Category: Governance

The rigid DNA of the European Central Bank

Posted on by Bridget Ansel

The slow-motion economic crisis in Greece finally accelerated into high gear this week, with the imposition of a bank holiday and capital controls following the withholding of another round of capital by the European Central Bank from Greece’s insolvent banks. This is what it looks like when a fixed exchange regime such as the euro fails. To date, the critique of the Eurozone as a system has mostly centered on the inability of an independent central bank to govern an economy as large as Europe’s without also having a European Treasury with tax and spending authority to back it up.

The problems inherent in the Eurozone, however, run deeper than a mismatch between the European Union’s wide monetary and narrow fiscal authority. They can be found in the DNA of the Eurozone. The European Central Bank’s institutional priorities and rules baked in from the start doomed the euro project to eventual failure in exactly this way. The relevant DNA emerged from a strand of the macroeconomic research born in the “neoclassical revolution” of the 1970s. That literature gave intellectual heft to the idea that a central bank should be run completely independent of any democratically-accountable political authority and that the only policy objective a central bank should uphold is to prevent inflation at all costs.

That is how the European Central bank is set up. Its governors are appointed from the central banks of the Eurozone member states and are not subject to any confirmation, as determined by a treaty that supersedes the domestic law of all European Union members. The same treaty requires that the central bank only care about “price stability,” meaning preventing inflation.

The most influential paper undergirding the euro is “Rules Rather Than Discretion: the Inconsistency of Optimal Plans,” published by economists Fynn Kydland of Norway and Edward Prescott of the United States in 1977 (see here for a lucid summary of their work). Their paper is very much a creature of the “stagflation” of the early 1970s. The key idea is that central bankers face a “Time Consistency Problem,” in which no policy they announce will ever be believed.

To understand Kydland and Prescott’s argument, first imagine that unemployment is negatively related to inflation. That is, higher inflation is associated with lower unemployment, and vice versa. Central bankers want to minimize both unemployment and inflation, so they face an inherent tradeoff. Now imagine that the economy exists for only two periods. In the first period, workers and firms form their expectation for inflation in the second period. If the central bank can convince workers and firms in the first period that inflation will be low in the second, then when that second period rolls around the central bank will have a strong incentive to generate surprise inflation in order to lower unemployment. But if workers and firms understand how the game is played, they’ll know a surprise is coming, so they cannot be induced to believe inflation will be low when they form expectations in the first period. Once workers and firms expect high inflation, if the central bank doesn’t deliver it in the second period, then unemployment will be high.

Because of these dynamics around expectations and trust, in the world of Kydland and Prescott’s simple model, an outcome with no inflation and low unemployment is unattainable. Any central banker who tries to reduce unemployment will only end up increasing inflation by inducing workers and firms to expect inflation, which the central banker then has to deliver to avoid unemployment.

To sidestep the Time Consistency Problem, Kydland and Prescott proposed what they called “rules” whereby the central bank should use a mathematical formula to dictate policy, eliminating their discretion. There’s no role for individual central bankers weighing options, hence no opportunity for them to try to fool workers, and hence no concern that workers will be afraid of being fooled and not trust an announced discretionary policy.

As several later writers noted, there’s no guarantee this mathematical formula would not itself be changed, so even mathematically-defined rules based on independent economic data may be just as bad, in practice, as discretion! Instead, as the Time Consistency literature evolved, the critical element in setting anti-inflationary policy is to appoint central bankers with strong reputations for fighting inflation. Everyone will believe them.

The result of this academic wrangling was a consensus in favor of endowing central bankers with no political accountability and tasking them to do whatever’s necessary to curtail inflation should it ever arise. By doing so, they assure inflation never arises in the first place. That is the view that was incorporated into the Maastricht Treaty of 1992 that set up the European Central Bank and gave it the notorious single mandate to reduce inflation, with no eye to mitigating recessions or unemployment.

Inflation has indeed been low in the Eurozone, including in countries such as Spain, Italy, and Greece where inflation had been famously high when these nations had their own currencies and autonomous central banks. But over this same period, inflation has also been low across developed economies, including those without independent central banks (such as the United Kingdom, where the move to independent policymaking only happened in 1997, to no discernible effect), and those with central banks charged with an explicit dual mandate for price stability and full employment (as in the United States).

The problem is that sometimes an economy needs inflation to function well. That’s true of Greece now—a massive debt overhang means that Greek businesses and households aren’t spending. Creditor-enforced austerity means the Greek government isn’t spending either. And an over-valued currency means that no one outside Greece wants to buy what the country is selling.

But the European Central Bank can never allow inflation because in the strange world of the Time Consistency Problem, that would reveal it to be irresponsible and secretly pro-inflation, ultimately leading to hyperinflationary expectations throughout the Eurozone. So despite the massive economic, social, and political cost, the European Central Bank insists on a prolonged depression rather than moderate inflation to work off Greece’s debt overhang. The idea that 25 percent of Greece should be unemployed and the other 75 percent should lose half their income in order to uphold the rigid diktat of an economics paper published in 1977 is not politically sustainable. Instead, Greece may well leave the euro. Either way, the Eurozone will leave its rigid DNA intact. Until the next crisis.

Bolstering the bottom by indexing the minimum wage to the median wage

Posted on by Casey Schoeneberger

The federal minimum wage today stands at $7.25 an hour, unchanged since 2009 despite rising prices and rising nominal wages for other workers. Without legislative action by Congress every year—a very difficult policy endeavor—the minimum wage for the nation will continue to stagnate. New legislation now before Congress seeks to overcome that perennial policy hurdle by proposing to index the minimum wage to the median wage—the exact middle point in the overall distribution of wages in the U.S. economy—after first raising it to $12 an hour in 2020.

Indexing the minimum wage to the median wage would automatically increase the minimum wage so that it keeps pace with the typical worker’s wage. Currently, 15 states and the District of Columbia index or have future plans to index the minimum wage to the annual rate of inflation, so that when prices rise each year the minimum wage rises accordingly. Indexing the minimum wage instead to the median is different because it links the minimum wage to overall conditions in the labor market rather than to the general level of prices. In this way, those earning the minimum wage experience annual wage gains according to overall demand for labor in the market rather than a less-direct measure of prices. Moreover, wage indexing improves the ability of the minimum wage to reduce inequality.

Indexing the minimum wage to the median is preferable to indexing it to the average wage. Raising the minimum wage would affect average wages, whereas pegging the minimum wage to the median wage would not. This issue brief explains all of these economic reasons for indexing the minimum wage to the median or typical worker’s wage, and shows what an indexed minimum wage would like over time.

View full PDF here alongside all endnotes

Indexing the minimum to median wage is good economics

Indexing the minimum wage to prevailing wage levels accomplishes two goals. First, indexing to wage levels increases the efficacy of the minimum wage as a policy tool to reduce wage inequality. In particular, wage indexing ensures those earning the minimum wage will not increasingly fall behind the typical worker.

Economic research on the minimum wage shows that between 1979 and 2012, more than 38 percent of the rise in inequality between the wage paid to the 10th percentile wage (the bottom ten percent of U.S. workers earn this wage or less) and the median wage is due to the minimum wage failing to keep up with the median wage. By indexing the minimum wage to the median wage, policymakers will help prevent widening disparities between those at the bottom and the middle of the wage distribution.

Second, wage indexing allows the minimum wage to rise in ways that the labor market can easily accommodate. Indexing the minimum wage to the general wage level means that roughly the same proportion of workers will earn the minimum wage year after year when the minimum wage rises. As long as underlying wage inequality does not change too much, fixing the distance between the minimum and median wage will keep constant the share of workers earning at or near the minimum wage.

What’s more, because a minimum wage increase will not alter the share of workers earning the minimum, employers will more easily adjust to regular increases in the minimum wage based on wage-indexing—as opposed to the irregular and larger increases typical of the current federal procedure, and many of the state and local procedures, for setting the minimum wage. Indexing to the median wage would require employers to raise wages for roughly the same proportion of their employees each year, whereas failing to index typically results in employers being required to raise wages for a much larger share of their workforces on less predictable basis.

What an indexed minimum wage would look like

Examining how the minimum wage would change over time if it were indexed to other measures of economic activity, such as prices or wages, is fairly straightforward. Immediately after the increase in the federal minimum wage back in 1996 and 1997, Congress could have indexed the new federal minimum wage of $5.15 an hour. Figure 1 shows how the minimum wage would have risen had it been indexed to the median wage or inflation from 1998 to 2014.

Figure 1

06XX15wage-indexing

Because Congress did not index the minimum to either prices or wages, the federal minimum remained unchanged for a decade before increasing in three successive increases in 2007, 2008, and 2009, to a level in between where it would have been if it had followed the path traced out by either indexing policy. The same figure also illustrates that the federal minimum wage has remained flat now for six years since the 2009 increase.

The median-wage indexed minimum wage is higher today than the minimum wage indexed to the Consumer Price Index because during the late 1990s and early 2000s nominal wages grew faster than inflation, resulting in real wage growth (after accounting for inflation). As a result, the median-index minimum wage would have been more than $8.25 in 2014. The inflation-indexed minimum wage would have been just over $7.60. Either way, the current federal minimum wage is lower than both indexed minimum-wage levels, standing at $7.25 an hour.

View full PDF here alongside all endnotes

We can also consider what the minimum wage would be had we indexed it to the median wage in 1968, which was the high point of the minimum wage relative to the median wage. In 1968, the minimum wage was more than 52 percent of the median wage of full-time workers, whereas in 2014 the minimum wage is about 37 percent of the full-time median wage.  If Congress had indexed the minimum wage to the median wage starting in 1968 than the minimum wage in 2014 would have been $10.21—more than 40 percent higher than the current minimum of $7.25.

Indexing the minimum to the median wage in 1998 or 1968 would have obtained substantially different minimum wages in 2014. The $10.21 minimum wage resulting from an increase in 1968 would have been almost 24 percent larger than the $8.26 that would have resulted from an increase in 1998. This underscores the importance of setting the appropriate level of the minimum wage before indexing it to the median wage. The minimum wage will only help a small portion of the workforce if it is set at a low fraction of the median wage and subsequently indexed. Wage indexing only maintains the position of the minimum wage relative to the typical wage, but indexing does not help set the initial level of the minimum wage.

By linking the minimum wage to the median wage, wage indexing keeps the minimum from falling to levels that many consider to be unfairly low or out of step with broader wage growth in the labor market. In addition, economists and political scientists alike recognize that economic fairness—and specifically the relationship between the minimum wage and the overall distribution of wages in the U.S. economy—is a major determinant of what the American public thinks is appropriate minimum wage policy.

There are precedents for wage indexing the minimum wage

Using the median wage as an index is natural to economists because they typically compare the minimum wage to the median wage in order to gauge the strength of the minimum wage. Where the minimum wage lies in the overall distribution of wages across the economy is central to contemporary economic theory. Academic research on so-called wage-spillover effects relies on comparisons of the minimum to the median wage. And when assessing the strength of minimum-wage policies across countries and across time periods, economists contrast national minimum-to-median wage ratios.

Keeping the minimum from slipping away from the typical wage also has policy precedents. In the run-up to increase minimum wages in the late 1980s and early 1990s, congressional bills included provisions to index the minimum wage to 50 percent of the average wage. And in the United Kingdom today there is an independent body called the Low Pay Commission, which advises the government on the appropriate annual minimum wage increase by factoring in the distance of the minimum wage to the to the median wage.

The median wage is the best wage to use as an index

To index the minimum wage to the general wage level, policymakers should use the median hourly wage instead of the average wage. The median wage is a good index because it is unaffected by the minimum wage. Minimum wages in the United States today cover less than ten percent of the workforce. When the minimum wage rises, it directly increases the wages of these low-paid workers. It also indirectly increases the wages of many of the workers who earn above minimum wage but still fall within the bottom 25 percent of wage earners, leaving the middle or median of the wage distribution unaffected.

This approach is better than using the average wage, or mean wage, as the peg for the index. If the minimum is indexed to the mean wage, when minimum-wage workers receive a raise, the average wage rises, which then increases minimum wages, and so on. Over time this process increases the share of the workforce earning the minimum wage, compelling employers to bear continually larger increases in labor costs.

In contrast, if the minimum is increased in line with the median wage, then the share of the workforce earning the minimum wage will remain roughly constant over time. This is because the median wage moves independently of the minimum wage. The benefit of keeping the minimum wage constant as a share of overall wages is that workers competing for low-wage jobs would find demand for their labor among employers equally constant.

In practice, the potential feedback effects from indexing to the average wage are small in a given year, but they may accumulate to economically meaningful sizes over time. Similar feedback effects would also be present in initiatives to index the minimum wage to the Consumer Price Index. If employers pass minimum wage increases onto their customers as price increases, then the minimum wage would indirectly affect the rate of inflation. These inflationary feedback effects, however, would be much smaller than feedback effects of indexing the minimum wage to the average wage because labor costs comprise only a part of the total costs of the production of goods and services.

The lack of any feedback effects from indexing the minimum wage to the median wage is yet another point in favor of this method of raising the minimum wage on an annual basis. Policymakers in Congress should seriously consider such legislation now in order to institute this new way of raising the minimum wage beginning in 2020.

View full PDF here alongside all endnotes

U.S. scholars need access to public and private big data

Posted on by Casey Schoeneberger

Big Data holds the promise of a wealth of information to uncover new insights into how our economy works but also the peril of exposing private information that could harm individual citizens. We all know that commercial ventures primarily use data gathered on their customers to track their purchases and spending habits—promising to varying degrees to protect such individual information—but now some private companies are allowing select scholars access to this information for research usage after the companies “anonymize” it.

One case in point is the JPMorgan Chase Institute, which last week unveiled its first report on the financial habits of retail banking customers at JPMorgan Chase & Co. The new research institute tapped into the commercial banking arm’s internal administrative data to determine how income and consumption fluctuate on a monthly and yearly basis. These findings will have important policy implications for lawmakers seeking to improve citizen’s financial well-being.

Researchers are constantly looking for new sources of information in order to answer the most challenging economic questions. But it is important to understand that by definition, JPMorgan Chase’s data can only tell us about their own customers. It cannot give us insight into the whole U.S. population—or even specific demographic groups. To create effective policies, we must gather information on all banking customers, not just those from one bank.

Still, researchers are flocking to private sources of data such as those released by JPMorgan Chase as well as credit-reporting companies. Yet the private sector is not the sole source of administrative data out there. Not by a long shot. The U.S. government holds tax records, school district filings, social security information—the list goes on—in order to administer its tax and benefit programs. Such recordkeeping has gone on for decades, but recent technological advances have made it easier to process these large datasets. Most importantly, government administrative data is representative of the entire population.

But because of perfectly reasonable privacy concerns this data is difficult to access, making a critical source of information—one that could allow us to investigate deep into our economy and provide better questions for policymakers to consider. But when handled correctly, these privacy concerns can be resolved. Those who are able to access the data have done amazing things. Work done using information from the U.S. Internal Revenue Service, for example, has transformed our understanding of the composition of incomes for those at the very top of the income ladder. And using administrative data, Harvard University economics professor Raj Chetty has repeatedly illustrated the extent to which your family and place of birth shape your success later on in life.

Such findings, however, are limited to the few scholars who have the means to gain access to this information. Even Chetty, the most well-known user of government data, must occasionally rely on European countries—many of whom have created secure data systems for researchers—to do his research on retirement subsidies, unemployment insurance, the effects of taxes on labor supply, among others.  Such research tells us a great deal about European economies and their labor markets, but cannot directly translate into usable information for policymakers in the United States—a tragedy for U.S. researchers and policymakers alike.

Because scholars do not have the necessary access to U.S. government data in the same way that European countries provide access, it is welcome indeed that researchers can turn to the private sector. But this is a temporary solution to a much bigger problem. Yes, privacy surrounding government data is an issue. At the same time, we tacitly allow private companies to track our information with little vocal apprehension. What companies find out about us—and in the case of banks, they find out quite a lot—can be used to answer important economic and behavioral questions, but also by firms seeking to expand their profits.

Without full access to public administrative data, U.S. researchers cannot explore Big Data in pursuit of meaningful research. And firms like JPMorgan Chase cannot completely fill that gap. We need both public and private sources of information and, right now, public access is far behind.

 

 

Is finance doing what it’s supposed to?

Posted on by Bridget Ansel

Seven years after the financial crisis and five years into the Dodd-Frank era of better supervision of the financial services industry, why is finance still controversial? Despite significant progress in reducing large financial institutions’ risk of failure and some financial abuses reined in around the edges, there’s ample evidence that much of finance is still detracting from rather than contributing to economic wellbeing.

How do we know? Because even as the global supply of capital soars to new heights, thanks to both expansionary monetary policy and excess private saving, corporate profits, particularly in finance, hit record levels, while average people are still paying a high price for borrowing. This paradoxical confluence of abundant capital for the well-connected and high corporate profits implies that corporations face little competition, because in theory abundant capital would make it easy for competitors or incumbents to expand their profitable operations, driving margins down. These same facts also suggest that what economists sometimes call the “real economy” hasn’t been cut in on the sweet deal available to banks, quasi-banks, and others with access to the privilege of cheap money. The result is a profusion of economic rents—unearned resource extraction by economic actors in a lucky position to profit from their advantage.

These were some of the conclusions from events held by the Roosevelt Institute last Wednesday to release a report called “Rewriting the Rules,” and by the Institute for New Economic Thinking on “Finance and Society” the week before. Both offered alternative interpretations of what the financial sector does, why it has become so large, powerful, and profitable, and what can or should be done to reform it without harming the economy as a whole.

Everyone from Federal Reserve Board chair Janet Yellen and International Monetary Fund managing director Christine Lagarde to Nobel Prize-winning economists Joseph Stiglitz and Robert Solow (as well as random ranters in the audience) noted that finance is a necessary feature of the economy. The sector provides liquidity and channels capital from savers to borrowers. So why were two major conferences held on the premise that something is wrong with a sector whose existence benefits us all?

Because, as most of both events’ participants argued, finance isn’t doing that job. The process of moving capital from savers to borrowers is inefficient and funds are actually flowing in the opposite direction—out of corporations and the real economy and into the hands of the wealthy, providing them with a healthy return on their savings at the expense of everyone paying high prices for loans, for telecommunications, housing, education, and other important products and services. Finance may even be shrinking the pie by redirecting human resources away from productive activities and toward strategizing new ways to divert the flow of cash to narrow private benefit.

That entire structural re-engineering of the economy is the fundamental driver of rising inequality at the top of the wealth and income ladder while everyone else is struggling to make a living in a slack labor market.

At the Institute for New Economic Thinking event, both Esther George, the President of the Federal Reserve Bank of Kansas City, and Claudia Buch, the Deputy President of the German Bundesbank, agreed that by supplying so much liquidity, central banks had in effect done all they could for society. But the rules of the economy, both written and unwritten, don’t automatically translate abundant, low-cost capital for financial institutions into gains for the real economy.

The idea that abundant capital would automatically benefit the rest of the economy is a part of the economics mythology of the “invisible hand”—that the free market will allocate resources–in this case capital–most productively, benefiting everyone. It’s a useful theory when it comes to defeating any challenge to the status quo, but it isn’t actually true. As Robert Solow said at the Roosevelt Institute’s event, we have enough evidence at this point to add a fifth universal element to the classical Greek four: “Bullshit.”

The Roosevelt Institute’s report is a good place to start when it comes to reforming the financial sector and the economy as a whole. But important as individual proposals are, a new narrative is emerging that rejects the false promise of a self-regulating, naturally welfare-promoting economy, with its gears greased by a large and powerful financial sector. There’s nothing natural or foreordained about the economy we inhabit, and past experience shows that meaningful progressive policies do not destroy the foundations of economic wellbeing, but rather create them. That doesn’t mean such reforms are easy to enact, but it does mean that it’s time the ideological walls protecting ever-increasing inequality were breached.

 

 

 

 

 

 

 

 

How raising the minimum wage ripples through the workforce

Posted on by David Evans

States and cities across the United States are increasing minimum wages within their jurisdictions, sparking other policymakers around the nation and on Capitol Hill to consider whether these changes affect the wages of all workers—not just those at the very bottom of the hourly pay scale who immediately benefit from a higher minimum wage. This question is especially timely this year, as 19 states have already increased the minimum hourly wage. And many cities are also boosting low-income workers’ pay, among them Oakland, CA, which increased its minimum wage to $12.25 an hour, and Seattle, WA, which now requires large employers to pay at least $11.00 an hour.

So how will these boosts in pay across the nation affect workers? In addition to minimum wage workers who receive a direct increase, which portions of the workforce receive indirect raises from a minimum wage increase? When our nation’s capital, Washington, DC, raises its minimum wage in July to $10.50, how will that affect the wages of workers who already earn $11.50?

This question is important for gathering a more complete understanding of the effects of raising the minimum wage beyond the lowest-paid workers. This issue brief explores the available economic research on these ripple effects, finding that increases in the minimum wage do raise the wages of those earning above the minimum wage. These ripple effects are critical to reducing wage inequality between those earning low- and middle-class wages.

Boosting Wages

New Equitable Growth book presents broad structural changes to boost U.S. wages for more equitable and sustainable economic growth

By Kate Bahn,

January 14, 2021

Coronavirus Recession

Why minimum wages are a critical tool for achieving racial justice in the U.S. labor market

By Ellora Derenoncourt, Claire Montialoux, Kate Bahn,

October 29, 2020

Boosting Wages

Strategic enforcement and co-enforcement of U.S. labor standards are needed to protect workers through the coronavirus recession

By Janice Fine, Daniel Galvin, Jenn Round, Hana Shepherd,

January 14, 2021

Although the minimum wages enhances the bargaining power of many low-wage workers, an increased minimum wage’s effectiveness in doing so dissipates as it spreads across the wage spectrum, essentially disappearing for middle-class wage earners. At the same time, assessing the exact impact of raising the minimum wage on specific earners may require higher-quality hourly wage data on all workers than is currently available in standard household surveys in the United States.

View full PDF here alongside all endnotes

What are minimum wage ripple effects and how do they occur?

In a recent study, Arindrajit Dube of the University of Massachusetts-Amherst, Laura Giuliano of the University of Miami, and Jonathan Leonard of the University of California-Berkeley find substantial evidence of a ripple effect in a large U.S. retailer’s pay policies. In 1996 and 1997, the federal government raised the minimum wage of $4.25 an hour in two steps to $4.75 and $5.15. The authors find that the large retail company, which was promised anonymity in order to provide data for study, raised its wages by 30 to 40 percent across its entire hourly workforce even though only 5 to 10 percent of this national firm’s employees earned less than the minimum wage.

There are good reasons to expect to see this same kind of ripple effect of raising the minimum wage more broadly in the U.S. labor market. In particular, economic theory suggests that increasing the minimum wage will raise the wages of other workers when employers need to compete for workers, as in some search-and-matching models of the labor market. Imagine all firms occupy rungs on a ladder, ranked by how well they pay their workers. After a minimum wage increase, the lowest paying firms raise their wage to the new minimum. This leads the next rungs of higher-paying firms to raise wages as well—to increase their ability to recruit and retain workers who would have better options elsewhere due to the minimum wage increase. The minimum wage then filters its way up the labor market, with ripple effects declining in influence further up the ladder.

Alternatively, workers may care about how they are paid relative to other workers at in their own workplace. After a minimum wage increase, will a supervisor be content with a wage similar to her now more highly paid staff? To the extent that employees are concerned about relative wages within a business, firms may raise wages in accordance with their institutional norms.

Whether ripple effects are largely market-mediated across firms or are instead based on relative pay concerns within the firm are open questions that get to the heart of wage-setting mechanisms in the labor market. The research by Dube, Giuliano, and Leonard on the large U.S. retailer suggests that within-firm pay concerns may matter a great deal because they affect how employees search for jobs. The retail industry famously boasts a high rate of employee turnover, and the authors find that workers quit their job significantly less often after minimum wage increases. This effect, however, largely occurs through relative pay concerns, such as when a worker receives a pay raise relative to her peers, she is far less likely to quit than if she had not received that relative increase in pay.

In addition to this one case study, economists find general evidence of these kinds of ripple effects from raising the U.S. minimum wage. The best estimates, though, appear in research conducted by economists David Autor of the Massachusetts Institute of Technology, Alan Manning of the London School of Economics, and Christopher Smith of the Federal Reserve Board. They study all state and federal minimum wage increases from 1979 through 2012, and measure the effect of the raises at each point of the wage distribution.

The authors find that the sharpest wage increases due to raising the minimum wage occur for workers at the bottom five percent of the wage scale, where U.S. minimum-wage workers are most likely to be concentrated. A ten percent increase in the minimum wage raises that 5th percentile wage by about 2.9 percent. The study also finds evidence of ripple effects as the minimum wage increases wages for workers who make more than the minimum—and that these ripple effects dissipate the further one moves up the wage ladder. The same ten percent minimum wage increase raises the wages of workers at the 10th percentile of wages by about 1.6 percent and raises the wages of those in the 20th percentile by a statistically significant 0.7 percent. After the 25th percentile, wage effects are typically very small and statistically indistinguishable from zero. (See Figure 1.)

Figure 1
How do ripple effects affect wage inequality?

A ripple effect for the bottom 20 percent of workers has important implications for wage inequality among workers in the United States. Over the 1979-2012 period studied by Autor, Manning, and Smith, the real (inflation-adjusted) value of the minimum wage fell and wage inequality increased, with those workers at the bottom 10 percent of the wage scale falling relative to the median wage by more than 22 percent. The authors estimate that the declining minimum wage during that period was responsible for nearly 39 percent of the increase in wage inequality between the typical worker at the middle of the wage spectrum and the worker at the bottom ten percent. Without ripple effects, the minimum wage may not have affected inequality at all because most minimum wage workers fall below the tenth percentile wage during the study period. (See Figure 2.)

Figure 2

Because women are generally paid less than men and therefore fall closer to the bottom of the wage spectrum, the minimum wage has larger effects on female wage inequality. For wage inequality among women, Autor, Manning, and Smith find that the minimum wage had particularly strong consequences. Between 1979 and 2012, the declining minimum wage was responsible for 48 percent of the increase in female wage inequality between the bottom and middle of the wage distribution. (See Figure 2.) This finding highlights that raising minimum wages in general disproportionately affects women. A female employee is more than 60 percent more likely to be a minimum wage worker than a male employee.

As significant as these ripple effects seem on their own and for the causes of wage inequality, Autor, Manning, and Smith themselves raise an important concern about these estimates: could the results simply be a product of survey measurement error? The authors rely on the best available source for U.S. wage data, the Current Population Survey of households, but misreported wage data in this survey poses a problem for distinguishing true ripple effects from fiction.

To understand why misreported data may skewer the findings about ripple effects, consider the current minimum wage of $7.25, which is roughly at the 4th percentile of wage earners. If the wage data contain substantial measurement errors, then some of these workers earning the minimum wage may misreport higher wages, perhaps reporting wages up to the 10th percentile. In that case, even if there were no ripple effects, raising the minimum wage above $7.25 will appear in the data as though it increased wages at the 10th percentile, even if that didn’t happen in reality.

Although measurement error in the U.S. survey data may complicate estimates of the size of the ripple effects of raising the minimum wage, better quality data suggests these ripples do exist. The recent study of a U.S. retailer by Dube, Giuliano, and Leonard, which used high-quality payroll data, is one case in point. Similarly, using employer-reported data in the United Kingdom that may be more accurate than U.S. household survey data, Richard Dickens of the University of Sussex and Alan Manning and Tim Butcher of the UK Low Pay Commission find that although the minimum wage only affected the bottom 5 percent of the wage distribution, ripple effects extended to the 25th percentile.

Conclusion

Both the theoretical and empirical research point to economically meaningful ripple effects from raising the minimum wage, although even the best measurements of the exact size of these effects in the United States are not completely certain. The evidence also seems clear that in the short run, minimum wages do not appear to have ripple effects for those workers earning middle-class wages or higher. In particular, the kinds of changes in the minimum wage that the United States experienced over the past three decades do not seem to affect the median wage or the wages of those at the top. Minimum wages, then, are an important piece of the policy toolkit affecting wage inequality and boosting stagnant wages at the bottom of the wage ladder. Improving middle-class wages will require other strategies.

—Ben Zipperer is a research economist at the Washington Center for Equitable Growth

How to bypass U.S. estate taxes

Posted on by Bridget Ansel

Tomorrow is tax day, which comes amid a burgeoning debate around our annual payments to Uncle Sam. One particular fight has broken out this week over the estate tax— which affects only those Americans with estates worth more than $5.43 million per person or $10.86 million per married couple. This means the estates of 99.85 percent of all Americans will not be subject to the estate tax.

What’s more, this debate on estate taxes misses one critical point—that a loophole allows many of those who are wealthy enough to face estate taxes to largely bypass the law altogether. This is done through the clever usage of a complicated tax-preferred savings vehicle called a Grantor Retained Annuity Trust, or GRAT, which those at the tippy top of the U.S. wealth and income ladder use to pass on their estate to heirs without it being subject to the full estate tax.

Here’s a basic description of how GRATs work. The first step for the very wealthy is to place a large amount of assets into a GRAT, with instructions that the entire amount should be returned to them over a specified period of time (two annual payments over two years, for example). Because individuals are not taxed when gifting to themselves, no taxes are levied on the original value of the assets placed in GRATs. Any earnings in excess of the assets originally placed in these GRATs accrue to trusts that are bequeathed to specified beneficiaries.

When the GRAT is first set up, the U.S. Internal Revenue Service gives a “gift value” estimate, based on the IRS “Section 7520 Code,” of how much they expect the assets to increase in value. Once the term of the GRAT expires, the wealthy individual who set up the GRAT, the grantor, will have received the original contribution plus this theoretical interest. In December 2014, the 7520 code was 2 percent. So if an individual created a two-year GRAT in that month and put in $1 million, they should receive back the original $1 million contribution and 2 percent interest via annuity payments through December 2016. Any excess appreciation earned beyond the original contribution and IRS assumed rate of return can be transferred to beneficiaries—estate tax free.

Not surprisingly, many savvy grantors will put assets in GRATS that have a good chance of increasing exponentially more than assumed rate of return set by the IRS. Many of the original Facebook founders, for example, put their pre-IPO Facebook stock into GRATs—and then saw their value increase well beyond the predicted IRS rate. Or consider gambling magnate Sheldon Adelson, who set up GRATs during the Great Recession of 2007-2009 using much of his Las Vegas Sands Corp. stock, which had plummeted in value. Because the stocks’ value rebounded as the U.S. economy recovered, he was able to shelter a half a billion dollars for his heirs, none of which will be taxed.

Even if the initial value of the GRAT does not increase to these degrees,  giving heirs $100,000, let’s say (which, relative to some GRATS, is actually small), grantors can continuously reinvest their money in GRATS again and again until they die. And doing so is court approved. A 2000 U.S. tax court ruling found that Audrey Walton’s (part of Walmart Stores, Inc.’s Walton family) GRAT was indeed legal as long as the grantor is alive. If grantors die during the term of their GRATs, all the assets are indeed subject to the inheritance tax (making a short-term, two-year GRAT a popular option among older grantors).

The Obama administration proposes to discourage such practices by requiring a 10-year minimum term on GRATs, but the probability of enacting any kind of restriction in the current political climate is slim. Very wealthy Americans are not obligated to report how much each GRAT passes on to heirs. But one estimate puts the amount that bypassed the estate tax at $100 billion since 2000.

The estate tax was originally enacted in 1916 in order to break up the oligarchic power base created during the Gilded Age of the late 19th century. The soaring wealth gap was not squashed altogether, although it was greatly diminished in part through the economic transformation instigated by World War II. Yet the estate tax has provided a relatively consistent stream of government revenue ever since. What’s troubling, though, is the loophole’s contribution to today’s widening wealth gap, which is at its highest point since the 1920s.

There is a valid debate to be had over the degree to which the estate tax can help alleviate rising wealth and income inequality in the United States, but it remains somewhat irrelevant if the premise on which that debate is based—the existence and enforcement of an estate tax for the wealthiest families—is far too easy to circumvent through the use of GRATs or other similar loopholes.

 

Determining the optimal U.S. tax rate for higher earners

Posted on by Nick Bunker

There are two constants in life: death and arguments about the optimal top marginal tax rate. The proper level of income taxation in the United States has been a hotly contested topic since the creation of the first federal income tax more than a century ago. The debate over the optimal tax rate has only intensified in recent years, as income and wealth inequality in the United States increases while taxes on the rich decline. Policymakers need an empirical answer to the question of the optimal level of taxation on top incomes.

How exactly do economists calculate an optimal level? Until very recently, economic research sought to determine the optimal rate by using just one concept—the highest tax rate that would maximize the amount of revenue collected, bearing in mind the disincentive to work created by taxation. Yet the most cutting-edge evidence tells us that our current estimates of the optimal tax rate are inaccurate because they’re missing important additional pieces of information about the behavioral response to taxes.

View full PDF here alongside all endnotes

So what is the optimal tax rate for top incomes? In order to determine that rate, policymakers should instead consider the following three ways that top earners might respond to tax changes:

  • by varying the supply of their own labor (working less)
  • by shifting between different types of income (wages and capital) to avoid taxes
  • by bargaining for different compensation levels from their employers

 

In this brief, we examine these three possible responses to higher taxes among the wealthy—responses that economists call elasticities—as posited by economists Thomas Piketty of the Paris School of Economics, Emmanuel Saez at the University of California-Berkeley, and Stefanie Stantcheva of Harvard University. Cutting to the chase, the three authors find that the optimal rate of taxation is much higher when we consider the responses quantified by three different elasticities as compared to one elasticity.

The analysis by Piketty, Saez, and Stantcheva finds that the optimal top tax rate is 83 percent. In contrast, the optimal rate using only one elasticity is 57 percent, which in turn compares to the current higher marginal tax in the United States of 39.6 percent. While 83 percent seems like a very high number, the underlying analysis of the paper is persuasive. Yet, the real take-away is the way the three authors calculate the much higher tax rate, and the importance of top earners bargaining for their compensation in calculating the optimal rate. U.S. policymakers need to understand the more complex responses of high earners to different tax rates. This new understanding is important given the country’s rising economic inequality and the relationship this rising inequality has to economic growth.

What is an elasticity?

Economists often seek to examine how responsive one economic variable will be to a change in another: What is known as an elasticity. Often they’ll explore the elasticity of Variable A to Variable B, such as the elasticity of employment to changes in the minimum wage or the elasticity of work hours to increases in the marginal tax rate. The resulting calculation of that elasticity would tell you how responsive the one variable is to changes in another. The larger the magnitude of the number, the larger the change.

In the case of the employment and the minimum wage, think of an elasticity of -0.1. This would mean a 10 percent increase in the minimum wage would result in a 1 percent decline in the level of employment. Or consider the elasticity of work hours to increases in the marginal tax rate, which economists calculate at -0.2—meaning a 10 percent increase in the marginal tax rate would result in a 2 percent decline in hours worked.

Obviously, many variables are at play in the complex U.S. economy. This is why factoring in other elasticities is important for economists to explore and for policymakers to understand.

A new approach to the problem

In their paper “Optimal Taxation of Top Labor Incomes: A Tale of Three Elasticities,” economists Piketty, Saez, and Stantcheva overturn conventional wisdom on optimal taxation by introducing empirical tests of three key ways that taxes can affect the behavior of top earners. The authors argue that the optimal tax rate for top earners isn’t the result of just one kind of response to taxation, but rather three different kind of behavioral responses, or three elasticities.

When the authors calculate the optimal tax rate using only the criteria that the rate not reduce the amount of time top earners spend working, they find that the optimal top tax rate would be 57 percent. But the authors argue that using just one elasticity misses out on too much that’s going on with the behavior of top earners when tax rates are changed. Instead, we should consider three elasticities.

Elasticity 1: Labor supply response

The labor supply of top earners is economic parlance for the amount of time wealthy individuals will put into work instead of leisure. To find the elasticity of their working hours to the tax rate  they pay, the authors first calculate the “supply side” elasticity, which measures the sensitivity of the labor supply of top earners to changes in the top tax rate. As the tax rate increases, individuals start to re-evaluate the trade-off between working and earning more money and not working and enjoying leisure time. If top earners were to stop working then the reduction in the labor supply would not only reduce the amount of taxes collected but, more importantly, could be harmful to economic growth.

That’s why policymakers need to know how responsive to taxation top earners are when it comes to their willingness to work. If they are very responsive, then the optimal tax rate could be lower, all other things being equal. If they are less responsive, the rate could be higher.

The authors calculate this first elasticity by looking at how both the share of income going to top earners in the United States, and U.S. economic output (measured by gross domestic product per person) changed as the top tax rate changed. They find that as the top tax rate went down between 1960 and the end of 2012 the top 1 percent of earners were able to keep more of their income but the growth of GDP per capita didn’t increase. That means this first elasticity is pretty low, at most 0.2.

In short, top earners do not substantially vary their labor supply in response to tax rates.

Elasticity 2: Tax avoidance

Another way that top earners can respond to taxation is to change the kind of income they receive so that they can avoid a higher tax rate. If the tax rate on labor income increases then top earners might shift their income toward capital income that is taxed differently. A chief executive officer at a big corporation, for example, might get his compensation shifted from purely salary to include stock options, which when exercised are treated as capital income. This doesn’t mean the top earners are changing their behavior. Rather they are trying to shelter their money from taxation.

A higher elasticity, or responsiveness, means that an increase in the tax rate would make the earner very likely to change the kind of income they earn. A low elasticity means that an earner would not change her source of income based on changes in the tax rate. In a situation where this elasticity is high, top earners will simply shift all their income away from labor and toward capital—so a high tax on the labor income of top earners would yield little revenue.

To calculate this elasticity, the authors compare the trends in the share of labor income going to the top 1 percent to the trends in the share of income that includes capital gains. What they find is that the trends of the two data series are nearly identical. In fact, Piketty, Saez, and Stantcheva say their estimate for the second elasticity is 0. Top earners in the United States do not tend to shift their income sources in response to changes in the tax rates.

Elasticity 3: Executive compensation bargaining

Many top earners are corporate executives. According to one estimate, 41 percent of the top 0.1 percent of income earners are executives, managers, or supervisors. A growing body of research demonstrates that corporate CEOs and other members of the corporate “C suite” (chief financial officers, chief information officers, chief operating officers, and the like) are not  responsible for all the gains in the company’s economic performance for which they are compensated. In other words, a substantial share of top corporate executives’ earnings are comprised of funds from the firm that might otherwise go to other workers, investments in the firm, or to shareholders.

When tax rates are lower, executives have a stronger incentive to bargain for higher compensation. And since this compensation isn’t necessarily due to higher productivity, the struggle is zero-sum—if executives receive compensation for productivity gains they aren’t responsible for then funds that would go toward other ends get diverted. Piketty, Saez, and Stantcheva explain that a higher elasticity means that executives are more likely to bargain for higher pay when tax rates are lower and therefore receive funds that might go elsewhere within the firm.

In order to calculate this elasticity, the economists look at international data to determine the relationship between top tax rates and CEO pay. The data show that in countries with lower tax rates, CEOs have higher average incomes after accounting for the kinds of industries in which their companies compete. Piketty, Saez, and Stantcheva interpret this finding as the sign of a high third elasticity, which they calculate conservatively at 0.3 at the lowest. This would mean a higher tax rate on top earners would reduce what economists call rent-seeking—the taking of undue compensation—within the firm, potentially increasing wages for average workers.

Conclusion: The optimal rate?

The research presented in “Optimal Taxation of Top Labor Incomes: A Tale of Three Elasticities,” by economists Piketty, Saez, and Stantcheva, suggests that the conventional wisdom around optimal taxation rates for top earners is missing some nuance in how top earners respond to taxation. Including the bigger picture would seem to leave substantial room for an increase in rates on top earners. Piketty, Saez, and Stantcheva’s calculate that the optimal top tax rate comes out to 83 percent, once their three elasticities—labor supply, tax avoidance, and bargaining—are combined.

Compare that result to the result of 57 percent when economists only consider the overall elasticity of income to tax rates. This level is also much higher than the current top federal income tax rate of 39.6. This isn’t to say that our current top tax bracket should be raised to 83 percent tomorrow. Rather, this is the optimal rate for those at the very top of the income ladder. The top tax bracket would have to be changed in order to tax only those individuals or households at the very tippy top at this new 83 percent rate.

The results of this research also indicate that the rise in income inequality at the very top of the income spectrum was driven primarily by the decline in tax rates, which allowed top earners to get higher incomes without increasing the pace of economic growth. So the main take-away from latest research is clear: Tax rates in the United States on incomes at the very top could be much higher without affecting output growth and potentially boost wages for average workers.

What Do Americans Think Should Be Done About Inequality?

Posted on by Casey Schoeneberger

A new online survey of some 10,000 Americans’ reaction to growing income inequality offers novel insight into public perceptions of inequality and what—if anything— should be done about it. The survey first presents some respondents with information about the extent of inequality—for example, by displaying how much more income a respondent would earn if increases in economic growth since 1980 had been more evenly distributed—and then assesses their attitudes toward inequality and policies aimed at ameliorating gaps between rich and poor, compared to other respondents who did not see the information. The survey shows that while respondents who view information about inequality are more likely to believe that inequality is a serious problem, they show no more appetite for many government interventions to reduce inequality— with the notable exceptions of increasing the estate tax and the minimum wage.

View full pdf here alongside all endnotes.

Our working hypothesis is that those surveyed alighted on the estate tax because it applies to many fewer Americans than respondents had assumed. And respondents favored increasing the minimum wage because doing so does not necessitate heavy government involvement (unlike, say, the Supplemental Nutrition Assistance Program, or food stamps for low-income Americans). The survey reveals a deep mistrust of the federal government’s ability to administer programs effectively and efficiently even after confronted with the importance of these programs in alleviating poverty among those Americans at the bottom of the ladder.

There are a number of nuances, of course, to the findings from the survey, which are detailed in our forthcoming paper, “How Elastic Are Preferences for Redistribution? Evidence from Randomized Survey Experiments.” Our conclusions bear directly on public policy debates in Washington, D.C. and in statehouses across the country as the U.S. public grapples with what, if anything, to do about a wealth and income gap now as wide as just before the onset of the Great Depression in 1929 (See Figure 1.)

Figure 1

inequality-surv-webart1

The survey

Over 10,000 respondents completed the surveys we designed for this project. The mix of respondents gives us some confidence that the results we find would mirror the attitudes of typical Americans. While online surveys do disproportionately draw from certain groups such as younger adults, our sample compares favorably with both the CBS News election survey from 2011 and the Rand Corporation’s American Life Panel online survey.

For our study, respondents were randomly assigned to a treatment group who viewed a short online presentation conveying information about income inequality, or a control group who did not view this presentation. This customized approach was made possible by an online platform that enabled us to gather detailed income data on the respondents and in turn inform them interactively about where they fell in the U.S. income distribution.

Respondents were also asked to self-report their political preferences using a five-point scale, from very liberal to very conservative. Then, both treatment and control groups answered a series of questions about their views on inequality and which policies, if any, they favored to address it. We call the difference between the percent of liberal and the percent of conservative control group respondents agreeing on these various issues the “political gap”—and we examine how our treatment might “close the gap” between liberals and conservatives on these various issues.

The findings

There are several novel findings that emerge from our survey. When respondents are given the actual data on the growing income gap in the United States, their concern about the problem increases by a staggering 35 percent—an effect equal in size to roughly 36 percent of the liberal-conservative gap on this question. Moreover, viewing information about inequality also significantly influences attitudes toward two redistributive policies: the estate tax and the minimum wage (See Figure 2).

Figure 2

Figure 2

 

When respondents in the treatment group learn the small share of estates subject to the estate tax (roughly one in 1,000), they support increasing it at three times the rate of the control group—akin to cutting the political gap in half (See Figure 3). This finding is mirrored in a recent study by political scientist John Sides of George Washington University, who finds that accurate information on the small number of families subject to the estate tax substantially reduces support for repealing the tax.

Similarly, after reviewing the presentation on income inequality, support for raising the minimum wage jumped by 4 percent (from an already high baseline of support of 69 percent) in the treatment group relative to the control group, sufficient to close the political gap by about 10 percent (See Figure 3). (The federal minimum wage now stands at $7.25 an hour; 28 states and the District of Columbia boast slightly higher minimum wages alongside other several cities).

Figure 3

inequality-survey-webart3

 

At the same time, attitudes toward some policies were unaffected, including increasing top income tax rates and support for the Earned Income Tax Credit for low-wage workers and the Supplemental Nutrition Assistance Program—more commonly known as the food stamps program—which on average provides $150 a month toward food purchases for eligible recipients.

Importantly, our results also suggest that this aversion to government intervention is due to a deep level of distrust in government. In a sense, respondents who have learned the role of government in creating the current level of inequality seem to be telling us they do not trust that government is also the entity to address the problem.

The policy implications

This last finding is, to our knowledge, the first direct evidence of the causal effects of trust in government on redistributive policy preferences. Our findings highlight the potential role of mistrust in government in limiting enthusiasm among the general public to certain kinds of government policy programs—such as the Supplemental Nutrition Assistance Program and the Earned Income Tax Credit—designed to help close the wealth and income gap.

Research into the connection between mistrust of government and policy preferences is only just beginning. For instance, economists Paola Sapienza at Northwestern University’s Kellogg School of Management and Luigi Zingales at the University of Chicago’s Booth School of Business find that Americans support higher auto fuel standards over a carbon tax-and-rebate program because they do not trust the government to in fact rebate the tax. Given that by most measures, Americans trust in government is at record lows, future work on its consequences would be welcome.

Finally, while beyond the scope of our paper, our results do point to an intriguing possibility: that the rise in inequality may have in fact led to the rise of distrust in government. If such a connection existed, then inequality may in fact be self-reinforcing—decreasing trust in government and undercutting support for the very policies aimed to reduce inequality.  We look forward to future work on the possible connections between inequality, trust in government, and support for redistribution.

View full pdf here.

—Ilyana Kuziemko is an economics professor at Princeton University, Michael Norton is a professor of business administration and marketing at Harvard Business School, Emmanuel Saez is an economics professor at the University of California-Berkeley, and Stefanie Stantcheva is an economist at Harvard University.

The links between institutions and shared growth

Posted on by Bridget Ansel

Increasing inequality in the United States and its relationship to economic growth is getting a lot of attention lately. It is now clear that sharply rising inequality is not necessary for good economic performance and, indeed, growing evidence suggests that high and rising inequality is harmful, especially if the mechanisms generating the rise in incomes at the top of the ladder contribute to stagnant and falling incomes for the rest of us.

What matters most for the vast majority of American households is not inequality per se. Rather, individuals are mainly concerned with generating a higher standard of living via better wages and employment outcomes. Too many households confront a labor market in which it is increasingly difficult to find living-wage jobs. Despite substantial national productivity growth in the last 30 years, the average pre-tax income for the bottom 90 percent of American households was lower in 2013 than in 1980, and the wage for full-time male college graduates has not grown since the late 1990s.

This wage stagnation stands in contrast to the years following World War II until the late 1970s, when worker compensation grew in line with labor productivity. It’s true that compensation-productivity gaps have also appeared in other rich countries, but with one crucial difference—outside the United States this gap has increased despite substantial growth in worker compensation. While U.S. manufacturing workers have suffered from technology and global competition, that‘s also the case for manufacturing workers in all rich countries. For example, I’ve shown that American manufacturing workers experienced a paltry 4 percent wage increase between 1980 and 2013 (after accounting for inflation) while manufacturing compensation increased by 42 percent in Germany, 43 percent in France, and 56 percent in Sweden.

Many argue that the United States makes up for poor compensation performance with strong employment growth. Yet over the past few decades, U.S. employment rates have not only been lower than that of many European countries, but also have fallen dramatically in recent years—especially for men. In addition, while U.S. employment and GDP grew in tandem between the late 1940s and the mid-1990s, a yawning gap has developed since then, with GDP continuing upwards and employment flattening, despite a growing population.

In short, since the 1980s, U.S. economic growth failed to produce enough jobs, and equally important, enough “decent jobs, ” which I defined as those paying adequate wages with adequate hours of work.  Through my research—funded in part by a 2014 grant from the Washington Center for Equitable Growth—I am seeking to uncover the conditions and mechanisms through which economic growth gets translated into a sufficient numbers of decent jobs. More specifically, I aim to identify the institutions and public policies that help explain best practices for the creation of decent jobs.

What happened to shared growth? Most economists continue to explain the explosion of earnings inequality with conventional supply-and-demand stories, in which worker compensation is believed to accurately reflect the contribution workers make to production. Thus, in this view, CEOs and financiers have received skyrocketing salaries, especially since the mid-1990s, because they are now contributing dramatically more to their firms and to the economy as a whole.

Similarly, the bottom 90 percent have seen stagnant and falling wages because they’ve fallen behind in the “race between education and technology.” The computerization of the workplace requires greater cognitive skills, but workers have not kept up, as indicated by the slowdown in college graduation rates. Assuming (nearly) perfectly competitive markets, the explosion in wage inequality in this view must reflect a similarly explosive increase in skill mismatch (too many low skill workers, too few high skill ones).

Such arguments leave little or no room for labor market institutions and public policies in determining changes in the distribution of earnings up and down the income ladder. An alternative view is that institutionally-driven bargaining power is a critical piece of the story, whether it is the noncompetitive “rents” earned by top managers and financiers, or the collapsing power of hourly wage employees. As Thomas Piketty argues in “Capital in the Twenty-First Century:”

In order to understand the dynamics of wage inequality we must introduce other factors, such as the institutions and rules that govern the operations of the labor market in each society [and explain] the diversity of wage distributions we observe in different countries at different times.

All rich countries face challenges from technology and globalization, but only the United States and the United Kingdom show inequality rising to extreme levels.

In order to understand wage inequality and unshared productivity growth in the United States, we must take a much closer look at the ways in which institutions affect labor market outcomes. In my forthcoming research, I will compare the United States with Canada, Australia, Germany, and France to answer questions such as:

  • How does the distribution and growth of decent jobs in these countries compare by economic sector, occupation, and demographic group?
  • To what extent can these outcomes be attributed to the effects of differing institutions across the five countries?

 

Underlying this research design is the view that institutions matter a great deal for market outcomes. The United States can do a better job of generating decent jobs, and a sensible first step is to learn from the experiences of other countries.

 

—David Howell is a professor of economics and public policy at The New School in New York City

 

The future of work in the second machine age is up to us

Posted on by Nick Bunker

“Big Thinkers” about the role of technology in the U.S. economy are roughly divided into two camps when it comes to the consequences of rapid technological change on the U.S. workforce. There is the techno-optimist view that better technology complements workers and hence benefits them by raising wages. And there’s the pessimistic view that better technology substitutes for workers and therefore displaces and harms them. A debate between the two views was probably what the organizers intended for an event last week hosted by The Brookings Institution’s Hamilton Project entitled “The Future of Work in the Age of the Machine.”

The impetus for the forum was the influential 2014 book “The Second Machine Age” by professors Erik Brynnjolfsson and Andrew McAfee at the Massachusetts Institute of Technology. The authors argue that increasingly “smart” technology displaces workers by reducing the range of tasks that require human ingenuity, and by enabling economic arrangements such as off-shoring that rely on instantaneous global communication and replicability. Brynnjolfsson and McAfee are clearly in the pessimists’ camp.

Until recently, economists were largely in the optimist camp. Sure, some jobs—think buggy whip manufacturers, typists, or travel agents—might disappear, but others would arise to take their place. In the long run, increased productivity would benefit everyone in the form of higher wages.

Yet the debate last week actually highlighted a third position. If either the techno-optimists or the techno-pessimists are right, then we should see a major positive impact on worker productivity. But it just isn’t there in the data. If anything, the rate of technological change in the United States has decreased since at least 2003, specifically in the technology sectors widely thought to be most innovative.

In contrast, we definitely see worker displacement, stagnant earnings, a failing job ladder, rising inequality at the top, “over-education” (workers taking jobs for which they’re historically overqualified), and declining rates of employment-to-population and household and small business formation. What we do not see are the productivity gains, either on a micro or macro level, that are supposedly driving worker displacement. (See Figure 1.)

Figure 1

 

fernald-graphic

Former Treasury Secretary Larry Summers made this point forcefully at the Hamilton Project event. He said “people see there’s already a lot of disemployment but not a lot of productivity growth.” And he continued by asserting that “the core problem is that there aren’t enough jobs,” and that it’s hard to believe the future promise of labor-supplanting technology is driving current displacement. The reason, he said, is that we’d expect to see the installment of new labor-saving systems that would cause a temporary increase in labor demand during the transition.

Summers noted that back when he was an undergraduate at MIT in the 1960s, his professors said labor would not be displaced by technology. In those days, the non-employment rate for prime-age male workers was 6 percent. Now it’s 16 percent. Summers’ co-panelist David Autor added that since 2000, the education wage premium has reached a plateau and the rate of over-education has increased, both of which are hard to square with the argument that the reason for rising inequality is the advance of technology. Summers added that the idea that more education solves the problem of displaced labor is “fundamentally an evasion.” Summers’ arguments and Autor’s observation imply that if we’re wondering how things got so bad for workers, it’s not because we live in the Second Machine Age.

So if not technology, what explains labor displacement?

Broadly speaking, the explanation is this: market practices and public policies that favor managers over workers, and those who make their living by owning capital over those who make their living by earning wages. That choice lurks behind the decline in full employment as a priority in macroeconomic policymaking. It’s also behind a shift in the legal standards, mores, and incentives of corporate management in favor of the interests of owners over other stakeholders. That choice is also evident in the abandonment of long-term productive investment as a priority in public budgeting in favor of upper-income tax breaks and retirement programs for the elderly.

As Summers noted at the Hamilton Project’s event, there seems to be a lot of so-called rents—economics speak for excessive payment for something beyond its actual value—in corporate profits that can’t be understood as the fruits of productive investment. The big question is: who gets those rents? In 1988, Summers wrote an article fleshing out the idea that the division of rents between corporate stakeholders is what drives rising inequality. More than a quarter century later, he could not have been more prescient.

The good news is that if such a profound shift played out over only three or four decades, then it’s reversible. That wouldn’t be true if it were the result of the technological trends detailed in “The Second Machine Age.” So what should be the focus of public policy is to figure out ways for workers to accrue more of corporate earnings, for more unemployed and underemployed people to find full-time, productive jobs, and for the broader economy to serve the interests of the actual people who inhabit it—those who overwhelmingly derive their living from their labor.

We know what needs to be done and how to do it, because we’ve done it before. (See Figure 2.) But it’s a lot harder to actually do than doubling the number of logic gates on a computer chip every two years—the ostensible tech explanation for our current economic woes.

Figure 2

incomegrowth-quintile1