Category: Disaggregating Growth

Will the United States give up on data collection?

Posted on by Meghan Miller
Former Census Bureau Director Robert Groves announces results for the 2010 U.S. census at the National Press Club, December 21, 2010, in Washington.

Federal data collection isn’t an exciting political topic, but it is critically important to running the country. Data collection by the U.S. government has often led the rest of the world. Two examples are the first U.S. decennial census in 1790, which predates the comparable census in the United Kingdom by 11 years, and the National Income and Product Accounts, one of the world’s first attempts—in 1937—at creating a comprehensive system of national accounts, which are now conducted by virtually every nation. Today, however, the federal statistical agencies that collect these data are threatened by insufficient funding proposed by the Trump administration and approved by the House and Senate appropriations bills after the Senate Appropriations Committee yesterday approved legislation that provides inadequate funding.

Timely and accurate data are critical to diagnose and respond to a wide variety of policy problems across the nation—and increase efficiency and effectiveness in the government. Speaker of the House Paul Ryan (R-WI), when announcing the Evidence-Based Policymaking Commission, discussed how the collection of data gives government “the tools to make better decisions and achieve better results.” Insufficient funding means insufficient data—and programs that don’t reflect the reality for families across the United States.

The threat of less funding is already hampering the ability of the U.S. Census Bureau to conduct a robust 2020 decennial census. The Census Bureau already cut two of its test sites for a 2018 “dress rehearsal,” leaving just one site that may not be representative of the many challenges faced by the agency in different areas of the country. The agency’s former director resigned unexpectedly shortly after sparring with Congress over the Census Bureau’s budget, which is $300 million too low, according to some estimates.

The importance of the decennial census cannot be overstated. It plays many roles in the administration of the U.S. government, most notably in the reapportionment of the 435 seats in the House of Representatives to the states. The census and several ancillary surveys produced by the Census Bureau also are commonly used in academia in varied fields that include economics and epidemiology, which in turn help inform the policymaking process in these and other critical areas. Further, the allocation of federal funding to states is based on formulas that rely on accurate population counts. Inaccurate reporting means that the programs are not providing support where they are needed.

The Census Bureau is not the only statistical agency facing cuts. Both the administration and the House appropriations bills are calling for a 10 percent cut to the budget of the U.S. Bureau of Economic Analysis, or BEA, compared with 2017. The BEA is best known for producing the National Income and Product Accounts, which include a measure of the nation’s gross domestic product and quarterly growth of the economy. The Obama administration was slowly increasing funding for the agency in hopes of improving some of the its economic indicators. The funding cuts will put those improvements on hold.

U.S. statistical agencies perform an important role in steering the U.S. economy. In the 1930s and 1940s, as the nation weathered the Great Depression and prepared for war, policymakers leaned on the nation’s economists and statisticians to devise a means of tracking the economy’s progress and potential. Other nations adopted similar systems of national accounts soon after. The National Income and Product Accounts played a critical role in World War II by helping planners estimate maximum feasible military output by the economy.

Today, the nation continues to see weak wage growth in the wake of the Great Recession, but the agency that collects this data, the U.S. Bureau of Labor Statistics, will have its funding frozen at fiscal year 2017 levels in the House Labor, Health and Human Services, Education, and Related Agencies appropriations bill as reported out of committee. According to an analysis by the Council of Professional Associations on Federal Statistics, the agency needs another $30 million per year just to perform the duties that are mandated of it. An increasingly complex and data-driven world requires strong national statistics. The Trump administration and Congress would be wise to re-examine these cuts against the need for the accurate and empirical work of these critical statistical agencies.

Why current definitions of family income are misleading, and why this matters for measures of inequality

Posted on by Nick Bunker

What is “equitable growth” and how do we measure it? The following essay, part of a series, asks economists, other researchers, and practitioners to explore these questions. Equitable growth means an economy that raises living standards for all families. We have seen decades of economic growth in the U.S.—commonly measured by GDP. Yet that success has not meant significant income growth for most American families. Clearly GDP doesn’t provide the full picture. How do we know we’re on the right track? There is little consensus around what specific of indicators are required to quantify whether the economy is growing on behalf of all Americans. Is it a matter of looking at different already existing measures? Should new data using existing concepts of income and well-being be created? Do our concepts of what’s important to measure need updating as well? A better understanding of equitable growth—and how to measure it—can improve our understanding, inform decisions and lead to better outcomes for all.

Researchers studying income distribution in the United States seem reluctant to acknowledge the family as an important unit of production and distribution. As a result, they often rely on statistics that provide a misleading picture of inequalities based on class, race or ethnicity, and especially gender.

Incomplete definitions of both family and income either obscure or render invisible transfers between and within households, including the value of housework and family care. Evidence from specialized surveys—such as the Health and Retirement Survey, the Panel Survey of Income Dynamics, the Survey of Income and Program Participation, and the American Time Use Survey—clearly demonstrate the quantitative relevance of these omissions.

Conventional measures

What, exactly, do economists mean by income, and what, exactly, is the presumed income-receiving unit? Usually, income refers to direct market income (labor earnings plus income from capital such as interest or dividends, and including, where feasible, indirect market income such as the dollar value of transfers from private pensions or government).

Many shortcomings of this measure are widely recognized. For instance, conventional estimates do not include any valuation of the flow of implicit service income from capital assets such as housing or the increase in wealth due to capital gains appreciation.1 Sources of income that take the form of in-kind benefits and/or tax expenditures such as the Earned Income Tax Credit are seldom included. These problems, however, have received more attention than those related to largely unmeasured aspects of the family economy.

Most individuals in the United States pool at least some of their income with other family members over a significant portion of their lifecycles. As a result, family income is a better indicator of material living standards than individual earnings. Family-based measures are especially relevant to the economic welfare of children, the elderly, and individuals who are sick or disabled, as well as those supporting or providing direct care for such dependents.

Many unrelated individuals live together in households without pooling income, but benefit from household public goods and economies of scale in household production. That’s why it is difficult to measure the extent to which individuals pool their income and what share of family or household income should be imputed to them. While it is often assumed that married couples equally share their market income, empirical research suggests that is not always the case. 2 The proliferation of informal partnerships such as cohabitation further complicate the story. 3

Intrafamily transfers of money and time—mediated by the public-good aspects of household consumption and economies of scale in household production—potentially affect both the size and the distribution of individual income. For instance, improvements in women’s earnings relative to those of men may be counterbalanced by a decline in intrahousehold or intrafamily transfers related to nonmarriage, loss of household economies of scale, or increases in the percentage of children maintained by women alone. 4

Defining the family in family income

Seeking a practical solution to a complex problem, the U.S. Census Bureau enforces a clear distinction between family and household. Specifically:

A family consists of two or more people (one of whom is the householder) related by birth, marriage, or adoption residing in the same housing unit. A household consists of all people who occupy a housing unit regardless of relationship. A household may consist of a person living alone or multiple unrelated individuals or families living together. 5

Note, however, that the definition of family provided here is limited to family members living in the same household. In this sense, it represents a truncated and, in some respects, misleading definition. While the U.S. Current Population Survey asks some questions relating to intrahousehold family transfers, these are largely been considered a private matter, except where they represent a traditional obligation rendered visible by child support agreements.

By contrast, the Health and Retirement Survey asks respondents to report financial help, defined as:

Giving money, helping pay bills, or covering specific types of costs such as those for medical care OR insurance, schooling, down payment for a home, rent, etc. The financial help can be considered support, a gift or a loan.

This is a much broader definition of intrafamily transfers than that in the Current Population Survey, and a recent empirical analysis of the Health and Retirement Survey finds that households with an adult between the ages of 50 and 64 transferred an average of $8,350 to family members over a two-year period between 2008 and 2010. Both the probability and the size of these transfers were positively correlated with income, and the overall likelihood of such transfers increased substantially between 1998 and 2010. 6 In other words, relatively affluent adults approaching retirement age have provided an increasingly significant economic boost to their adult children, which is not factored into conventional family-income calculations.

This analysis of the Health and Retirement Survey does not break out transfers by race or ethnicity, but other research utilizing data from the 2005 and 2007 Panel Study of Income Dynamics, as well as the Survey of Consumer Finance, shows that middle- and upper-income African Americans are more likely to provide informal financial assistance than whites with similar characteristics. 7 Not surprisingly, black families are more likely to have needy family members and friends—what might be termed a negative network effect. This difference can account for a significant portion of the racial gap in wealth.

Overall, such transfers may have an equalizing effect because they generally flow from those with more market income to those with less. But young white adults are more likely to receive transfers from relatively affluent parents, while young black adults are more likely to transfer income to those closer to the bottom of distribution.

Equivalence scales and intrafamily transfers

Comparisons of family income are often unadjusted for family household composition or are adjusted on a per-capita basis, simply divided by the number of household members. Both approaches are misleading. A family of two is much better off with an income of $50,000 than a family of six. In contrast, a family of six does not need three times as much income to be as well off as a family of two, even though it has three times as many members.

For this reason, family income is often adjusted by an equivalence scale that assigns a different weight to children and adults, and takes economies of scale into account. The U.S. poverty line and benchmarks based upon it—such as the 200 percent of the poverty line—represents an implicit equivalence scale. Another common measure divides family income by the square root of family size. 8 Such scales represent an approximation of what might be termed an intrafamily transfer. 9

Virtually all conventional equivalence scales assume that children are less costly than adults because the cost of feeding and clothing them is lower. 10 Further, virtually all applications of equivalence scales to family income in the United States apply the same scales at every point in time. Since the mid-1960s, however, children have become more costly relative to adults, and family budgets have shifted away from food and clothing toward services such as childcare and education.

A major factor behind increased childcare costs is the significant increase in the labor-force participation of mothers between 1975 and the mid-1990s. Another is the steady climb in the percentage of children living in families maintained by a mother alone, since mothers are required to engage in paid employment in order to even qualify for public assistance.

While longitudinal data are scarce, a recent Census Bureau report based on the Survey of Income and Program Participation estimates that overall expenditures on childcare doubled between 1985 and 2011, from $84 to $143 per week in constant dollars. In the most recent year, families with incomes below the federal poverty line spent about 30 percent of their income on childcare, compared to 8 percent for families not in poverty. 11

Parental spending on higher education has also increased, the combined result of increasing college enrollments (despite relatively stagnant graduation rates) and significant increases in tuition and fees, particularly over the past 15 years. 12 Research also shows that home buyers and renters pay a significant premium for houses in high-quality school districts, indirectly increasing the cost of children. 13

These factors have important implications for considering the distribution of adjusted family income today. Whites in general are less likely than African Americans or Hispanics to live in households with children, and college-educated women are significantly less likely than other women to become single parents. Current equivalence scales significantly understate the economic significance of these demographic differences. Yet many dual-earner families with children sit squarely in the middle of the (conventionally measured) family-income distribution.

In current economic parlance, disposable income is typically defined as income after taxes. One could conceptualize “adult disposable income” as the net of taxes and benefits and transfers to children and other dependents. Instead, current assumptions treat spending on children or other needy family members merely as another form of consumption, no different than spending on restaurant meals or automobiles.

The value of nonmarket work

Housework and family care are now widely recognized as forms of work that yield economic benefits. Recent data from the American Time Use Survey show that productive activities that someone else could, in principle, be paid to perform constitute roughly half of all time devoted to work in the United States. 14

A number of studies impute a market value to this work on the aggregate level, simply multiplying the number of hours by an estimate of quality-adjusted replacement cost. This exercise suggests that the contribution of nonmarket work to an expanded definition of Gross Domestic Product in the United States lies somewhere between 30 percent and 40 percent. 15 Yet, with a few notable exceptions, the value of nonmarket work is largely ignored in estimates of family income on the microeconomic level. 16

Consider two family households of identical composition consisting of two adults and two children under the age of 5, both with a family income of $50,000 (ignoring both taxes and benefits, for the sake of simplicity). Conventional measures would place both of these families at exactly the same place in the distribution of income. But what if the first family includes two wage-earners, both working 40 hours per week and earning $25,000 per year, and the second family includes one wage-earner, working 40 hours per week and earning $50,000 per year, along with one stay-at-home parent who prepares meals, does shopping, and provides childcare.

Surely the second family is significantly better off than the first, if only because it does not incur the childcare costs alluded to in the section above.

What effect does imputation of the value of nonmarket work have on estimates of the distribution of family income in the United States? Empirical work today suggests that it has an equalizing effect in the cross-section, not because low-income families devote more time to it, on average, but because any imputation of the value of that work represents a larger percentage of their market income. 17

The implications for trends over time are quite different. The equalizing effect of valuing nonmarket work was almost certainly greater in the 1960s, when a relatively large percentage of married women were full-time homemakers. As women entered wage employment and substituted market employment for at least some of their nonmarket work, this equalizing effect diminished. Inequality in women’s earnings is also far greater today than it was in the 1960s, with high-earning women likely to marry high-earning men. 18

Whatever the gender implications of the traditional breadwinner/homemaker family, it may well have mitigated some aspects of class inequality among whites (it was never widespread among blacks). But most studies of the impact of women’s increased labor-force participation on income inequality completely ignore the value of nonmarket work, essentially assigning homemakers a contribution of “zero” in their empirical analysis. 19

Implications

Changes in the family economy of the United States have probably had only small effects on the relative income of the top 1 percent or the top 5 percent. They have larger implications for both the reality and the perception of relative income among households with divergent patterns of female labor-force participation and family responsibility.

Many public benefits in the United States—from the Supplemental Nutrition Assistance Program to financial aid for college—are conditioned on conventional measures of family income. Because these measures provide an incomplete and misleading picture of relative well-being of families, reliance on them may breed frustration and resentment. 20

Many of the policy proposals emerging from both political parties speak to concerns about the costs of family care: increased public provision of care and education, as well as child and dependent care tax credits. The potentially equalizing effect of such policies deserves serious consideration. In principle, many of the data sources cited above offer the potential to enlarge appreciation of the family in family income.

—Nancy Folbre is the director of the program on gender and care work at the Political Economy Research Center at the University of Massachusetts, Amherst.

Improving the measurement and understanding of economic inequality in the United States

Posted on by Nick Bunker

What is “equitable growth” and how do we measure it? The following essay, part of a series, asks economists, other researchers, and practitioners to explore these questions. Equitable growth means an economy that raises living standards for all families. We have seen decades of economic growth in the U.S.—commonly measured by GDP. Yet that success has not meant significant income growth for most American families. Clearly GDP doesn’t provide the full picture. How do we know we’re on the right track? There is little consensus around what specific of indicators are required to quantify whether the economy is growing on behalf of all Americans. Is it a matter of looking at different already existing measures? Should new data using existing concepts of income and well-being be created? Do our concepts of what’s important to measure need updating as well? A better understanding of equitable growth—and how to measure it—can improve our understanding, inform decisions and lead to better outcomes for all.

There has long been interest in extending and improving the National Income and Product Accounts, or NIPA, to turn it into a better indicator of general economic welfare and its progress. The accounts, and the summary Gross Domestic Product number, were not initially intended as an indicator of welfare, but rather as a measure of economic activity—even misdirected activity such as the production of cigarettes. Even on that basis, there is room for improvement: Perhaps the most commonly suggested extension has been the explicit inclusion of environmental degradation and improvement, along with other instances of resource depletion. I would certainly be in favor.

Nevertheless, I want to begin with a retrograde suggestion. Whatever is eventually done with NIPA and GDP, I hope it is done in such a way that it will always be possible easily to extract from the new NIPA most of the components of the old NIPA. (It will not be possible to extend the new NIPA back in time very far because the basic data will never have been collected.) For those of us who want to go beyond measurement to understand macroeconomic behavior and policy, having a fairly consistent quarterly time series going back to 1949 is a wonderful thing. Even with those data, it is a hard problem—without them, it would be hopeless. Another 65 years of compatible data would be more than welcome.

From the very beginning of national accounting, it has been understood that the focus on gross investment and gross product is a standing temptation to error. If GDP increases from one year to the next only because there is a bigger charge for depreciation of fixed capital, it is obvious that nothing has really gotten better. Taking account of depreciation to yield net product—and correspondingly net national income—would provide a better measure of productive economic activity. The prominence given to gross investment and GDP derived from the realization that measures of depreciation taken from business accounts would be not only inaccurate but also biased in odd ways. The practice of depreciation accounting was too badly infected by tax incentives and cosmetic considerations; it was thought better to sidestep the resulting not-even-random errors.

I am told that the situation has not improved, so my second suggestion would be to devote more resources to the improvement of depreciation (and depletion) accounting. If we could move in the future to a focus on net output and net income, NIPA would be a small step closer to providing better elementary building blocks for the measurement of welfare and its changes through time.

These suggestions amount only to shifting or not shifting deck chairs. A useful and substantial extension of the standard statistical picture of the economy would be the regular publication of distributional information. I would want to go well beyond the familiar Gini coefficient, which hides more than it reveals. For starters, I would urge the regular calculation of the distribution of personal income by size, say by income deciles. (The usual quintile distribution is a lot better than nothing but still too rough.) It would be very informative to have such distributions for both market incomes—before taxes and transfers, and after taxes and transfers. It would be especially nice if regular official publications could familiarize the interested public with the Lorenz curve, which is probably the best graphic way of grasping shifts in inequality. The choice between quarterly and annual publication is a trade-off. A year is a long time to wait for news if you care about changes in income inequality, but for an individual or group of individuals, shifts in the interquarterly timing of income may be essentially meaningless. Only experience will clarify the pros and cons.

It would also be useful, in a different way, to have a serious accounting of the assets and liabilities of the public sector at all levels of government. Governments own land, buildings, vehicles, and other equipment. They invest in these objects and undergo depreciation. Even if we do not adequately measure their contribution to output, we should at least take account of their existence. It is a familiar plaint that infrastructure in the United States—bridges, roads, port facilities—is badly decayed and in need of repair and replacement. Regular measurement might lead to more rational decision-making.

There is another, more speculative reason to want better wealth accounting in the public sector. A common cause for alarm these days is the threatened advance of the robots and the possibility that human labor may become all but obsolete. If the threat is real—who knows?—one significant policy response might be the establishment of large sovereign wealth funds that would indirectly own some of those hypothetical robots, and thus provide a source of income to replace the missing wages and salaries. It would then be an advantage to have a tradition of accounting for public-owned wealth and its productivity.

It hardly needs to be added that better and more complete tracking of the distribution of private wealth would also contribute to our understanding of economic welfare. A lot has been accomplished by a group of scholars making use of tax data here and elsewhere. The tax files necessarily miss quite a lot of private wealth, and even apart from that, the development of further sources of information could provide a much-needed check on information and disinformation gleaned from tax returns. Both measurement and understanding would be improved.

Finally, sticking close to familiar economic indicators, I would urge a major effort to collect more longitudinal data, especially on employment status, earnings, and income. Of course, many scholars have used the Panel Survey of Income Dynamics and other surveys for this purpose. I am hoping for a larger, more systematically designed effort aimed specifically at producing the transition frequencies that could serve as the basis for an understanding of the random process that generates lifetime incomes.

(Personal history: When I was writing my Ph.D. thesis in 1949, I came upon transition matrices for covered earnings that the Social Security Administration had compiled for three or four years in the late 1930s. They made the basis for an excellent chapter. Coverage is much broader now, so there’s the possibility for a fresh start with large enough numbers to take account of age and other personal characteristics.)

It is obvious that my wish list has steered away from other vital indicators of social and economic welfare that have often been proposed. I have not mentioned such important issues as health status, access to education and training, exposure to crime, and subjective feelings of security in general. All of these and others are part of a complete picture of average well-being and its distribution in society. I have omitted them not out of abject ignorance or indifference, but out of respect for the principle of comparative advantage.

—Robert Solow is a Nobel Laureate and professor emeritus at the Massachusetts Institute of Technology.

The once and future measurement of economic inequality in the United States

Posted on by Emilie Openchowski
A man drives a golf cart from his house to his golf club as a group of landscape workers take a break in Vista, California.

A slew of research into economic inequality replete with serious looking graphs may give the impression that measuring inequality in the United States is a solved problem. This is misleading. Inequality is still measured incompletely because existing U.S. government statistics do not attempt to match their estimates to the National Income and Product Accounts. NIPA is the source of the most reported and well-understood economic statistics such as the nation’s Gross Domestic Product and quarterly GDP growth figures.

Because existing estimates of economic inequality are not pegged to NIPA, they don’t account for all sources of income. They may exclude, for example, fringe benefits provided by employers such as employer-provided health insurance and retirement benefits, government transfers such as supplemental nutrition assistance or the child tax credit, government services such as public education, and tax expenditures such as the home mortgage tax deduction and tax breaks for employer-provided insurance. These exclusions, big and small, make many existing estimates of inequality fundamentally incomparable to our most well-established measures of economic growth.

This wasn’t always the case. The Office of Business Economics—the precursor to the U.S. Department of Commerce’s Bureau of Economic Analysis—compiled inequality data decades ago, starting in 1947 and ending in 1971. These estimates were relatively simple: They divided households into five quintiles and reported the accumulated income of each quintile. The bottom 20 percent of households, for example, held 5 percent of all personal income in 1956, while the top 20 percent held 44.9 percent of all personal income. (See Figure 1.)

Figure 1

Inequality remained fairly stable during this period. Because economic prosperity was broadly shared from the end of World War II until the early 1970s, the distribution of income rarely changed. As one researcher noted, “the relative distribution of income has remained virtually constant over the post-war period.” Unfortunately, these estimates were discontinued due to a lack of resources at an inopportune time: Income inequality would increase slowly starting in the 1970s and more rapidly in the succeeding decades.

Today, with inequality increasing, there is rekindled interest within the Bureau of Economic Analysis in measuring inequality alongside growth. A new paper in the BEA’s Survey of Current Business attempts to reconstruct measures of inequality that are pegged to NIPA, similar to those compiled in the middle of the 20th century. The team of current and former government economists—Dennis Fixler at the BEA, David Johnson at the University of Michigan, and Andrew Craig and Kevin Furlong at the BEA—merge the Current Population Survey and the Consumer Expenditure Survey to construct estimates of income for each quintile of the U.S. population between 2000 and 2012.

They find that the top 20 percent now hold about 52 percent of all personal income. Moreover, the paper moves beyond the BEA’s older efforts in key ways. The authors provide estimates of inequality for regions of the United States and for each state individually, as well as the change in inequality between 2000 and 2012. (See Figures 2 and 3.)

Figure 2

Figure 3

In the paper, “Toward National and Regional Distributions of Personal Income,” the four authors also decompose personal income by category, showing how Social Security income, Medicare benefits, and more are shared by each quintile of the income distribution. They find, for example, that 86 percent of all dividend income flows to the top 20 percent of U.S. households, highlighting the near-monopoly that upper-class households have over financial assets.

There are limits to what can be done with the tools that are currently available. Respondents to the surveys under- and over-report income, for example. Some components of income are missing entirely and must be estimated based on what clues are available. The two surveys used are linked using a procedure that introduces some error into the estimates. Other estimates of inequality show that it is important to break out the top 10 percent of income earners or even the top 1 percent of earners, and these groups are not addressed in this paper.

Another approach, by Thomas Piketty at the Paris School of Economics and University of California, Berkeley economists Emmanuel Saez and Gabriel Zucman, uses tax data to look at very high earners, showing that the top 1 percent and even the top 0.1 percent have been the foremost beneficiaries of recent increases in inequality. Ultimately, better surveys and more interagency access to U.S. government administrative data is necessary to address the challenges of providing better inequality statistics.

The ability to look at the geographic distribution of inequality and at slices of income within different income groups teases the possibilities of a more robust project to disaggregate the National Income and Product Accounts statistics that are currently the most referenced statistics of economic progress in the nation. Devoting federal resources to the project could allow us to track inequality not only by income bands, but also by age, geographic location, gender, ethnicity, and type of income.

Distributional National Accounts and measuring 21st century growth

Posted on by Bridget Ansel
In this June 1, 2016 photo, vacant buildings stand near the Pyramid which houses a Bass Pro Shops megastore that opened in 2015, in Memphis, Tenn. Statistics describe an America that is nearly recovered from the Great Recession, but the national averages don’t give a complete or accurate picture. Wealth is flowing disproportionately to the rich, skewing the data used to measure economic health.

The U.S. Bureau of Economic Analysis late last week released revised data showing that in the third quarter, the U.S. economy grew at an annual rate of 3.2 percent. This was deservedly celebrated as good news—growth was higher than previously reported, which means more economic activity and, with that, presumably, more jobs and better incomes. Yet that same data revealed nothing about how the economy is performing for people across the income spectrum.

New data released this week from economists Thomas Piketty, Emmanuel Saez, and Gabriel Zucman addresses this knowledge gap. Based on careful research that matches data on aggregate economic growth to individual incomes, the researchers show that between 1980 and 2014, on average, pre-tax income grew by 61 percent over that period, yet most of the U.S. population did not benefit from this growth. The bottom 50 percent of the population saw only a 1 percent growth in their pre-tax incomes (after adjusting for inflation) while those in the top 1 percent saw their incomes rise by 205 percent.

These Distributional National Accounts—developed by the three economists and co-collaborators working at the Paris School of Economics, the University of California-Berkeley, Oxford University, and Harvard University—provide policymakers and economists alike with a better way of understanding economic growth—one that directly connects the analysis of aggregate economic data with the real-life circumstances of individuals.

For generations, economists relied on very broad national income and product accounts to report on economic activity. These data—the National Income and Product Accounts—aggregate information from all the businesses, households, and governments across the economy to discern the total value of goods and services sold, the total incomes received, and what share comes from various sources, such as earnings, interest, rent, or government payments. The data also show how much the United States sells to other countries and buys from abroad.

This data is central to understanding how the U.S. economy works, but it is important to remember that policymakers more than a half-century ago made a choice about how to discern what was happening in the economy—one that did not take into consideration the consequences of economic growth on individuals but which suited the economic issues of the era. In the 1930s, in the wake of the Great Depression, the U.S. Commerce Department commissioned Nobel laureate Simon Kuznets to develop a set of national economic accounts. This was a time when promoting growth (and the jobs that come with it) was the nation’s priority. Prior to this, policymakers, business leaders, and families had to rely on a hodgepodge of data to infer what was going on in the economy.

Make no mistake, Kuznets’ National Income and Product Accounts have served their purpose over time and are among the most significant data developments of the 20th century. The data remain one of the most important tools that the Federal Reserve Board and other policymakers have to understand and manage the U.S. economy toward full employment. Historians credit the implementation of these accounts as one of the key reasons the United States so effectively marshaled economic resources to fight in World War II.

Today, there is a new data frontier—understanding what growth looks like for individuals and families throughout the U.S. economy amid growing income inequality. The data that underpins the new Distributional National Accounts can help policymakers understand why, even though the economy grew by 16 percent in the wake of the Great Recession, millions of Americans report that the economy is not working for them any better than it was amid the worst economic downturn since the Great Depression. Those millions of individual Americans and their families get it—most have not benefitted from more than seven years of growth. Distributional National Accounts enable policymakers to understand whether and how income inequality affects economic growth.

Economic growth in the United States: A tale of two countries

Posted on by Bridget Ansel

Overview

The rise of economic inequality is one of the most hotly debated issues today in the United States and indeed in the world. Yet economists and policymakers alike face important limitations when trying to measure and understand the rise of inequality.

One major problem is the disconnect between macroeconomics and the study of economic inequality. Macroeconomics relies on national accounts data to study the growth of national income while the study of inequality relies on individual or household income, survey and tax data. Ideally all three sets of data should be consistent, but they are not. The total flow of income reported by households in survey or tax data adds up to barely 60 percent of the national income recorded in the national accounts, with this gap increasing over the past several decades.21

This disconnect between the different data sets makes it hard to address important economic and policy questions, such as:

  • What fraction of economic growth accrues to those in the bottom 50 percent, the middle 40 percent, and the top 10 percent of the income distribution?
  • What part of the rise in inequality is due to to changes in the share of national income that goes to workers (labor income) and owners (capital income) versus changes in how these labor and capital incomes are distributed among individuals?

A second major issue is that economists and policymakers do not have a comprehensive view of how government programs designed to ameliorate the worst effects of economic inequality actually affect inequality. Americans share almost one-third of the fruits of economic output (via taxes that help pay for an array of social services) through their federal, state, and local governments. These taxes collectively add up to about 30 percent of national income, and are used to fund transfers and public goods that ultimately benefit all U.S. families. Yet we do not have a clear measure of how the distribution of pre-tax income differs from the distribution of income after taxes are levied and after government spending is taken into account. This makes it hard to assess the extent to which governments make income growth more equal.22

In a recent paper, the three authors of this issue brief attempt to create inequality statistics for the United States that overcome the limitations of existing data by creating distributional national accounts.23 We combine tax, survey, and national accounts data to build a new series on the distribution of national income. National income is the broadest measure of income published in the national accounts and is conceptually close to gross domestic product, the broadest measure of economic growth.24 Our distributional national accounts enable us to provide decompositions of growth by income groups consistent with macroeconomic growth.

Related

How to deliver equitable growth: 14 strategies for the next administration

In our paper, we calculate the distribution of both pre-tax and post-tax income. The post-tax series deducts all taxes and then adds back all transfers and public spending so that both pre-tax and post-tax incomes add up to national income. This allows us to provide the first comprehensive view of how government redistribution in the United States affects inequality. Our benchmark series use the adult individual as the unit of observation and split income equally among spouses in married couples. But we also produce series where each spouse is assigned their own labor income, allowing us to study gender inequality and its impact on overall income inequality. In this short summary, we would like to highlight three striking findings.

Our first finding—a surge in income inequality

First, our data show that the bottom half of the income distribution in the United States has been completely shut off from economic growth since the 1970s. From 1980 to 2014, average national income per adult grew by 61 percent in the United States, yet the average pre-tax income of the bottom 50 percent of individual income earners stagnated at about $16,000 per adult after adjusting for inflation.25 In contrast, income skyrocketed at the top of the income distribution, rising 121 percent for the top 10 percent, 205 percent for the top 1 percent, and 636 percent for the top 0.001 percent. (See Figures 1 and 2.)

Figure 1

Figure 2

It’s a tale of two countries. For the 117 million U.S. adults in the bottom half of the income distribution, growth has been non-existent for a generation while at the top of the ladder it has been extraordinarily strong. And this stagnation of national income accruing at the bottom is not due to population aging. Quite the contrary: For the bottom half of the working-age population (adults below 65), income has actually fallen. In the bottom half of the distribution, only the income of the elderly is rising.26 From 1980 to 2014, for example, none of the growth in per-adult national income went to the bottom 50 percent, while 32 percent went to the middle class (defined as adults between the median and the 90th percentile), 68 percent to the top 10 percent, and 36 percent to the top 1 percent. An economy that fails to deliver growth for half of its people for an entire generation is bound to generate discontent with the status quo and a rejection of establishment politics.

Because the pre-tax incomes of the bottom 50 percent stagnated while average national income per adult grew, the share of national income earned by the bottom 50 percent collapsed from 20 percent in 1980 to 12.5 percent in 2014. Over the same period, the share of incomes going to the top 1 percent surged from 10.7 percent in 1980 to 20.2 percent in 2014.27 As shown in Figure 2, these two income groups basically switched their income shares, with about 8 points of national income transferred from the bottom 50 percent to the top 1 percent. The gains made by the 1 percent would be large enough to fully compensate for the loss of the bottom 50 percent, a group 50 times larger.

To understand how unequal the United States is today, consider the following fact. In 1980, adults in the top 1 percent earned on average 27 times more than bottom 50 percent of adults. Today they earn 81 times more. This ratio of 1 to 81 is similar to the gap between the average income in the United States and the average income in the world’s poorest countries, among them the war-torn Democratic Republic of Congo, Central African Republic, and Burundi. Another alarming trend evident in this data is that the increase in income concentration at the top in the United States over the past 15 years is due to a boom in capital income. It looks like the working rich who drove the upsurge in income concentration in the 1980s and 1990s are either retiring to live off their capital income or passing their fortunes onto heirs.

Our second finding—policies to ameliorate income inequality fall woefully short

Our second main finding is that government redistribution has offset only a small fraction of the increase in pre-tax inequality. As shown in Figure 1, the average post-tax income of the bottom 50 percent of adults increased by only 21 percent between 1980 and 2014, much less than average national income. This meager increase comes with two important limits.

First, there was almost no growth in real (inflation-adjusted) incomes after taxes and transfers for the bottom 50 percent of working-age adults over this period because even as government transfers increased overall, they went largely to the elderly and the middle class. Second, the small rise of the average post-tax income of the bottom 50 percent of income earners comes entirely from in-kind health transfers and public goods spending. The disposable post-tax income—including only cash transfers—of the bottom 50 percent stagnated at about $16,000. For the bottom 50 percent, post-tax disposable income and pre-tax income are similar—this group pays roughly as much in taxes as it receives in cash transfers.

Our third finding—comparing income inequality among countries is enlightening

Third, an advantage of our new series is that it allows us to directly compare income across countries. Our long-term goal is to create distributional national accounts for as many countries as possible; all the results will be made available online on the World Wealth and Income Database. One example of the value of these efforts is to compare the average bottom 50 percent pre-tax incomes in the United States and France.28 In sharp contrast with the United States, in France the bottom 50 percent of real (inflation-adjusted) pre-tax incomes grew by 32 percent from 1980 to 2014, at approximately the same rate as national income per adult. While the bottom 50 percent of  incomes were 11 percent lower in France than in the United States in 1980, they are now 16 percent higher. (See Figure 3.)

Figure 3

The bottom 50 percent of income earners makes more in France than in the United States even though average income per adult is still 35 percent lower in France than in the United States (partly due to differences in standard working hours in the two countries).29 Since the welfare state is more generous in France, the gap between the bottom 50 percent of income earners in France and the United States would be even greater after taxes and transfers.

The diverging trends in the distribution of pre-tax income across France and the United States—two advanced economies subject to the same forces of technological progress and globalization—show that working-class incomes are not bound to stagnate in Western countries. In the United States, the stagnation of bottom 50 percent of incomes and the upsurge in the top 1 percent coincided with drastically reduced progressive taxation, widespread deregulation of industries and services, particularly the financial services industry, weakened unions, and an eroding minimum wage.

Conclusion

Given the generation-long stagnation of the pre-tax incomes among the bottom 50 percent of wage earners in the United States, we feel that the policy discussion at the federal, state, and local levels should focus on how to equalize the distribution of human capital, financial capital, and bargaining power rather than merely the redistribution of national income after taxes. Policies that could raise the pre-tax incomes of the bottom 50 percent of income earners could include:

  • Improved education and access to skills, which may require major changes in the system of education finance and admission
  • Reforms of labor market institutions to boost workers’ bargaining power and including a higher minimum wage
  • Corporate governance reforms and worker co-determination of the distribution of profits
  • Steeply progressive taxation that affects the determination of pay and salaries and the pre-tax distribution of income, particularly at the top end

The different levels of government in the United States today obviously have the power to make income distribution more unequal, but they also have the power to make economic growth in America more equitable again. Potentially pro-growth economic policies should always be discussed alongside their consequences for the distribution of national income and concrete ways to mitigate their unequalizing effects. We hope that the distributional national accounts we present today can prove to be useful for such policy evaluations.

We will post online our complete distributional national accounts micro-data. These micro-files make it possible for researchers, journalists, policymakers, and any interested user to compute a wide array of distributional statistics—income, wealth, taxes paid and transfers received by age, gender, marital status, and other measures—and to simulate the distributional consequences of tax and transfer reforms in the United States.

—Thomas Piketty is a professor of economics at the Paris School of Economics. Emmanuel Saez is a professor of economics and director of the Center for Equitable Growth at the University of California-Berkeley. Gabriel Zucman is an assistant professor of economics at the University of California-Berkeley. They are co-directors of the World Wealth and Income Database, together with economists Facundo Alvaredo at the Paris School of Economics and Anthony Atkinson at Oxford University.

Distributional national accounts: Methods and estimates for the United States

Download File
120716-WP-distributional-national-accounts

Read the full PDF in your browser

Authors:

Thomas Piketty, Professor of Economics, École des Hautes Études en Sciences Sociales, Paris School of Economics
Emmanuel Saez, Professor of Economics & Director of the Center for Equitable Growth, University of California, Berkeley
Gabriel Zucman, Assistant Professor of Economics, University of California, Berkeley

Abstract:

This paper combines tax, survey, and national accounts data to estimate the distribution of national income in the United States since 1913. Our distributional national accounts capture 100% of national income, allowing us to compute growth rates for each quantile of the income distribution consistent with macroeconomic growth. We estimate the distribution of both pre-tax and post-tax income, making it possible to provide a comprehensive view of how government redistribution affects inequality. Average pre-tax national income per adult has increased 60% since 1980, but we find that it has stagnated for the bottom 50% of the distribution at about $16,000 a year. The pre-tax income of the middle class—adults between the median and the 90th percentile—has grown 40% since 1980, faster than what tax and survey data suggest, due in particular to the rise of tax-exempt fringe benefits. Income has boomed at the top: in 1980, top 1% adults earned on average 27 times more than bottom 50% adults, while they earn 81 times more today. The upsurge of top incomes was first a labor income phenomenon but has mostly been a capital income phenomenon since 2000. The government has offset only a small fraction of the increase in inequality. The reduction of the gender gap in earnings has mitigated the increase in inequality among adults. The share of women, however, falls steeply as one moves up the labor income distribution, and is only 11% in the top 0.1% today.

U.S. top one percent of income earners hit new high in 2015 amid strong economic growth

Posted on by David Evans

The top 1 percent income earners in the United States hit a new high last year, according to the latest data from the U.S. Internal Revenue Service. The bottom 99 percent of income earners registered the best real income growth (after factoring in inflation) in 17 years, but the top one percent did even better. The latest IRS data show that incomes for the bottom 99 percent of families grew by 3.9 percent over 2014 levels, the best annual growth rate since 1998, but incomes for those families in the top 1 percent of earners grew even faster, by 7.7 percent, over the same period. (See Figure 1.)

Figure 1

Overall, income growth for families in the bottom 99 percent was good again in 2015 as it had been last year, marking the second year of real recovery from the income losses sparked by the Great Recession of 2007-2009. After a large decline of 11.6 percent from 2007 to 2009, real incomes of the bottom 99 percent of families registered a negligible 1.1 percent gain from 2009 to 2013, and then grew by 6.0 percent from 2013 to 2015. Hence, a full recovery in income growth for the bottom 99 percent remains elusive. Six years after the end of the Great Recession, those families have recovered only about sixty percent of their income losses due to that severe economic downturn.

In contrast, families at or near the top of the income ladder continued to power ahead. These families at or near the top of the income ladder did substantially better in 2015 than those below them. The share of income going to the top 10 percent of income earners—those making on average about $300,000 a year—increased to 50.5 percent in 2015 from 50.0 percent in 2014, the highest ever except for 2012. The share of income going to the top 1 percent of families—those earning on average about $1.4 million a year—increased to 22.0 percent in 2015 from 21.4 percent in 2014.

Income inequality in the United States persists at extremely high levels, particularly at the very top of the income ladder. Figure 1 shows that the incomes (adjusted for inflation) of the top 1 percent of families grew from $990,000 in 2009 to $1,360,000 in 2015, a growth of 37 percent. In contrast, the incomes of the bottom 99 percent of families grew only by 7.6 percent–from $45,300 in 2009 to $48,800 in 2015. As a result, the top 1 percent of families captured 52 percent of total real income growth per family from 2009 to 2015 while the bottom 99 percent of families got only 48 percent of total real income growth. This uneven recovery is unfortunately on par with a long-term widening of inequality since 1980, when the top 1 percent of families began to capture a disproportionate share of economic growth.

The 2015 numbers on income have been built using the new filling-season statistics by size of income published by the Statistics of Income division of the IRS. These statistics can be used to project the distribution of incomes for the full year. We have used these new statistics to update our top income share series for 2015, which are part of our World Top Incomes Database. These statistics measure pre-tax cash market income excluding government transfers such as the disbursal of the earned income tax credit to low-income workers.

Timely statistics on economic inequality are key to understanding whether and how inequality affects economic growth. Policymakers in particular need to grasp whether past efforts to raise taxes on the wealthy—in particular the higher tax rates for top U.S. income earners enacted in 2013 as part of the 2013 federal budget deal struck by Congress and the Obama Administration—are effective at slowing income inequality.

The latest data from the IRS suggests the 2013 reforms proved to be fleeting in terms of reducing income inequality. There was a dip in pre-tax income earned by the top one percent in 2013, yet by 2015 top incomes are once again on the rise—following a pattern of growing income inequality stretching back to the 1970s.

—Emmanuel Saez in a professor of economics at the University of California-Berkeley and a member of the Washington Center for Equitable Growth’s steering committee.

Photo by Steve Johnson, via flickr

U.S. income inequality persists amid overall growth in 2014

Posted on by Casey Schoeneberger

Income inequality in the United States grew more acute in 2014, yet the bottom 99 percent of income earners registered the best real income growth (after factoring in inflation) in 15 years. The latest data from the U.S. Internal Revenue Service show that incomes for the bottom 99 percent of families grew by 3.3 percent over 2013 levels, the best annual growth rate since 1999. But incomes for those families in the top 1 percent of earners grew even faster, by 10.8 percent, over the same period. (See Figure 1.)

Figure 1

Saez1_6262015

Overall, real average incomes per family in 2014 grew by a substantial 4.8 percent. For the bottom 99 percent of income earners, this marks the first year of real recovery from the income losses sparked by the Great Recession of 2007-2009. After a large decline of 11.6 percent from 2007 to 2009, those families saw a negligible 1.1 percent in real income gains from 2009 to 2013. But a full recovery in income growth for the bottom 99 percent is still not in sight. In 2014, these families recovered slightly less than 40 percent of their income losses due to the Great Recession.

Those at or near the top of the income ladder did substantially better in 2014. The share of income going to the top 10 percent of income earners—individuals making an average of $300,000 a year—increased to 49.9 percent in 2014 from 48.9 percent in 2013, the highest ever except for 2012. The share of income going to the top 1 percent of families—those earning on average about $1.3 million a year—increased to 21.2 percent in 2014 from 20.1 percent in 2013. Income inequality, then, remains extremely high, particularly at the very top of the income ladder. (See Figure 2.)

Figure 2

Saez2_6262015

More broadly, the top 1 percent of families captured 58 percent of total real income growth per family from 2009 to 2014, with the bottom 99 percent of families reaping only 42 percent.

The release time for this latest income data from the IRS usually lags behind other key indicators of U.S. economic performance. Aggregate economic growth statistics are typically available a month after the end of each quarter. In contrast, U.S. Census Bureau official income and poverty measures are not available until mid-September of the following year, or 8.5 months after the end of the year. Complete individual income tax statistics, the only statistics that can capture top incomes, are usually not available until 19 months after the end of the year.

This difference in timing explains why economic growth statistics are much more widely discussed than income inequality statistics in public debates about economic inequality and growth. But for the first time, this income data is now more readily available because the Statistics of Income division of the IRS now publishes filing-season statistics by size of income. These statistics can be used to project the distribution of incomes for the full year.

My colleagues and I used these new statistics to update our top income share series for 2014, which are part of our World Top Incomes Database. These statistics measure pre-tax cash market income excluding government transfers such as the disbursal of the earned income tax credit to low-income workers. For the first time, we can produce inequality statistics less than 6 months after the end of the year.

Timely statistics on economic inequality are central to informing the public policy debate about the connections between economic growth and inequality. One case in point: The higher tax rates for top U.S. income earners enacted in 2013 as part of the Obama Administration and Congress’ federal budget deal seem to have had only a fleeting impact on the outsized accumulation of pre-tax income by families in the top 1 percent and 0.1 percent of income earners.

To be sure, there was a shifting  of income among high-income earners from 2013 to 2012 as these wealthy families sought to avoid the higher rates enacted in 2013. This adjustment created a spike in the share of top incomes accumulated by the very wealthy in 2012 followed by a trough in 2013. By 2014, however, top incomes shares were back to their upward trajectory. This suggests that the higher tax rates starting in 2013, while not negligible, will not be sufficient by themselves to curb the enormous increase in pre-tax income concentration that has taken place in the United States since the 1970s.

—Emmanuel Saez in a professor of economics at the University of California-Berkeley and a member of the Washington Center for Equitable Growth’s steering committee

Exploding wealth inequality in the United States

Posted on by Ed Paisley

There is no dispute that income inequality has been on the rise in the United States for the past four decades. The share of total income earned by the top 1 percent of families was less than 10 percent in the late 1970s but now exceeds 20 percent as of the end of 2012.  A large portion of this increase is due to an upsurge in the labor incomes earned by senior company executives and successful entrepreneurs. But is the rise in U.S. economic inequality purely a matter of rising labor compensation at the top, or did wealth inequality rise as well?

Before we answer that question (hint: the answer is a definitive yes, as we will demonstrate below) we need to define what we mean by wealth. Wealth is the stock of all the assets people own, including their homes, pension saving, and bank accounts, minus all debts. Wealth can be self-made out of work and saving, but it can also be inherited. Unfortunately, there is much less data available on wealth in the United States than there is on income. Income tax data exists since 1913—the first year the country collected federal income tax—but there is no comparable tax on wealth to provide information on the distribution of assets. Currently available measures of wealth inequality rely either on surveys (the Survey of Consumer Finances of the Federal Reserve Board), on estate tax return data, or on lists of wealthy individuals, such as the Forbes 400 list of wealthiest Americans.

Download the pdf version of this brief for a complete list of sources

In our new working paper, “Wealth Inequality in the United States since 1913: Evidence from Capitalized Income Tax Data,” we try to measure wealth in another way.  We use comprehensive data on capital income—such as dividends, interest, rents, and business profits—that is reported on individual income tax returns since 1913. We then capitalize this income so that it matches the amount of wealth recorded in the Federal Reserve’s Flow of Funds, the national balance sheets that measure aggregate wealth of U.S. families. In this way we obtain annual estimates of U.S. wealth inequality stretching back a century.

Wealth inequality, it turns out, has followed a spectacular U-shape evolution over the past 100 years. From the Great Depression in the 1930s through the late 1970s there was a substantial democratization of wealth. The trend then inverted, with the share of total household wealth owned by the top 0.1 percent increasing to 22 percent in 2012 from 7 percent in the late 1970s. (See Figure 1.) The top 0.1 percent includes 160,000 families with total net assets of more than $20 million in 2012.

Figure 1

102014-wealth-web-01

Figure 1 shows that wealth inequality has exploded in the United States over the past four decades. The share of wealth held by the top 0.1 percent of families is now almost as high as in the late 1920s, when “The Great Gatsby” defined an era that rested on the inherited fortunes of the robber barons of the Gilded Age.

In recent decades, only a tiny fraction of the population saw its wealth share grow. While the wealth share of the top 0.1 percent increased a lot in recent decades, that of the next 0.9 percent (families between the top 1 percent and the top 0.1 percent) did not. And the share of total wealth of the “merely rich”—families who fall in the top 10 percent but are not wealthy enough to be counted among the top 1 percent—actually decreased slightly over the past four decades. In other words, family fortunes of $20 million or more grew much faster than those of only a few millions.

The flip side of these trends at the top of the wealth ladder is the erosion of wealth among the middle class and the poor. There is a widespread public view across American society that a key structural change in the U.S. economy since the 1920s is the rise of middle-class wealth, in particular because of the development of pensions and the rise in home ownership rates. But our results show that while the share of wealth of the bottom 90 percent of families did gradually increase from 15 percent in the 1920s to a peak of 36 percent in the mid-1980, it then dramatically declined. By 2012, the bottom 90 percent collectively owns only 23 percent of total U.S. wealth, about as much as in 1940  (see Figure 2.)

Figure 2

102014-wealth-web-03

The growing indebtedness of most Americans is the main reason behind the erosion of the wealth share of the bottom 90 percent of families. Many middle class families own homes and have pensions, but too many of these families also have much higher mortgages to repay and much higher consumer credit and student loans to service than before. For a time, rising indebtedness was compensated by the increase in the market value of the assets of middle-class families. The average wealth of bottom 90 percent of families jumped during the stock-market bubble of the late 1990s and the housing bubble of the early 2000s. But it then collapsed during and after the Great Recession of 2007-2009.  (See Figure 3.)

Figure 3

102014-wealth-web-02

Since the housing and financial crises of the late 2000s there has been no recovery in the wealth of the middle class and the poor. The average wealth of the bottom 90 percent of families is equal to $80,000 in 2012—the same level as in 1986. In contrast, the average wealth for the top 1 percent more than tripled between 1980 and 2012. In 2012, the wealth of the top 1 percent increased almost back to its peak level of 2007. The Great Recession looks only like a small bump along an upward trajectory.

How can we explain the growing disparity in American wealth? The answer is that the combination of higher income inequality alongside a growing disparity in the ability to save for most Americans is fuelling the explosion in wealth inequality. For the bottom 90 percent of families, real wage gains (after factoring in inflation) were very limited over the past three decades, but for their counterparts in the top 1 percent real wages grew fast. In addition, the saving rate of middle class and lower class families collapsed over the same period while it remained substantial at the top. Today, the top 1 percent families save about 35 percent of their income, while bottom 90 percent families save about zero.

The implications of rising wealth inequality and possible remedies

If income inequality stays high and if the saving rate of the bottom 90 percent of families remains low then wealth disparity will keep increasing. Ten or twenty years from now, all the gains in wealth democratization achieved during the New Deal and the post-war decades could be lost. While the rich would be extremely rich, ordinary families would own next to nothing, with debts almost as high as their assets. Paris School of Economics professor Thomas Piketty warns that inherited wealth could become the defining line between the haves and the have-nots in the 21st century. This provocative prediction hit a nerve in the United States this year when Piketty’s book “Capital in the 21st Century” became a national best seller because it outlined a direct threat to the cherished American ideals of meritocracy and opportunity.

What should be done to avoid this dystopian future? We need policies that reduce the concentration of wealth, prevent the transformation of self-made wealth into inherited fortunes, and encourage savings among the middle class. First, current preferential tax rates on capital income compared to wage income are hard to defend in light of the rise of wealth inequality and the very high savings rate of the wealthy. Second, estate taxation is the most direct tool to prevent self-made fortunes from becoming inherited wealth—the least justifiable form of inequality in the American meritocratic ideal. Progressive estate and income taxation were the key tools that reduced the concentration of wealth after the Great Depression. The same proven tools are needed again today.

There are a number of specific policy reforms needed to rebuild middle class wealth.  A combination of prudent financial regulation to rein in predatory lending, incentives to help people save—nudges have been shown to be very effective in the case of 401(k) pensions—and more generally steps to boost the wages of the bottom 90 percent of workers are needed so that ordinary families can afford to save.

One final reform also needs to be on the policymaking agenda: the collection of better data on wealth in the United States. Despite our best efforts to build wealth inequality data, we want to stress that the United States is lagging behind in terms of the quality of its wealth and saving data. It would be relatively easy for the U.S. Treasury to collect more information—in particular balances on 401(k) and bank accounts—on top of what it already collects to administer the federal income tax. This information could help enforce the collection of current taxes more effectively and would be invaluable for obtaining more precise estimates of the joint distributions of income, wealth, saving, and consumption. Such information is needed to illuminate the public debate on economic inequality. It is also required to evaluate and implement alternative forms of taxation, such as progressive wealth or consumption taxes, in order to achieve broad-based and sustainable economic growth.

Emmanuel Saez is a professor of economics and director of the Center for Equitable Growth at the University of California-Berkeley. Gabriel Zucman is an assistant professor of economics at the London School of Economics.