AUTHOR: TUCKER GAUSS – FEBRUARY 4, 2023
EDITOR: PALLAVI MURTHY
Michael Reich is a Professor of Economics and Chair of the Center of Wage and Employment Dynamics at the Institute for Research on Labor and Employment at UC Berkeley. His research specializes in labor economics, minimum wage, socioeconomic mobility, historical economics, and public policy.
Gauss: What inspired you to get into the field of economics? What made you interested or motivated to study some of the topics that you’ve researched?
Reich: I didn’t know anything about economics until I was in college. I come from a family of Holocaust survivors, and was born in Europe at the end of the Second World War. We lived in a Displaced Persons camp in Germany for over three years and then my family came to New York City as refugees in 1949. My father, my mother, my aunts and uncles all went to work doing unskilled jobs in the garment industry, which was then and still is a low wage industry. My father eventually became an employer in the industry. Still, I grew up with a sense that low wage labor markets were places where people got stuck and very hard to escape from. As a child I was already very conscious of racial and economic inequality.
My first passion was science. In 1957, when I was nearly twelve, the Russians sent up the first orbiting satellite, called Sputnik. Just a few months later, the U.S. also launched a satellite, called Vanguard. Soon after, the Russians put the first astronaut in orbit. Those were exciting times. So I was a child of the Sputnik era, really interested in science and math and especially in physics. I went to college thinking I would be a physicist. I’d already participated in a science honors program for high school seniors at Columbia University, where I was exposed to some famous physicists. After my freshman and sophomore college years I worked for two summers in the U.S. Naval Research Laboratory, in Washington DC. That lab built the first Vanguard satellite. When I was 17, I was already working in the same lab that had produced our first satellite just five years earlier!
I did some interesting work in applied physics during those two summers, which led to a journal article in Physical Review. But I also learned that physics in the laboratory was very different from physics in the classroom. It was a lot messier, it was harder to do and the cutting edge of physics involved phenomena that were not that relevant to the real world. Anyway, as a result I was already in D.C. in the summer of 1963. I was therefore able to attend the 1963 March on Washington, the one with the famous Martin Luther King “I Have a Dream” speech. I wrote about that experience in August 2023 for a magazine—American Prospect; ten years earlier, The New York Times published a short item about how the march changed my life.
The march showed me the power of unions, civil rights movements and collective action to achieve some measure of racial and economic justice. I began to think that I won’t major in physics, maybe I’ll switch into economics, which was my minor at the time. I did go to grad school in economics, where I learned that I could make a difference in the world by using essentially the same scientific methods that I’d learned in physics classes. That was exactly what I was hoping for.
During my graduate student years I was able to work on two research projects. One was my doctoral dissertation, which examined who benefits from racial inequality (spoiler: not white workers); the other examined how low wage labor markets differ from other labor markets. I’ve been working on those topics ever since, as well as on many other topics, over the course of my career. I joined the Berkeley Economics Department in 1974, and I’ve been a professor here ever since.
Gauss: Wow, that sounds remarkable. Regarding those lower wage labor markets, I wanted to hear your perspective on political analysis a bit. Studying economics—graphing models, making proposals—is a very different thing in terms of how that’s actually going to be put into policy. So I’m curious, what’s stopping Congress from implementing a $15 minimum wage?
Reich: So you’re referring to the work that I’ve been doing for the last 15 years or so on minimum wages. Let me talk about the economics of minimum wages a bit first and get to the politics afterward. Thirty years ago, the conventional wisdom in economics drew upon the simple observation that when the price of something goes up, the demand for it goes down. So if you go to the grocery store and the price of beef is higher than last week, you’ll buy more chicken and less beef. This simple economic model assumes that you don’t experience any switching costs if you buy more chicken and less beef. It also assumes the markets will clear— at the new price the quantity of chicken supplied will equal the quantity demanded. In other words, in this perfectly competitive model markets work smoothly and instantly—without any friction—and prices among competing businesses must all be the same..
The same theoretical argument, this same model of perfect competition, was used to explain the effects of the minimum wage. If you mandate an increase in the price of labor, then low-wage employers, such as fast food stores, will want to hire fewer workers. So a higher minimum wage will hurt, not help, the workers for whom the policy was intended. That was the old consensus thirty years ago. But a lot of very good empirical work since has shown that the consensus was factually wrong. This new work used quasi-natural experiments to identify the causal effects of minimum wages on employment.
A quasi-natural experiment refers to an event that is very similar to a controlled laboratory experiment or a clinical trial of a new drug, in which one group gets a treatment and a similar group gets a placebo. The older empirical studies used methods and data that were not credibly able to distinguish a correlation from a causation. We all know that people bring umbrellas with them on days that are rainy, but that correlation does not mean the umbrellas caused the rain.
Quasi-natural experiments became possible because, beginning in the late 1980s, a growing number of states began to raise their minimum wages above the federal level. In the early 1990s, for example, New Jersey raised its minimum wages, but neighboring Pennsylvania did not. So one could compare employment changes in fast food chains in areas of New Jersey that were near the Pennsylvania border to employment changes in the same fast food in Pennsylvania areas that were near the New Jersey border. Both areas were affected similarly by other economic factors, such as national and regional business cycles, but only New Jersey raised its minimum wage.
Gauss: The New Jersey and Pennsylvania study by David Card!
Reich: Yes, you’re well versed on this topic. That paper, which found employment grew as fast in New Jersey as in Pennsylvania, had an enormous influence. But it did not convince all economists, because it was just one case study with very limited before and after treatment data. So I did a study about 15 years ago that generalized the Card and Krueger paper on New Jersey and Pennsylvania. My study compared all the border counties in the U.S. where there was or had been a minimum wage on one side of a state border and a different one on the other side. We collected restaurant employment data on several dozen such border differences over an 18 year period. So with this much larger dataset we could deploy more advanced econometric methods than Card and Krueger could.
Surprisingly, we found the same results as Card and Krueger. The employment effect wasn’t exactly zero for every one of these border pairs, there was a bit of a distribution reflecting idiosyncrasies in individual cases. But the central tendency was a statistically precise zero. This paper received a great deal of attention from other economists and helped change the conventional wisdom of the profession. And it has stood the test of time.
So how can we explain the absence of a disemployment effect? The answer is that labor markets are different from highly competitive markets, such as the markets for chicken and beef. It takes workers a while to search for a job that’s a good match for their needs and preferences, especially if they already have a job and have limited time to search. It also takes a while for employers to find workers that are a good match for jobs in their companies, even when they are hiring for low skill jobs. As a result, at any point in time some workers who want a job don’t have one (unemployment) and some job openings are not filled (vacancies). Economists refer to these characteristics as frictions.
In economics, unlike in Newtonian mechanics, frictions play more than a small role. They fundamentally change how markets operate. For example, employers in frictional labor markets face a tradeoff between the wages they offer workers and the costs they face to recruit and retain their workers. A low wage will mean more employee turnover and higher job vacancy rates. A high wage will generate lower turnover and vacancy rates. Either of these choices could generate similar profits. The availability of these choices means that employers possess wage setting power. This model especially applies to low-wage labor markets, even when, for example, many competing fast-food restaurants are concentrated in a very small area.
Employers use their wage-setting power to keep wages below what they would be in a competitive market, even if they then experience higher employee turnover. As I found when I was a graduate student studying low-wage labor markets, and has been shown many times since, low wage firms have very high rates of employee turnover, greater than 100 percent per year. The overall turnover rate in the U.S. among all employers is about 30 percent.So low-wage firms constantly have vacancies. McDonald’s and Walmart are continuously hiring, trying to reduce those vacancies.
A higher minimum wage then makes it easier for employers to attract workers and workers will also then stay longer with the same employer. That’s what we found when we did a follow-up econometric study to our border state paper, using the same econometric methods. We found that employee separation rates went down when the minimum wage went up. So the minimum wage didn’t kill jobs, it killed job vacancies. Minimum wages overcome employer power to set wages at a level lower than would obtain if the labor market was more competitive.
Let me at last turn to your policy question. If minimum wages have only minimal effects on employment, why hasn’t the federal minimum wage been increased to $15? I heard senior Republicans oppose $15 in 2019, when I testified at a Congressional committee hearing on the bill. Their response to my testimony was that the minimum wage is socialism. Their ideology, in other words, was labeling as bad any government policies that help people and that give the government a good name. These Republicans were not primarily concerned with whether raising the minimum wage to $15 was good or bad for employers. Their response was driven by an anti-government ideology.
I was also surprised by the number of Republican senators who voted against raising the federal minimum wage, even those that represented states that already had a higher minimum wage than the border states around them. If the federal minimum wage went up, then employers in those other bordering states would have to raise wages, but the employers in the state that already had higher minimum wages would not. You might think the senators from those higher minimum wage states would support a policy that would reduce low wage competition from their bordering states. But many of these senators—all of them were Republicans—still voted against the bill. They were voting for their anti-government ideology, not the perceived economic interests of businesses in their state.
Gauss: Do you think that model makes more sense, whereby different states, counties, regions can tailor their minimum wages based on whatever their living standards might be like?
Reich: It’s a common argument that one size doesn’t fit all—because living costs vary so much across the U.S. We already have a system that allows for such differences: we have a federal floor and we allow states and localities to go higher; many—over 30 states and several dozen cities—have done so, some to $15. But it’s important to set the federal floor so that households in the poorest areas are able to make ends meet. If you look at the lowest cost of living counties—say in Mississippi and Alabama—the cost of housing and childcare have gone up so much that $15 does not allow a family to raise children and make ends meet. The federal floor needs to be a living wage for these families. $7.25 is not a living wage. An unregulated labor market will not generate a living wage.
Gauss: What about alternative solutions that’ve been seen outside the United States? What do you think about the efficacy of Denmark’s no minimum wage model, with unions instead driving living wages? Or Universal Basic Income or flexicurity? Do you think that those alternative models might be more successful than minimum wage in the United States?
Reich: It’s not an either or choice. Europe is really interesting because it contains a variety of models for setting wage floors. Some countries have a national statutory minimum wage; in others, unions negotiate different minimum wages in each industry (There’s no place that really has UBI in the sense that the UBI level by itself is close to a livable income). Denmark doesn’t have a statutory minimum wage. Its minimum wages are negotiated by unions in each industry at the industry level, not at the level of the individual firm, as is the case in the U.S. The minimum wage in these union agreements applies to everybody in the industry, even if you’re not a member of the union or if your workplace doesn’t have any union members. It’s called industry-level bargaining.
Germany, France, Italy, Norway and Sweden also have their own versions of industry-wide wage determination. When you have strong unions, these agreements cover 90+% of the workers. Then industry-level bargaining becomes a much better approach than a workplace by workplace approach. Strong unions will obtain better wage minimums and enforcement mechanisms and hold wage inequality in check. In Denmark, McDonald’s workers get paid more than $22 an hour and get six weeks of paid vacation each year. Yet the price of a Big Mac is about the same as in the U.S. But as I said, the Danish system is predicated on strong unions.
Gauss: Well, really since Reagan, union power has seriously declined in the United States. But what do you think about recent dynamics, like the localization of supply chains and expanded paid benefits for workers post-COVID? This isn’t representative of low wage labor markets to the same extent, but notably we had like the actors and writers strike in Hollywood. Do you think that these instances might represent growing labor power in the United States that could effectuate industry-level bargaining? Or do you think that we’re still a long way off of that?
Reich: The atmosphere for union organizing today is more positive than in many decades. Besides the successful actors’ and the writers’ strikes that you mentioned, there have been other major strikes, such as the graduate student strike at UC, strikes in the auto industry, in health care and among Starbucks baristas. In California, SEIU has obtained substantial industry-wide wage increases for fast food workers and for healthcare workers. So we are already seeing a lot of examples of growing union power.
But labor law is stacked heavily against unions who are trying to organize new members. The National Labor Relations Act gave workers the right to choose their own representatives. But there are no effective penalties against companies that violate the rights of workers who want to choose their own representatives. Indeed, in 1983 Ronald Reagan permanently fired striking air traffic employers, despite their right to strike. Since then, private sector employers have felt free to violate the law, for example, by permanently firing union organizers and strikers.
The law needs to change to create a level playing field. The analogous laws for public sector workers require their employers—school boards, public hospitals, the Postal Service—to be neutral about unionism. As a result, unionism in the public sector remains as strong as it was forty years ago, while private sector unionism has declined enormously.
Gauss: Interesting. To pivot from minimum wage, I’d love to hear your thoughts on the Social Structure of Accumulation theory. If you’d like, maybe you could talk about that a little bit. From the last piece that I read of yours, from like 2009, it was kind of forecasting that the 2008 recession might catalyze a “crisis phase” in the social structure of accumulation. Do you think the 2008 recession was an instance of that? Do you think that the COVID-19 pandemic was an instance of that, and what might the implications of a crisis phase be?
Reich: Let me first check your date real quick, because I don’t remember myself.
Gauss: I’m probably wrong, you’d know better than I do.
Reich: This book on my shelf says it was in 2010. You were essentially correct though, as it appeared right after the Great Financial Crash of 2008-09. Let’s dive into Social Structure of Accumulation (SSA) theory, which I helped develop in the 1970s and 1980s and have continued to work on since.
We originally developed SSA theory while examining the history of labor unions, wages and labor markets in the United States. We began with the recognition that postwar U.S. labor markets were highly segmented; there was not just one labor market, but three distinct ones. Low-wage jobs tended to be dead-end jobs with high turnover rates, clerical jobs had few advancement opportunities but did provide employment security, while professional and managerial jobs provided opportunities for career advancement within the same company or industry.We soon realized that labor market segmentation was not a defining feature of labor markets in the nineteenth century or the first third of the twentieth century.
To understand how segmentation came about, we studied the history of labor markets and then placed that history in the context of how other institutions in the economy had evolved. We noticed that major economic institutions did not change gradually, but in spurts. Moreover, the pace of institutional change seemed related to long periods of rapid economic growth that alternated with long periods of slow growth. Each of these periods lasted about 20 to 25 years. Interestingly, dramatic changes in economic and political institutions bunched just before each new period of rapid growth.
Beginning in 1876, for example, the U. entered a long slump period and then a crash in the so-called Panic of 1893. The crash and what to do about the economy was the defining issue of the presidential election of 1896. In this party realignment election a political coalition led by big business and bankers decisively defeated an alliance of populist farmers and industrial workers. The 1892 election represented a party realignment election—in the sense that the Republican Party dominated national politics for some time. Immediately after the election, the U.S. experienced a big merger wave—from 1897 to 1903. Huge monopolistic companies were formed in almost every major industry and economic growth restarted. The defeat of labor meant that workers shared only marginally in economic growth. Indeed, employers de-skilled the jobs of craft workers, leading to relative wage declines among skilled workers. The result was a more homogeneous labor market than had existed before.
The economic growth model of this period worked for a while, but was not sustainable. In particular, the growth of profits was much greater than the growth of wages, especially in the 1920s. The imbalance between the two, and the accompanying stock market speculation, led directly to the Great Depression.
The Depression ushered in another period of debate about what reforms were needed to restart economic growth, what the new institutions should be. On the one side, some said the depth and length of the Depression was the fault of Federal Reserve policy and America going off the gold standard. The other side argued that we need to put constraints on the banks and on monopolies, and to create a more inclusive model of economics. So Franklin Roosevelt originally came to office with an anti-monopoly reform program. Soon after, he shifted over to a program that regulated finance and gave workers and retirees a bigger share of the economic pie. Roosevelt’s New Deal was marked politically by the party realignment elections of 1932 and 1936,
After the Second World War, we had a new stage of capitalism, one that lasted into the 1970s and which spread prosperity more broadly than before. Unions became widely accepted in many industries and inequality was lower than in the previous stage. A larger role for government in managing the economy became the fabric that connected the disparate elements of the New Deal coalition, upheld not only by Democratic presidents but also by President Eisenhower.
Strains began to appear in the 1970s. Groups that were newly included in the workforce, but still disadvantaged—black workers, Latino workers and female workers—organized themselves and demanded greater shares of the growing economic pie. These demands, and those of powerful unions, placed strains on the economy, increasing inflation and reducing profits. So in the 1970s, the stage of capitalism that was characterized by a limited accord between management and some segments of labor entered a long period of crisis.
The 1980 election of Ronald Reagan represented the response of business and the so-called Reagan Democrats (mainly blue-collar workers) to the crisis of the 1970s. The election represented another realignment of American party politics, as we wrote in our 1982 book. This was no longer the era of big government, a sentiment shared both by Republican Presidents and President Bill Clinton.
Indeed, when Ronald Reagan became president, one of his first acts was to permanently fire all the striking air traffic controllers. Previously, workers on strike would be rehired after labor and management agreed on a settlement that ends the strike. Permanent firing of strikers essentially eliminated the right of workers to strike. Reagan’s actions signaled to many large corporations that the government would look the other way if they also openly opposed their own unions. Many then did exactly that.
The Reagan election thus represented the beginning of a new Social Structure of Accumulation, one with very weak unions, in which wages grew more slowly, and inequality soared to heights not seen since the 1920s. The Reagan coalitions also rolled back restrictions on mergers and deregulated major industries. The deregulation of finance began under President Clinton and continued under President Bush.This era has come to be called the era of neoliberalism.
The 2008-09 financial crisis resulted directly from the excesses of the era of neoliberalism. Since then, there’s been a lot of debate about how to move forward. We’ve been in a situation of very close elections and divided government, so we clearly haven’t entered a new social structure of accumulation.
It’s still possible that 2024 could be a realignment election, though. But a lot of the elements of the neoliberal ideology like free trade were dropped by both sides. So clearly, there’s something afoot that’s different. There are people who might point to industrial policy under Biden with the Inflation Reduction Act, the Chips Act, the American Recovery Act, and the Bipartisan Infrastructure Bill. Some argue it’s a new industrial policy, a new paradigm. I see that as a new way to make fiscal policy more popular and effective, but I don’t think it really moves the needle in terms of inequality.
To summarize, the SSA approach identifies different stages in American history. Each stage ended with a crisis, different from a normal recession. During the crisis period, competition develops among political coalitions, each with a different view of how to get out of the crisis and what reforms were needed. Eventually a single political coalition becomes dominant in a realignment election. It then solidifies its power through reforms that represent a new institutional model of economic growth. Each new social structure of accumulation requires a new vision of economic institutions and a new political coalition supporting the reforms. The new vision becomes institutionalized after a new realignment of the political party system.
Gauss: I guess you should take the opinion of a freshman with a grain of salt, but from what I can observe, there has been a post-2008 shift in social accumulation. At least from the U.S. perspective, there’s a retreat from globalization. You mentioned how both party platforms have moved away from free trade, COVID showed the dangers of global supply chains, and Biden has invested in domestic infrastructure and manufacturing. There’s an argument that maybe the market forces of globalization are inevitable, and governments can only do so much to address the inequalities that come with globalization. It might be something we’ll have to wait 20 years to see, but I think this trend could be an extrapolation from the post-2008 recession period.
It’s still a tug of war between these different ideologies though, and little has been resolved as yet.
Reich: That’s correct, we still have a very divided government. And your point about globalization is important.China will not always be a better place to invest, indeed, it isn’t right now. International capital has been getting out of China and going to Vietnam and Mexico.
Gauss: Exactly! Mexico is now our biggest trading partner. So those political forces are still at play—if the U.S. government disagrees with China ideologically, then it’ll take a stand economically.
Reich: I like to frame these conflicts in terms of political economy, because they are about political and economic power. I was taught a theory of international trade in which trade left both parties better off because they exercised their comparative advantage. In truth, trade doesn’t mean that there’s an absolute advantage to one firm; a large bloc of one side might actually be worse off than it was before. For instance, workers in a tradeable industry might be worse off, even if consumers were better off. We were taught that some of the gains would be used to compensate the losers, so everyone will be better off. It was an argument about Pareto improvement.
The fallacy in this theory is that the losers were rarely compensated and the compensation was often inadequate. For example, some free trade agreements included provisions for retraining displaced workers, but often there were no new jobs in the area—such as the industrial midwest.Often, the displaced workers were over 50 years old. For older manual workers, learning new skills was not easy, and it’s difficult emotionally when you’ve been in one industry for a long time to start all over again.
In Denmark, retraining is nearly mandatory and the government spends about thirty times—as a share of GDP—what we do on training. Workers then become more flexible over what jobs they hold, while retaining economic security. The Danes call this flexicurity. It’s a much better system than ours.
Gauss: It’s hard to make that adjustment just due to the disparity between growth and inequality. If I recall correctly, political scientist Dani Rodrik wrote about the openness mantra of international trade. His research on free trade found that for every $1 gained from efficiency, from that comparative advantage, there’s $20 of unequal distribution that was lost by some people. Is it possible to even really balance those forces, when the gap is that big?
Reich: Rodrik has a point. He refers to the context when institutions don’t change. The Social Structure of Accumulation theory argued that institutional reform can occur. But there must be a coalition that assembles parts of business, parts of labor, parts of other groups and brings them together with policies that benefit more people. people.
Gauss: That’s a perfect transition to my next question! You’ve written about the segmented labor market, which complicates that coalition and undermines class consciousness. So I’m curious where you think that idea stands today with changing technology, AI, and globalization. Do you think any of that might change some of the structural labor divisions between high wage and low wage labor? What are the consequences of any of these dynamics?
Reich: That’s a great question, but predicting the effects of AI is very difficult. AI is likely to displace much entry-level programming and much office work. More of the workforce may begin to experience economic insecurity.
The Democratic Party faces a major challenge in creating a coalition that brings high wage and low wage workers together.They have to appeal to people without more than a high school degree, the so-called working class, as well as people with college degrees who are professionals and more middle class. Both polls and voting studies show that the Democrats have gained a lot of support among college graduates, but they’ve lost support among working class voters.
The challenge is whether Democrats can continue to get the support of more educated workers and gain back the support of those who don’t have a college degree, who are still the majority of the workers and the voters in this country. It’s difficult to make them excited about being in the same coalition with each other.
Racial inequality is a big part of this, which is what I wrote about my dissertation. Liberal Democrats have had thoroughly good intentions to advance programs that would help black Americans. But unfortunately, these efforts were presented to whites in a zero sum framework, when in fact, many institutions, like the minimum wage or unionism or maintaining full employment, reduce racial wage and employment inequality while also helping most white workers.
Gauss: Or Affirmative Action?
Reich: Exactly, that’s a perfect example. In truth, these policies should not be thought of as a zero sum, but a system in which you can have gains for both groups. And there was another issue. Republicans since Nixon used race as a political tactic to gain votes of white workers, using terms that were like dog whistles.
Gauss: Nixon with the Southern Strategy?
Reich: Yes, exactly. Republicans criticized liberal governors for being soft on crime and supposedly letting black rapists out of jail and allowing criminals to immigrate. It’s an uphill battle.
You don’t have that kind of a problem in Canada, which has had more immigrants per capita in the last few decades than the U.S. But Canada has a policy of active integration of immigrants into the population, so they are not seen as a problem. In the U.S., conservative politicians present immigrants as a strain on the economy, taking away jobs from the native-born and creating burdens for social services and school systems.
Gauss: It’s interesting, you mentioned that this division on the basis of race is almost a uniquely American trend, whether it be economically or politically. Going back to economics, the neoclassical argument is that free market capitalism makes employers colorblind; employers just want to hire the best worker for the job. Why hasn’t that really taken effect?
Reich: Because employers have wage setting power. It’s only in a purely competitive model that competition erodes discrimination. If firms have wage-setting power, divisions among workers, including racial divisions, reduce worker bargaining and enhance firms’ power to keep wages lower for all their workers. This is the argument of my first book, Racial Inequality: a Political-Economic Perspective, which you can get as an ebook from Princeton University Press.
In the book, I go through all the economic theories of discrimination, which have not changed much since 1982. I develop a model in which stronger unions affect wages positively, but racial inequality weakens unions. When employers and/ or white workers treat black workers unfairly, unions are weaker and employers are better off. In this model, employers who discriminate obtain higher profits. Therefore competition among employers will not eliminate racial inequality.
But when black and white workers are united within the same union, or united to promote race-neutral public policies like full employment or minimum wages, both black and white workers will be better off and racial inequality will diminish. Racial justice and economic justice are interlaced with one another.
Gauss: Well, we’ve been discussing a lot of realistic but pessimistic dynamics in economics. On a more positive note, is there anything that you’re excited about in the future of economics?
Reich: I would point to two exciting developments. The first is that so much of the profession now recognizes that the effects of power on economic inequality should be at the center of economic inquiry. When I was a graduate student, the emphasis was on how markets were efficient; very few economists studied their effects on inequality.. We now recognize, especially at Berkeley, the role of economic power in creating inequality and the policies that can overcome it. I’m very happy that the Berkeley Economics Department is a recognized center for studies of the causes and cures of inequality.
Second, the methods available to economists to study economic questions have greatly improved. We can see this improvement both in macroeconomics and in microeconomics. When I was a graduate student in the late 1960s, macroeconomic theories and large-scale empirically-based macroeconomic forecasting models both showed that government policy could and did have positive effects on economic stability and growth. That was very exciting. However, the large-scale empirical models were better at forecasting in the short-term than at identifying the underlying relationships of macroeconomics. This meant that important questions, like who benefits from a tax cut, were not well understood. Fortunately, many younger macroeconomists have begun to use the causal identification tools of microeconomics (which I discuss below) to credibly identify causal macroeconomic relationships. This development will help put macroeconomics on a firmer theoretical foundation.
The state of microeconomics back then was totally different from today. Then it was dominated by purely theoretical research, much of which showed that perfectly competitive economies produced optimal outcomes. So social well-being and Pareto efficiency are maximized when markets are allowed to operate on their own—without taxes, subsidies, minimum wages, tariffs or any other government interventions. That view ignored how many markets are characterized by imperfect competition. And It left out key issues like how the distribution of wealth affects prices and the distribution of well-being, as well as how externalities like pollution create market failures that governments need to correct.
Microeconomics in those days was almost entirely about models, not about evidence for those models. You could read multiple issues of the American Economic Review and not come across an actual piece of empirical data. Empirical work was popular in some applied fields, such as industrial organization and labor economics, but econometric methods then had few convincing tests for distinguishing a correlation from a causal identification.
As one distinguished econometrician put it, there was a lot of “con” in econometrics. Fortunately, since then the profession has developed methods that credibly identify causal effects. The key innovation was to use quasi-natural experiments, a method Card pioneered in papers on immigration from Cuba to Miami and on the New Jersey/Pennsylvania minimum wage differences. As a result, more economists have turned to empirical work and are using the new causal methods.
The ability of economists to identify causal relationships represents a tremendous advance in the credibility and relevance of economic research. The search for and use of quasi-natural experiments has also led to a greater recognition of the importance of institutions in the economy and it has demonstrated the positive effects of many policies, such as minimum wages. The minimum wage research has also stimulated research on economic power. What are the conditions in which employers can set wages? Can unions, minimum wages and low unemployment rates overcome employer power that keeps wages low?
So economics is much more interesting now than it was when I started. I’m glad to have been a part of these changes, and I’ve been delighted to be in this economics department, which has been a powerhouse in bringing about these positive changes in our field.
Gauss: I love it, I really appreciate it! Thank you so much for taking the time to talk to me.
Reich: Of course!
Featured Image Source: Berkeley Department of Economics
Disclaimer: The views published in this journal are those of the individual authors or speakers and do not necessarily reflect the position or policy of Berkeley Economic Review staff, the Undergraduate Economics Association, the UC Berkeley Economics Department and faculty, or the University of California, Berkeley in general.
