What are the key factors affecting US middle-class incomes?

Jason Furman
A hand holding a looking glass by a lake
Crowdsource Innovation
Get involved with our crowdsourced digital platform to deliver impact at scale

The US economy has strengthened considerably. GDP grew at an annual rate of 2.8% over the past two years, compared with 2.1% in the first three-and-a-half years of the recovery. The improvement was particularly clear in the labour market, where 2014 recorded the highest job gains – on both an absolute and a percentage basis – since the late 1990s. The unemployment rate has declined faster than expected as the labour force participation rate has stabilised. Indeed, the underlying trends of the overall recovery are both strong today and indicative of continued strength.

This robust aggregate growth presents an opportunity to address a long-standing challenge for the US economy – the 40-year stagnation in incomes for the middle class and those working to get into the middle class, shown in Figure 1. As strong as the recovery has been, not all households have fully shared in the income growth. This column provides historical and international context for the key factors affecting middle-class incomes: productivity growth, labour force participation, and income inequality. The President’s approach to economic policies – what he terms “middle-class economics” – is designed to improve all three. These policies are described in more detail in the recently released Economic Report of the President (Council of Economic Advisers 2015).

Figure 1. US middle-class income growth

Sources: World Top Incomes Database, Census Bureau, Congressional Budget Office.
Note: Income levels from the World Top Incomes Database and the Census Bureau are deflated with the CPI-U-RS price index.

The evolution of middle-class incomes in the postwar period

Middle-class incomes have historically been shaped by three factors: how productivity has grown, how income is distributed, and how many people are participating in the labour force. Although many of these factors have evolved continuously, varying from year to year, it is instructive to divide the post-WWII years into three periods that capture major differences in the trends in these three variables. Specifically, these periods are: the Age of Shared Growth from 1948 to 1973, when movements in productivity, participation, and distribution aligned; the Age of Expanded Participation from 1973 to 1995, when women entered the labour force at a rapid pace but productivity slowed and distribution worsened; and the Age of Productivity Recovery from 1995 through 2013, when productivity improved (at least until the run-up to the financial crisis) but participation declined and income inequality continued to worsen.

Table 1. US middle-class income growth and its determinants, 1948–2013

Sources: World Top Incomes Database, Census Bureau, Congressional Budget Office, Bureau of Labor Statistics, Bureau of Economic Analysis, Council of Economic Advisers calculations, Saez (2015).
Notes: Income levels from the World Top Incomes Database and the Census Bureau are deflated with the CPI-U-RS price index, and income levels from the Congressional Budget Office (CBO) are deflated with the personal consumption expenditures price index. Income shares are provided by the World Top Incomes Database, median household income is provided by the US Census Bureau, and median household income including benefits, transfers, and taxes is provided by the CBO. CBO median income is extended before 1979 and after 2010 with the growth rate of Census median income.

The Age of Shared Growth (1948–1973)

All three factors – productivity growth, distribution, and participation – aligned to benefit the middle class from 1948 to 1973. The US enjoyed rapid labour productivity growth, averaging 2.8% annually. Income inequality fell, with the share of income going to the top 1% falling by nearly one-third, while the share of income going to the bottom 90% rose slightly. Household income growth was also fuelled by the increased participation of women in the workforce. Prime-age (25 to 54) female labour force participation escalated from one-third in 1948 to one-half by 1973. The combination of these three factors increased the average income for the bottom 90% of households by 2.8% a year over this period. This measure functions as a decent proxy for the median household’s income growth because it ignores the large, asymmetric changes in income for the top 10% of households. At this rate, incomes double every 25 years, or about once every generation.

While these levels of shared income growth and low income inequality worked to benefit the middle class, it is important to recognise that these factors do not capture the many non-economic dimensions (such as racial and gender discrimination) on which the US has made considerable progress over the past half-century. Accordingly, while this period illustrates the combined power of productivity, equality, and participation to benefit the middle class, it is not necessarily a model for other important aspects of domestic policy.

The Age of Expanded Participation (1973–1995)

Starting in 1973 and running through 1995, two of the three factors that had been driving middle-class incomes derailed. Labour productivity growth slowed dramatically to only 1.4% annually, in part due to the exhaustion of pent-up innovations from WWII, reduced public investment, dislocations associated with the breakup of the Bretton Woods international monetary system, and the oil shocks of the 1970s. Not only did the economy grow more slowly in these years, but these smaller gains were distributed increasingly unequally – the share of national income that went to the top 1% nearly doubled, while the share that went to the bottom 90% fell accordingly. As a result, productivity gains did not boost middle-class incomes, and average income in the bottom 90% declined by 0.4% a year during these years. One important factor that prevented a larger fall in middle-class incomes was greater labour force participation. The share of dual-income households rose as women surged into the labour force even faster than in the Age of Shared Growth.

Some alternative and likely more accurate measures of middle-class income show slight increases during these years. Real median household income as measured by the Census Bureau rose by 0.2% a year from 1973 to 1995. And after including employer-paid health premiums and adjusting for changing family size, the Congressional Budget Office (CBO) estimates that median income climbed 0.4% a year, and 0.7% a year after taxes and transfers. But regardless of how it is measured, middle-class income growth clearly slowed dramatically over this period.

The Age of Productivity Recovery (1995–2013)

The third period is defined as lasting from 1995 through 2013, though it will take a longer perspective to understand whether and how the Great Recession and the current recovery fit into this period. Amid the worst recession since the Great Depression, the average real income for households in the bottom 90% declined at a 0.2% annual rate during these years. When including employer-paid health premiums and adjusting for family size, median income rose 0.4% a year according to CBO data – still considerably slower than in the Age of Shared Growth. Largely as a result of substantial tax cuts, post-tax and post-transfer incomes rose at a 1.3% average annual rate in this third period.

Labour productivity grew at a 2.3% annual rate over the period as a whole, near the rates achieved in the first era, fuelled by a new economy that made unprecedented advances in the production and use of information technology. However, these gains did little to contribute to rising wages for the middle class as the trend of worsening inequality from the previous era continued into this period. The share of income going to the bottom 90% fell to 53%, well below the 68% earned by this group in 1973. Meanwhile, the labour force participation rate fell as women’s entry into the workforce plateaued and even started to drift down, albeit at one-half the pace of the decline in prime-age male participation – a notable trend over the entire postwar era. After 2008, the retirement of the baby boomers added to the decline in participation.

While productivity growth was high on average from 1995 to 2013, it varied substantially within this period. It was higher from 1995 to 2005, declined prior to the start of the crisis, and then was adversely affected by the crisis itself. Understanding the degree to which the years 1995 through 2013 should be considered a single regime for the productivity growth rate, or one with an adverse break in the trend during or just before the crisis, will take many more years of data and analysis.

The importance of productivity, inequality, and participation

As productivity, the income distribution, and participation evolved over the past 65 years, middle-class incomes went from doubling once in a generation to showing almost no growth at all by some measures. But if these three factors had recently continued the strong trends observed in earlier periods, the outcome for typical families would be quite different. Four counterfactual thought experiments give a sense of the magnitudes involved in this dramatic change:

  • The impact of higher productivity growth. What if productivity growth from 1973 to 2013 had continued at its pace from the previous 25 years? In this scenario, incomes would have been 58% higher in 2013. If these gains were distributed proportionately in 2013, then the median household would have had an additional $30,000 in income.
  • The impact of greater income equality. What if inequality had not increased from 1973 to 2013, and instead the share of income going to the bottom 90% had remained the same? Even using the actual slow levels of productivity growth over that period, the 2013 income for the typical household would have been 18%, or about $9,000, higher.
  • The impact of expanded labour force participation. What if female labour force participation had continued to grow from 1995 to 2013 at the same rate that it did from 1948 to 1995 until it reached parity with male participation? Assuming that the average earnings for working women were unchanged, and maintaining the actual histories of productivity and income distribution, the average household would have earned 6% more in 2013, or an additional $3,000.
  • The combined impact of all three factors. Finally, if all three factors had aligned – if productivity had grown at its Age of Shared Growth rate, inequality had not increased, and participation had continued to rise – then these effects would have been compounded and the typical household would have seen a 98% increase in its income by 2013. That is an additional $51,000 a year.

In combination, these factors would have nearly doubled the typical household’s income had they sustained their more favourable readings from earlier historical periods. Productivity, inequality, and participation constitute the fundamental challenges facing the future of middle-class incomes, and President Obama’s policies are designed to strengthen all three.

Table 2. Counterfactual scenarios for productivity, inequality, and participation

Sources: World Top Incomes Database, Census Bureau, Congressional Budget Office, Bureau of Labour Statistics, Current Population Survey, Bureau of Economic Analysis, Council of Economics Advisers calculations.
Notes: These thought experiments are intended to demonstrate the importance of these three factors for middle-class incomes. They do not consider second-order effects or interactive effects. The first thought experiment assumes that an increase in productivity is associated with an equal increase in the Census Bureau’s mean household income. The second thought experiment uses the Census Bureau’s mean income of the middle quintile as a proxy for median income. The third thought experiment assumes that newly participating women will have the same average earnings as today’s working women, and halts the growth of female labour force participation when it matches male participation. The first and third thought experiments assume that income gains are distributed proportionally such that mean and median incomes grow at the same rate. Dollar gains are calculated off a base of the Census Bureau’s median household income in 2013. The fourth thought experiment compounds the effects of the first three.

The drivers of middle-class Incomes: An international comparison

A wide range of advanced economies has faced similar challenges as the US regarding middle-class incomes. Most of today’s large advanced economies experienced rapid growth in the immediate post-WWII years followed by substantially slower growth and plateauing, as shown in Figure 2. That development took place relatively early in the US (around 1973) and later in other countries (for example, around 1980 in France and Canada). In Japan, middle-class incomes slowed in the 1970s and have substantially declined over the past two decades.

Figure 2. Growth in real average income for the bottom 90%

150220-bottom 90 percent income growth


Sources: World Top Incomes Database, Saez (2015), Council of Economic Advisers calculations.
Notes: Data for all countries exclude capital gains. For Germany, data excluding capital gains is unavailable after 1998, so this chart displays data including capital gains adjusted for the historical relationship between capital-inclusive and capital-exclusive incomes. Italian data begins in 1974 and is indexed to the average of the other series at that point. Italian data is calculated by the Council of Economic Advisers from the income level and share of the top 10% as provided by the World Top Incomes Database.

Labour productivity growth

The first driver of incomes – labour productivity growth – underlies the progress of both potential GDP and family income. Over the past year, the Organisation for Economic Co-operation and Development (OECD) and the International Monetary Fund (IMF) reduced their productivity growth estimates for many high-income countries. In recent years, the US has been somewhat better situated than many other advanced economies, in part because this country has been the centre of much high-tech innovation. In fact, the US has defied the trend in other high-income economies by experiencing a pickup in productivity growth over the last 20 years. In contrast, productivity growth has generally declined in most other high-income economies over the same period, as shown in Figure 3.

Figure 3. Labour productivity growth

Source: Conference Board, Total Economy Database, Council of Economic Advisers calculations.

Income inequality

The second important factor influencing the dynamics of middle-class incomes is inequality. This, too, is a global issue. In the US, the top 1% has garnered a larger share of income than in any other G-7 country in each year since 1987 for which data are available, as shown in Figure 4. From 1990 to 2010, the top 1%’s income share rose 0.22 percentage points a year in the US versus 0.14 percentage points a year in the UK. While comparable international data are scarce after 2010, the gains of the top 1% continued since then in the US, until a noticeable downtick in 2013.

Figure 4. Share of income earned by top 1%, 1975–2013

Sources: World Top Incomes Database, Saez (2015).
Notes: Data for all countries exclude capital gains. For Germany, data excluding capital gains is unavailable after 1998, so this chart displays data including capital gains adjusted for the historical relationship between the capital-inclusive and capital-exclusive ratios.

Labour force participation

The third driver of income growth is labour force participation. Although the US has enjoyed a strong labour market recovery amid surging employment, its labour force participation rate has fallen more than that of other high-income countries.

The recent decline in the labour force participation rate is largely the result of demographic changes. Since 2008, when the first of the baby boomers turned 62 and became eligible for Social Security, the baby boom has become a retirement boom. This loss of productive workers was compounded by the severe recession that hit around the same time. But even before either of these events, the economy already faced labour force participation challenges, including a long-running decline in male labour force participation and an end to the rapid increase in female participation.

Since the early 1990s, the US has experienced a marked decline in labour force participation among males aged 25 to 54 (‘prime age’), as shown in Figure 5. In this regard, the US experience has been something of an outlier compared to many other high-income countries. Since the financial crisis, US prime-age male participation has declined by about 2.5 percentage points, while the UK has seen a small uptick and most large European economies were generally stable. Of 24 OECD countries that reported prime-age male participation data between 1990 and 2013, the US fell from 16th to 22nd.

Figure 5. Prime-age male labour force participation rates

Source: OECD.

The story is somewhat similar among prime-age females. Historically, the US showed leadership in bringing women into the workforce. In 1990, the US ranked 7th out of 24 current OECD countries reporting prime-age female labour force participation – about 8 percentage points higher than the average of that sample. But since the late 1990s, women’s labour force participation plateaued and even started to drift down in the US while continuing to rise in other high-income countries, as shown in Figure 6. As a result, in 2013 the US ranked 19th out of those same 24 countries, falling 6 percentage points behind the UK and 3 percentage points below the sample average. A recent study found that the relative expansion of family leave and part-time work programmes in other OECD countries versus the US explains nearly one-third of the US’s relative decline (Blau and Kahn 2013).

Figure 6. Prime-age female labour force participation rates

Source: OECD.

The President’s approach to middle-class economics

The history of middle-class incomes has important policy implications. President Obama’s economic agenda is designed to improve each of the three key factors that drive middle-class incomes: productivity, labour force participation, and the income distribution.

Labour is more productive when it is mixed with productive capital. That is why the President supports substantial investment in infrastructure, an important component of the US’s physical capital, to repair crumbling bridges and roads. Workers are also more productive when they are better educated, and the President supports a range of new education investments from pre-kindergarten to college to help develop the US’s human capital. Moreover, the President’s approach to business tax reform boosts productivity through several distinct channels, including by encouraging domestic investment from the private sector and reducing the inefficiencies of the international tax system – while also financing infrastructure. All of these fiscal policies are complemented by expanding trade, including through the Transpacific Partnership (TPP) and the Transatlantic Trade and Investment Partnership (TTIP).

Labour force participation also matters. Although it is perfectly reasonable for labour force participation to fall as the population ages and people retire, in the US the participation rate for the prime-age men has been declining for decades and for women has plateaued and started to decline – trends that have been exacerbated by the recession and recovery. Expanding job creation would help create new opportunities and draw people into the workforce. Moreover, the President has proposed new tax credits for childcare and second earners in working families, as well as policies that promote workplace flexibility and paid sick leave. In addition, the President’s executive actions on immigration will help increase the labour force, though he supports comprehensive immigration reform that could do even more.

To help reduce income inequality and ensure that more of the economic growth is shared, the President supports increasing the minimum wage – now that its real value has declined 20% over the past three decades, and expanding the Earned Income Tax Credit to low-income workers without children and non-custodial parents. Moreover, continued investments in education and skills training will also curb the trend of worsening inequality by helping ensure that children of all backgrounds have much to offer the 21st century labour market. These measures build on earlier progressive steps to reduce inequality, including the Affordable Care Act’s exchange subsidies and Medicaid expansion, as well as the President’s progressive reforms to the individual income tax code.

Although workers have begun to reap the benefits of the US’s recovery, this has just served to reveal the long-standing challenges we face in terms of productivity growth, labour force participation, and inequality. As the US economy moves towards complete recovery, President Obama’s focus on middle-class economics is designed to foster growth in a shared and sustainable way.

This article is published in collaboration with Vox EU. Publication does not imply endorsement of views by the World Economic Forum.

To keep up with the Agenda subscribe to our weekly newsletter.

Author: Jason Furman is Chairman, Council of Economic Advisers.

Image: Shadows are seen on an American flag. REUTERS/Mike Segar

Don't miss any update on this topic

Create a free account and access your personalized content collection with our latest publications and analyses.

Sign up for free

License and Republishing

World Economic Forum articles may be republished in accordance with the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Public License, and in accordance with our Terms of Use.

The views expressed in this article are those of the author alone and not the World Economic Forum.

World Economic Forum logo
Global Agenda

The Agenda Weekly

A weekly update of the most important issues driving the global agenda

Subscribe today

You can unsubscribe at any time using the link in our emails. For more details, review our privacy policy.

About Us



Partners & Members

  • Join Us

Language Editions

Privacy Policy & Terms of Service

© 2024 World Economic Forum