Understanding economic inequality and growth at the bottom of the income ladder

Today the official poverty rate in the United States is back to levels we haven’t seen in 20 years, and the incomes of families at or near the bottom of the income ladder are at the same level they were in the early 1970s. Studies show poverty rates on the decline since the beginning of Great Society programs in the 1960s until the late 1990s, but seeing as wages have not improved, this decrease in poverty was almost entirely due to increased government transfers.

Poverty is back up because wages at the bottom have stagnated or fallen. Over the past 40 years, workers in the bottom 40 percent of the wage spectrum experienced negligible wage growth, and wages have fallen for those in the bottom 10 percent, after accounting for inflation. The trends have been worse for men than women, in no small part because women’s increased educational attainment and on-the-job experience have boosted their wages over the past few decades.

Researchers find that the lack of wage-and-income growth for families at the bottom of the income ladder in particular results in serious economic consequences. First, the continued lack of income growth harms low-income children’s development, which affects our nation’s future human capital. Second, a growing body of evidence suggests that the lack of income gains at the bottom have macroeconomic consequences because it either reduces consumption or encourages more debt, both of which are destabilizing.

But there are remedies for these problems, such as raising the minimum wage. Research by economists Daniel Aaronson and Eric French at the Federal Reserve Bank of Chicago and Sumit Agarwal of the University of Singapore find that increasing the minimum wage boosts the consumption of affected workers.And a battery of other research shows that raising the minimum wage does not reduce local employment and reduces employee turnover.

The three essays in this section of our conference report—by Christopher Wimer, a research scientist at Columbia Population Research Center, Arindrajit Dube, an associate professor of economics at the University of Massachusetts-Amherst, and Gavin Kelley, chief executive of the Resolution Foundation in the United Kingdom—look the overall exclusion of low-wage workers from the benefits of economic growth and how that affects the future growth and stability of our economy. They also consider whether the government should focus on raising market wages though policies such as the minimum wage, anti-poverty assistance or some better combination of the two approaches.

Reversing inequality at the bottom: the role of the minimum wage

by Arindrajit Dube

There are many factors affecting the growth in wage inequality in the United States over the past four decades. When it comes to workers on the bottom rungs of the income ladder, one important factor is the minimum wage.

The federal minimum wage reached its high-water mark in 1968, when it stood at $9.59 per hour in 2014 dollars, declining to a still-respectable $8.59 by 1979. During the 1980s, however, the real (inflation-adjusted) minimum wage declined substantially. And over the past 20 years, the minimum wage has largely treaded water, reaching a historical low of $6.07 per hour in 2006 just before the last federal increase in 2009. The minimum wage now stands at $7.25 per hour in today’s dollars.

The failure of the minimum wage to keep up with inflation means that, for workers earning the minimum wage, each hour of labor purchases less goods and services today than it did in the past.

Minimum wage workers are not only (contrary to popular belief) teenagers and young adults whose low wages are supplemented by their families. In fact, between 1979 and 2011, the share of low-wage workers—defined as those with wages of $10 or less in 2011 dollars—under the age of 25 years of age fell to 35.7 percent from 47.1 percent. Instead, minimum wage workers are increasingly adults who must rely exclusively on their meager earnings to support basic household consumption. The decline in the value of the minimum wage affects female workers in particular, as they tend to be paid lower wages.

Low minimum wages are also problematic when they deviate too far from the median wage because that means minimum-wage earners are falling farther behind on the income ladder. This is why economists often use the ratio of the minimum to the median wage. The so-called 50/10 wage gap—the median wage earner compared to those with earnings in the bottom 10 percent of the income ladder— captures this type of wage inequality over time. Since 1979, around a third of the changes in the 50/10 wage gap have been driven by changes in the minimum wage.

There are two main reasons to pay attention to this measure. First, a comparison of the minimum wage to the median offers us a guide to how many workers are affected by a particular minimum wage increase, and what level of minimum wage the labor market can bear. When this ratio is low—say around 0.2—the policy is not raising wages of many workers. In contrast, a high ratio—say around 0.8—indicates a highly interventionist policy where the minimum wage is dramatically compressing differences in wages for nearly half the workforce.

Second, the median wage provides a reference point for judging what is a reasonable minimum wage level. No one expects that the minimum wage should be set equal to the median wage, but fairness concerns matter when the minimum wage falls below say, one-fourth or one-fifth of the median wage.

A natural target is to set the federal minimum wage to half of the median wage for full-time workers. This target has important precedence historically in the United States. In the 1960s, this ratio was 51 percent, reaching a high of 55 percent in 1968. Averaged over the 1960–1979 period, the ratio stood at 48 percent. Today, the ratio stands at 38 percent. Raising the federal minimum wage to around $10/hour would restore the value of the minimum to around half of the median full-time wage, yet efforts at raising the minimum wage have largely stalled in a deeply divided Congress despite widespread political support around the country.

This federal inaction has led to a flurry of activities at the state and local level. States have stepped in during periods with a stagnant federal minimum wage in the past, especially the 2000s, but for the first time in U.S. history we have many major cities establishing citywide minimum wages for all (or most) private-sector workers. The growing list of cities with such a policy now includes Albuquerque, Chicago, San Francisco, San Diego, San Jose, Santa Fe, Seattle, and Washington, DC. Other cities such as Los Angeles and New York are actively exploring possibilities.

This push to increase minimum wages in big cities coincides with organizing by workers in fast-food chains in major metro areas. The target minimum wage in most of these areas is substantially higher in nominal (non-inflation-adjusted) value— with $15/hour a focal point for these campaigns. The confluence of these factors raises the possibility of substantially altering wage standards in the U.S. labor market.

How should we think about these sizable increases in the minimum wage? First, we should be careful not to overstate the size of the increases or the levels of the minimum wages because the cost of living and overall wage levels vary tremendously by region. Setting the minimum wage to half the full-time median wage would produce $10/hour policy nationally, but much higher figures in major metro areas such as Washington, DC ($13.51), San Francisco ($13.37), Boston ($12.85), New York ($12.25), and Seattle ($11.85).

Moreover, these higher nominal wages are usually phased in gradually. In Seattle, the hourly minimum wage will eventually rise to around $14 in 2014 dollars. This constitutes around 59 percent of the median full-time wage in that metro area, which is certainly higher than historical standards but not outlandishly so.

So what we do know about the impact of minimum wages over the past few decades and the importance of particular channels for the higher, local wage standards? First, most careful recent work points to relatively small impact on employment—be it for sectors such as restaurants or retail or for groups such as teens.3 As a result of wage increases and small impact on employment, family incomes rise at the bottom. A 10 percent increase in the minimum would reduce the poverty rate among the non-elderly population by around 2 percent, and generally raises family incomes for the bottom 20 percent of the family income distribution.

It is possible that the much larger increases in minimum wage may induce greater substitution of low-skilled labor with automation, or with fewer but more high-skilled workers? If this is true then we would expect evidence of growing “disem­ployment” (workers out of a job due to lack of skills or education) from these higher city-wide wage standards. Yet recent research also identifies some additional benefits that may be more important than larger wage increases. A growing body of research shows that while the impact on employment stock is small, there are larger reductions in employment flows or turnover. The reduction in turnover provides additional evidence that search frictions in the low-wage labor market are quantitatively important and offer some clues as to the way cost increases may be absorbed.

Given the cost of recruiting and training new workers, for example, reduction in turnover can be expected to offset about a fifth of the labor-cost increases associated with minimum wage hikes in this range. I think the large city-wide increases will provide us with some additional evidence on this topic. In particular, I believe it should be possible to assess whether the lower turnover regimes lead to substantially different training policies as would be predicted by some models incorporating “search friction”—things that prevent or make it more difficult for workers to find the kind of jobs they want. Moreover, it will be interesting to see whether change comes from the extensive margin (growth in high-training/low-turnover firms) or the intensive margin (change within firms).

The nature of high-cost metro areas means that a substantially higher minimum wage may allow more lower-wage workers to live closer to their place of work (inside the city) and reduce commute time. The labor-supply effect from this “in-migration” also can reduce recruitment costs and improve the quality of the service work force. These additional channels will be useful to keep in mind in future research.

Evidence also suggests that, in part, cost increases associated with a higher minimum wage are passed on to customers as price increases, especially for industries that employ high levels of low-wage labor. The best evidence suggests that a 10 percent increase in minimum wage would raise fast food prices by around 0.7 percent. There are reasons to believe that the higher income customers inside major cities are better able to absorb price increases without cutting back on demand. Limited evidence from San Francisco tends to confirm this observation.

Finally, there is some evidence that low-wage workers substantially increase consumption in response to wage hikes.  Daniel Aaronson and Eric French at the Federal Reserve argue that the higher marginal propensity to consume among low-wage work­ers is likely to lead to some short-term increases in economic growth from a minimum wage increase. My reading of the evidence is that it is somewhat difficult to accurately assess the importance of this channel, in part because the relatively small number of minimum wage workers makes any aggregate demand effect fairly small. But I do think that the size of increases and possible in-migration of low-wage workers into urban areas may increase the local demand impact of a city wage standard.

Minimum wage policies are a powerful lever for affecting wage inequality in the bottom half of the labor market. Modest increases in minimum wages can raise the bottom wage, and family incomes, without substantially affecting employment. But minimum wages are limited in their reach, and cannot be expected to solve all our problems when it comes to wage inequality. At the same time, the much higher wage standards being implemented in some of the cities offer the possibility of taking this policy “to scale.”

Along with this greater promise, however, come added risks. The reality is that we do not know very well how these policies will affect the local economy. Future researchers would do well to utilize the careful identification strategies that have been the hall­mark of recent minimum wage research to study these high city-wide minimum wage increases. Doing so will deepen our understanding of the functioning of the low-wage labor market, and help us gauge the proper scope of this important public policy.

Economic inequality and growth in the United Kingdom: Insights for the United States

by Gavin Kelley

After an extraordinarily long and deep economic downturn, the United Kingdom is finally enjoying belated but comparatively strong growth. The current recovery is jobs-rich, with employment growth massively outper­forming expectations relative to gross domestic product. That’s the good news. In stark contrast, however, pay growth remains unprecedentedly weak and productiv­ity has plummeted. Real (inflation-adjusted) wages have fallen for six years straight, with even nominal wages growing at less than 1 percent in recent months—the lowest increase ever recorded.

This apparent collapse in the link between economic growth and real wage gains is more extreme than anything we have seen before. But the trend has not emerged completely out of the blue. Even as the U.K. economy continued to grow steadily prior to the financial crisis and global recession in 2007-2009, workers across the earnings distribution experienced a major slow-down in wage growth.

This unhappy story about the weakening relationship between wages and growth is all too familiar in the United States. But the U.K. experience is different in important respects—and potentially offers some relevant insights for U.S. policymakers to ponder.

First, let’s look at what happened. The simple ratio of GDP growth to growth in median wages in the United Kingdom weakened markedly in the period from 2003- 2008 compared to the 1990s and 1980s. In those earlier decades, wage inequality grew sharply—those at the top pulled away from the middle, and the middle pulled away from the bottom—but pay was rising across the board. In contrast, a big deceleration in the growth rate of earnings characterized the early 2000s. For the first time, median pay trailed way behind growth in real GDP per capita.

Between 1977 and 2002, average annual real wage growth for workers at the median was around 2 percent, but from 2003 to 2008 it fell to around 0 percent to 1 percent (depending on the measure of inflation used). This stagnation happened even while real GDP per capita had an average annual growth rate of 1.4 percent. The squeeze was broadly felt: the only earners on the income ladder who experienced stronger growth were those near the bottom rungs (buoyed by increases in the minimum wage) and those at the very top (especially due to bonus payments in finance).

In the wake of the financial crisis of 2008 and amid the Great Recession of 2007- 2009, the fall in real wages (around 8 percent) has also been relatively evenly spread across the earnings spectrum, though it is far bigger if we include the self-employed (who are excluded from official data). Younger workers have suffered the most, while older workers have been the least affected.

Wages, however, don’t give the full-picture when it comes to living standards. If we look at household income growth, from 1994-95 to 2011-12, the bottom half of households took just 16 percent of pre-tax growth. Upper-middle households (those in the 50th to 90th percentiles) took 45 percent of household income pre-tax growth (44 percent post-tax), proportionate to their population share. The richest 10 percent of households took 38 percent of pre-tax growth (29 percent post-tax) while the richest 1 percent took 14 percent pre-tax (9 percent post-tax).

In short, redistribution boosted the bottom half’s share of income growth from 16 to 26 percent.

Why has the link between economic growth and wages weakened? The share of GDP flowing to the wages of those on the low and middle part of the income spectrum has fallen markedly since the mid-1970s, from 16 percent to just 12 percent—a decline of 25 percent. In simple accounting terms, this relationship depends on three factors:

• How much of GDP growth goes to profit rather than labor?
• How much of that share of economic growth goes to labor in the form of non-wage benefits and how much actually gets paid out to workers in wages?
• Of this wage share, how much reaches low- and middle-income earners?

It is often assumed that the United Kingdom and the United States alike face a long-term decline in the labor share of GDP as more of our national incomes are sucked up into corporate profits due to a mix of changing globalization, technology, increased financialization and, relatedly, deregulation spurred by the impact of big money on democratic politics.

From the U.K. perspective, there has been a slight shift in this direction over time, though it is an issue that is often overstated. Changes in the U.K.’s labor share of national income accounted for only a fifth of the cleavage that had opened up between pay and productivity since the early 1970s. The decline in the labor share of income has been less marked than in the United States.

Another U.K. perspective is that workers’ wages have primarily been under pressure because of the rising burdens on employers to provide more non-wage compensation such as higher national insurance and pension contributions. These employment costs have certainly risen, but again they can be overstated, with such increases accounting for a bit over a quarter of the gap between productivity and pay. That said, it is true that the rising cost of non-wage compensation appears to have played a more important role in the period of wage stagnation from 2003 in the United Kingdom.

But by far the most important factor explaining the declining share of the cake going to the bottom half of U.K. workers since the 1970s has been rising wage inequality, although this played a smaller role in the immediate pre-crisis period of 2003 to 2008.

How these three trends are likely to evolve over the next decade and beyond is far from clear. The intellectual zeitgeist expects there to be a redistribution of income over time from labor toward capital due to the “rise of the robot” (technology replac­ing workers) and French economist Thomas Piketty’s now famous observation that “r >g” (returns on capital are greater than the returns on economic growth).

Equally troubling is the outlook for non-wage costs. The tricky balancing act over the past decade of securing adequate pensions savings for an aging society and pro­tecting the wages of today’s workers in the United Kingdom is unlikely to go away. Similarly, most projections anticipate that, following the recent downturn period where wage inequality remained fairly level, it is now likely to increase again as the highest earners pull away from the rest.

Yet the idea that resumed growth is pre-destined to mean ever higher inequality is bogus. It was not long ago, after all, that the United Kingdom experienced broadly shared eco­nomic growth. So what observations can we make based on the U.K.’s experience?

First, standing still takes a lot of effort when the ground is shifting. A rising minimum wage and aggressive use of tax-credits made a significant and positive difference in the United Kingdom, but policymakers were pushing against the grain and didn’t do enough to confront the structural economic challenges such as inadequate business investment, lack of employee bargaining power, and weak demand for skilled labor.

Second, successive waves of “welfare reform,” together with the long-term decline in labor union collective bargaining, appears to have shifted the wage-unemployment relationship since the early 2000s. Wages have become significantly less responsive to falling unemployment than was the case in the 1980s and 1990s.  At the same time, and despite the gains from the minimum wage, working poverty has become far more pervasive. Arguably, these shifts put even more onus on aggressive monetary and fiscal policy to help generate a tight labor market and wage growth.

Third, the U.K.’s policy on the minimum wage was a success but we shouldn’t rest on our laurels. The Low Pay Commission, the body that oversees the minimum wage, is widely judged to have been highly effective if perhaps too cautious. The wage gap between the bottom and middle of the distribution has fallen (slightly) since its introduction. Fifteen years ago the whole notion of the minimum wage was highly partisan. Now each of the political parties jockey for position on this issue.

The Low Pay Commission’s blend of operational independence, technical expertise, and social partnership (employer and union representation) has worked well. And this flexibility has been an advantage; in the UK context, linking the national mini­mum wage to inflation would be a mistake. But there is now a sense that we need to revise our minimum wage framework to reflect learning over 15 years and to inject more ambition into the process.

Finally, policy wonks need to think hard about the political economy of tax credits. Most experts think tax credits increased the incentive to work (boosting single-parent employment rates in particular), helped bring about a major fall in child poverty, and shored up the post-tax transfer share of income going to the bottom 50 percent of society. Yet the rapid expansion of the policy (around 8 in 10 families with kids were eligible in 2010) raced ahead of popular support, making it surprisingly easy for the current governing coalition of Conservatives and Liberal Democrats to cut them. Tax credits have been characterized as “welfare” for the work-shy, whereas “tax-relief” is generally perceived more positively.

So what is the outlook for wage inequality in the United Kingdom? Broad-based economic growth is very unlikely to return by chance. Securing such an outcome will require a number of elements, including:

• A more aggressive strategy for raising the wage floor during the current period of economic recovery, drawing confidence from growing research about the capacity of buoyant labor markets to absorb steady minimum wage rises
• Tackling the extraordinary rents that have accrued to small numbers in the finance sec­tor over the past decade as the link between run-away rewards, financial instability, and fiscal retrenchment is all too clear (and is toxic for those on low and modest incomes)
• Ditching the notion that increasing payroll taxes (on employees and employ­ers) are a politically cute way of raising extra revenue (not least when large and regressive tax-reliefs remain untouched)
• Boosting the woefully inadequate business and public investment as there is no other path to higher labor productivity
• Remedying perennial weaknesses in U.K. education policy, especially the awful wage and productivity returns to many low and intermediate level vocational qualifications (respectively, the qualification level that a 16 or 19 year old is expected to attain)

This last point is key. Education may not be the panacea that political leaders claim it to be, but the wage-penalty arising from poor quality sub-degree level vocational qualifications in the United Kingdom is particularly punitive.

More speculatively, there is a desperate need for experimentation with new labor mar­ket institutions that could offer employees some greater form of bargaining power, but in a manner that is compatible with the realities of a relatively flexible, heavily service-dominated economy. This is pretty much a policy void in the United Kingdom today.

Recreating more equitable, broad-based economic growth requires as prerequisites a tighter jobs market together with a higher wage floor. But to restore the link between economic growth and wage growth also will involve bold policy experimentation in pursuit of higher wages for those on the low- and middle-income rungs on the economy in the United Kingdom.

Inequality and the wellbeing of the poor in the United States

by Chris Wimer

How does the rise in economic inequality affect workers and their families at the bottom of the income ladder? To begin to approach an answer to such a question, it is important to first understand the facts on the ground. What have these workers and their families experienced over the past several decades? A common but deeply flawed measure of their wellbeing over the years is the official poverty rate, which fluctuates over a fairly narrow band but remained essentially flat since President Lyndon B. Johnson’s declaration of the War on Poverty in the mid-1960s.

This is not the forum to rehearse the litany of reasons why the official poverty rate is fundamentally flawed. But perhaps its biggest shortcoming is that it doesn’t count the many resources directed toward low-income families when measuring income. These resources include in-kind benefits such as supplemental nutrition assistance (what we used to call food stamps) and housing assistance, but also after-tax benefits such as the Earned Income Tax Credit and the Child Tax Credit.

When these resources are properly accounted for in a poverty measure, my colleagues and I at Columbia University demonstrate that poverty rates fell by about 40 percent over the past half century, from 26 percent in 1967 to 16 percent today. We have made more progress than we thought in fighting poverty in the United States since the 1960s. That is the good news. The bad news is that the declines I note above have come entirely because of the work of government policies and programs—not because low-income workers and families have succeeded in the workplace.

Indeed, aside from the latter half of the 1990s, low-income workers and families generally fared poorly relative to their more advantaged peers in the middle class and especially compared to the wealthy in terms of income growth. Absent resources from government programs, poverty (properly measured) would have actually increased between the 1960s and today—from 27 percent to 29 percent, equal to about 37 million people.

Focusing exclusively on numbers and percentages surrounding a specific poverty line, however, obscures other trends in income and the wellbeing of the poor. Recent data that my colleagues and I are collecting for a new longitudinal study of New York City residents tells us that actual levels of material hardship—the inability to meet one’s routine expenses—are actually quite a bit higher than poverty rates, even as properly measured. This means we need to think about those at the bottom of the income spectrum as not just those who fall below some predetermined poverty line but also those who find themselves consistently struggling to keep pace with what it costs to get by in contemporary society.

So a key question is whether the run-up in income inequality over the past five decades is a driving force of the economic woes of the less fortunate or simply another measure of it. The poor are doing better than in the past thanks to govern­ment programs that help alleviate poverty and give them the opportunity to climb the bottom rungs of the income ladder, but at the same time we know the fortunes of those at the top are far outpacing those at the bottom.

If, as some contend, the wellbeing of the poor is dampened by the rise in inequality, then we are justified in attempting to reduce income inequality in order to improve the lots of the less fortunate. But if the two are merely jointly determined—say by the rising returns on a better education that are (partially) the result of market forces—then reducing income inequality by itself is likely do little to improve the long-run wellbeing of the poor aside from helping the poor to get by and consume more from their income.

What do we know about whether rising income inequality in the United States reduces the wellbeing of the poor? Unfortunately, not very much. Cornell University economist Robert Frank argues that as inequality rises we see a pattern of so-called “expenditure cascades” as people further down the economic ladder essentially try to consume enough to “keep up with the Jones’” just above them. University of Chicago economist Marianne Bertrand finds that rising inequality leads to reduc­tions in disposable income further down the income ladder, though she is not explicitly focused on the wellbeing of the poor.

But these studies spark very provocative questions. Does increased inequality not only lead to an increase in consumer prices but also changes in consumption patterns in a way that causes income to not go as far for the poor as it might? And do these pro­cesses have actual negative effects on the overall wellbeing of the poor? Identifying such effects using common econometric methods, however, remains challenging.

So it is still an open question whether rising levels of inequality harm less-skilled and lower-earning families. Even if government programs and policies keep disad­vantaged individuals and families afloat, sociologists still might question whether income that comes once a year in the form of tax refunds or once a month in the form of a Supplemental Nutrition Assistance Program card is as useful as income from a regular paycheck, which provides benefits both remunerative and potentially cumulative, given that over time, that job may turn into a career.

What is ultimately most important is not whether people have enough resources over the course of a year to meet a somewhat arbitrary line of what experts think they need. Rather, we need to know whether people are truly able to harness their resources to meet both their daily and monthly expenses while simultaneously investing in their own and their children’s future.

In short, understanding whether and how economic inequality affects those at the bottom of the income spectrum is central to the success and wellbeing of our nation.