the analytics, the question was: what were the impediments to moving people to work? Was it a lack of incentives, jobs, training, home support, or transportation? There were different views on the relative importance of each. In the end, it didn’t make much difference, because the parts of it that many of us thought were really important, like support and training, we couldn’t afford. When I think about it in retrospect, I just think it was terrible. What we were talking about was something that would cost something like $4 billion over 10 years, and when I think that we spent $150 billion to bail out AIG, and that we were spending $60 billion every couple of months in Iraq, that amount pales in comparison.
The other question we worried about was what would happen in an economic downturn. Welfare reform worked okay when jobs were being created, but what would happen in a downturn? The discussion in the previous session was very telling on that: we basically eliminated welfare as a safety net. It is does not play a countercyclical role anymore. Temporarily we put something in place instead — expanded food stamps — but only temporarily.
Alan Krueger: I tried to raise the understanding about issues related to inequality within the Obama administration and in the public. This was an area that came naturally to me because I did a lot of research on the causes of the rise in inequality prior to joining the government. I was struck by a number of things. I was always quite reluctant to use the word “inequality.” The Wall Street Journal wrote an article about me in 1995 where they said “Krueger prefers to use the term ‘income dispersion,’” which is a more neutral term. I think the problem has become has become so severe that I am now comfortable describing the problem as “income inequality.”
I think there are lots of reasons to be concerned. To me, one of the greatest concerns is that the rise in inequality is going to jeopardize opportunities for children born in disadvantaged families. We see some evidence of that already. I drew attention to something we called the Great Gatsby Curve, which is the relationship between inequality at a point in time and subsequent intergenerational mobility. If inequality is high in a given period, then intergenerational mobility seems to be lower. There are theoretical reasons why you would expect that to occur. You would expect that to be the case even in a simple human capital model where the returns to skills have increased and there is intergenerational correlation in skills. It is also the case that we don’t have good data yet to be able to show that intergenerational mobility has slowed down. I would often ask my staff, “When will we be able to know the answer to this?” and I was told, “In 30 years, when we have more data.” I don’t think we should wait.
I tried to expand the playing field in terms of the explanations for the phenomenon, because the way it is talked about matters tremendously for the policy. I think it is a grave oversimplification to say that income inequality expanded in the United States because of technological change and globalization, although I would put those two high on the list of significant factors. But we know when we look around the world that Canada, for example — which had very similar exposure to global trends and very similar technology — has had very different patterns in terms of rising inequality and job growth. Some of the difference is supply-related: Canada improved its educational system, and I think that has to be part of our solution.
But I also tried to emphasize the role of institutions. It is striking how much evidence there is that norms matter at work, that people care about fairness, that employers know about that, that institutions matter, and that institutions help to enforce those norms and help the labor market to function. Yet when we think about why inequality has risen, these factors get very little weight. We can certainly point to the decline in labor unions and say that that contributed to the rise in inequality for men. We can point to the minimum wage and say its decline has contributed, especially for women. But generally I think there has been an erosion in the norms and institutions that enforce fairness at the workplace. A lot of the rise in inequality has been between companies, not within companies, which suggests that a lot of it is company policy. What Richard Lester used to call “the range of indeterminacy” in management policy in setting pay seems to have an independent effect.
When we go through a period where we cut income taxes for the most well-off, where we gut the estate tax, when income inequality is rising rather dramatically and almost all of that growth is going to the top 1 percent — that sends a signal that maybe those norms should change. I think one of the frictions we are facing in American society is that American workers haven’t really changed much regarding how they view relative pay and norms at work. I suspect there are some things that the President can do by using the bully pulpit, which is fortunate because that’s the main instrument he may have with this Congress. For example, the President can shine a spotlight on companies that have leaned against the trend.
On structural unemployment:
Robert Solow: In trying to promote macroeconomic expansion, the Council of Economic Advisers found that there were two main objections to a reasonable expansionary policy. One was the abhorrence of budget deficits. I learned a lot about the English language trying to devise phrases that would allow the Kennedy administration to run a higher budget deficit, but not higher than whatever Eisenhower had done. The other objection was that the unemployment rate — then about 6.8 percent — was mostly structural, and was all or largely mismatches (geographical, occupational, educational), and that any attempt to have general macroeconomic expansion would be foiled by bottlenecks and shortages. At the time there were books about The Automation Revolution, The End of Work, things like that. I presume there still are books with titles like that.
In 1961 my job was to produce an answer to this structural unemployment argument — if an answer could be found — and I did a crash self-education on whatever labor-market data existed in 1961 (which couldn’t be done today; there is too much of it). I discovered that it was all nonsense. There was no strong argument at all to suggest that what was left over from the recession of 1960 was qualitatively different in terms of structural unemployment from 1954, 1957, or 1958.
That argument keeps coming back. I have been wondering ever since why the notion never dies. I think it comes from a lack of general-equilibrium imagination on the part of the general public, but it also comes from a lack of general-equilibrium imagination on the part of a lot of economists. It’s exactly as if you looked at a car with a flat tire and you figured that the hole in the tire had to be at the bottom, because that was where the tire was flat. Since we are looking at a labor-market problem, the defect has to be in the labor market, not anywhere else.
Joseph Stiglitz: Like it was for Bob, the issue of “structural unemployment,” or “jobless recovery,” was always there in the Clinton administration. We had the same feeling as Bob about whether there was any validity to it. I worry, though, that although it was not true then, it may be somewhat true in 2013, because we have lowered the cost of capital relative to wages in the attempt to stimulate investment. That means that the investments that are made are very capital-intensive, optimizing the use of technology. If you have a putty-clay model, that means that the technology you employ is very capital-intensive. We have high levels of unskilled unemployment, and yet grocery stores and drug stores are investing in expensive machines that will add to unemployment. From a social point of view, that is not very beneficial, and it may be that the price signals we are giving are encouraging that.
Alan Krueger: In some senses the issues we faced in 2008 were the same as in 1961 or 1993, although I think they were much more severe over the last four years. The recession was much deeper. We had to face a financial crisis, which I don’t think was the case in 1961 or 1993. We had 15 more years of growing inequality, so that made the problem even more difficult to address. We had eight years of irresponsible fiscal policy preceding us, two wars, and demographic changes taking place that are quite different now from what was taking place in the early 1990s. I think about half of the reason for declining labor force participation is aging. The other half is connected to the weakness in the economy. Looking ahead, I think we will see more and more profound effects of demographic changes in the United States. If you go back to earlier Economic Reports of the President, this is something that is not a surprise. It was quite helpful to point to Greg Mankiw’s Economic Report of the President in anticipating we would be seeing a decline in labor-force participation in this period.
We did a lot of analysis on long-term unemployment, and, when you look at the characteristics of the long-term unemployed, it really weighs against the stereotypes a lot of us have. The stereotype I had — and I can say people in various senior places in the White House had — was that the typical long-term unemployed person was an older man who was displaced from a blue-collar job. That describes 6 percent of the long-term unemployed. Thirty-six percent were in service jobs; those were disproportionately women. You really see the wreckage of the Great Recession, and it was quite dispersed across the economy and geographically. Sure, it was concentrated in some areas and more concentrated among construction workers, but it is very widespread, suggesting that any sectoral type of policy is only going to make a small dent in the problem.
Compiled by Joshua Malbin