Chapter 1. Overview

This introductory chapter gives an overview of the entire report drawing on the analyses carried out in the four subsequent chapters. The report looks at the multiple pressures on the middle class, in terms of their economic situation, their living expenses, and labour market insecurity. It documents that in many OECD countries, middle incomes have grown less than upper incomes and in some, they have not grown at all. The costs of some goods and services, such as housing, which are core for a middle class lifestyle, have risen faster than real median incomes. The report sheds light on the rising job insecurity for middle-income households in rapidly transforming labour markets, where middle-income jobs face high risk of automation. Finally, it discusses policy measures and initiatives to address the challenges with which middle-income households contend - boosting incomes, coping with increased costs of living, improving the fairness of taxation and building skills of middle-class households.


The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

1.1. The middle class dream is increasingly only a dream for many

Over the past decade, the OECD has reviewed the evolution of inequalities in income and opportunities in details and as part of its Inclusive Growth Initiative (OECD, 2019[1]; OECD, 2018[2]) (see Box 1.1 for an overview). The last flagship report of the inequality series focused on social mobility (OECD, 2018[3]). It showed that in many OECD countries, opportunities for low and middle-income families to move up the ladder became limited over the past decades. Moreover, many middle-class families faced a growing risk of falling down to a lower income or a lower status.

The present report zooms in onto the middle class as an economic and social group and sheds light on the multiple mounting pressures on this group. 1 In particular, the report analyses the main trends affecting “middle-income” households – their employment, consumption, wealth and debt, as well as perceptions and social attitudes. It finds that the middle class is being squeezed, both in the sense that the share of people in the middle income group has declined over time, especially among the new generations, and that the cost of a middle class lifestyle is rising faster than middle incomes. The report also discusses how policies should respond to these growing pressures for the middle class.

Why a focus on the middle class? The middle class used to be an aspiration. For many generations it meant the assurance of living in a comfortable house and affording a rewarding lifestyle, thanks to a stable job with career opportunities. It was also a basis from which families aspired to an even better future for their children. However, there are now signs that this bedrock of our democracies and economic growth is not as stable as in the past.

Is the middle-class under pressure? Middle incomes have barely grown, in both relative and absolute terms in many OECD countries; the cost of essential parts of the middle-class lifestyle has increased faster than income, notably housing and higher education; and job insecurity has risen in the context of fast transforming labour markets. Today, the middle class looks increasingly like a boat in rocky waters.

Is this insecurity temporary or here to stay? Fundamentally, rising uncertainty stems from fewer opportunities to climb up the ladder for middle-class people and their children than in the past, and higher risks to slide down. Social mobility at different stages of life is limited: education attainments, occupational status, earnings, and even health status are very persistent across generations. Two fifths of the earnings differences between fathers carry over to the next generation (OECD, 2018[4]). Many middle-income households face a considerable risk of sliding down into the lower-income class: one-in-seven households in the middle 60% of the income distribution and one-in-five of those living in the second-lowest income quintile slide into the bottom 20% over a four-year period. These risks have increased over the past two decades in many OECD countries. At the same time, more affluent middle-income households have somewhat lower risks today. This indicates a rising chance of fracturing among the middle-income class (OECD, 2018[4]).

What are the prospects for the middle-class? The labour market prospects of many middle-class workers looks uncertain. Already in the past two decades, rapid integration along global supply chains and, even more so, fast and transformative technological change and population ageing have resulted in a polarisation of jobs. This shift in employment towards high-skilled non-routine jobs and some low-skilled non-routine jobs has hollowed-out middle-skilled jobs. Those are often held by middle-class people (OECD, 2017[5]; Autor, 2015[6]). In the future, many new job opportunities will emerge with further penetration of new digital technologies – artificial intelligence, machine learning, internet of things etc. – but many jobs are likely to be automated or profoundly overhauled (Nedelkoska and Quintini, 2018[7]).

These trends paint an uncertain picture for workers with middle incomes, in particular those with low-medium skills in routine jobs. This comes on top of the secular decline of productivity gains and the shrinking labour share observed in many countries, which have weighed on the purchasing power of most middle-income families.

Box 1.1. Recent OECD work on inequalities of opportunities and outcomes

A number of recent OECD reports have highlighted structural inequalities that persist in social and economic outcomes in various domains. The evidence suggests that income inequality has increased in three quarters of the OECD countries over the past three decades and remains even higher in most emerging economies, albeit declining in some (OECD, 2011[8]). High and persistent income inequalities have negative consequences not only for social cohesion in our societies but also for economic growth to the extent they undermine the opportunities to access quality education and health services as well as good jobs (OECD, 2015[9]). OECD PISA studies (e.g. (OECD, 2018[10]) and OECD work on educational opportunities (OECD, 2017[11]) have clearly shown that too many children, students and adults from socio-economically disadvantaged backgrounds fall behind. Gender inequalities persist in all areas of social and economic life (OECD, 2017[12]). Such disparities in education, in income and resources go together with disparities in life satisfaction (OECD, 2017[13]). Moreover, inequalities compound over the life course leading to a process of unequal ageing (OECD, 2017[14]).

The OECD’s most recent inequality report A Broken Social Elevator? How to Promote Social Mobility drew attention on the increasing lack of social mobility in our societies. Socio-economic status heavily influences income and employment prospects, job quality, health outcomes, education, and other opportunities that matter to people’s well-being (OECD, 2018[4]). The analyses found that in many OECD countries, opportunities for low and middle-income families to move up the ladder became limited as the steps became larger. At the same time, many families witness a growing risk to fall down to lower income or a lower status.

The report makes a key contribution to the OECD’s Inclusive Growth initiative and its effort to develop a “people-centred inclusive growth model” in which well-being is the metric of success; where everyone has an equal opportunity to prosper; and where equity considerations are important in defining effective economic policies (OECD, 2018[2]).

1.2. The middle class is an engine for prosperity and economic growth

Why the middle class matters? The presence of a strong and prosperous middle class supports healthy economies and societies. Through their actions and activities, they improve not only their own position, but also that of others. The investment of the middle class in education, health, and housing, their support for good quality public services, their intolerance of corruption, and their trust in others and in democratic institutions, are the very foundations of inclusive growth.

In particular, middle classes generally invest heavily in their own education and that of their children, thus increasing current and future stock of human capital. (Brown and Hunter, 2004[15]; Pressman, 2007[16]; Bassanini and Scarpetta, 2002[17]). Human capital accumulation is therefore an important channel through which a strong middle class can support wellbeing and GDP per capita growth, as it boosts that share of population with secondary and tertiary education (Brueckner et al., 2017[18]; OECD, 2003[19]).

A strong middle class is also essential to support sustained productivity growth and to properly fund social protection and lift millions of families out of poverty. The middle class also contributes to economic growth and capital accumulation as a source of entrepreneurship and innovation. In countries with more middle-income households, entrepreneurship activities tends to have a positive impact on GDP growth (van Stel, et al., 2005[20]). In this sense, a strong middle class is considered an important element to foster small and medium-sized enterprises and to grow a robust entrepreneurial sector (OECD, 2010[21]). Overall, the evidence is clear: economic growth is stronger in countries where the middle class is strong (Easterly, 2001[18]; Brueckner et al., 2018[19])

1.3. Most people identify themselves as belonging to the middle class

Who is in the middle class? The majority of the people in OECD countries – on average around two-thirds of the population – think of themselves as part of the middle class. Such self-identification is highest in Nordic countries (except Finland), the Netherlands, Luxembourg and Switzerland, where four out of five people think they are part of the middle class. In contrast, just two out of five people or less identify with the middle class in Portugal, Brazil, Chile and the United Kingdom. In the United Kingdom, some studies suggest most people identify as working class.2

Many people self-identify with the middle class regardless of their own objective socio-economic circumstances. Such “middle class bias(Evans, 2004[22]) seems particularly high in Italy, Mexico and India, and in countries where large shares of the population identify themselves as “middle class”.3 This phenomenon can be partly explained by the implicit income level that people consider as a minimum threshold for being middle class: this level tends to be substantially lower for poorer people and noticeably higher for richer people.4

The extent of such self-identification with the middle class is only loosely related to the shares of middle-income households across countries. Middle-income households defined as households earning between 75% and 200% of the median national income (see Box 1.2) make up, on average, 61% of the population in OECD countries. This share ranges from around 50% in Chile, Mexico, the United States and Israel to around 70% in Nordic and some Continental European countries. In emerging economies, the middle-income group is smaller and ranges from around one third of the population in South Africa to about half in Russia (Figure 1.1).

Box 1.2. Defining and measuring middle class

Social class refers to a group of people who share the same socio-economic status. How this status is defined differs according to different academic traditions and the use to which the concept is being put. As a consequence, indicators used to define and analyse the “middle class” vary significantly, within and across disciplines (Reeves, Guyot and Krause, 2018[23]; Gornick and Jäntii, 2013[24]). Many economic analyses use income as key indicator to define middle class status. In the rest of the social sciences, most measures and indicators are based on occupation and employment status (Goldthorpe, 2016[25]), although social, cultural and economic capital are also used (Savage et al., 2013[26]; Savage, 2016[27]). The middle class has also been defined using subjective measures such as social class identification (Bird and Newport, 2017[28]).

This report uses income as a benchmark to analyse the middle class and highlights this choice by mainly using the term “middle-income class”, rather than “middle class”. That said, the report goes beyond the pure income perspective by analysing in detail the importance of job polarisation (Chapter 3) and consumption (Chapter 4) for the middle class. In both these cases, however, the middle-income class remains the benchmark.

This report defines “middle-income class” as the population living in households with incomes ranging between 75% and 200% of the national median. For some of the analyses, the middle-income class is sub-divided into three groups: lower middle-incomes (75% to 100% of median), middle middle-income (100% to 150% of median) and upper middle-income (150% to 200% of median). The population in households with income below 75% of the median are the “lower-income class” and those with income above 200% of the median are the “upper-income class”.

Middle class indicators based on median incomes produce fairly stable cross-country comparisons. Robustness checks indicate that country rankings show little variation when using different thresholds based on proportions of median income. Income shares tend to be more stable when using people-based thresholds (income deciles) than income-based thresholds (ranges around the median).

While middle class indicators based on income are linked to the level of income inequality and macroeconomic performance, these factors alone are not sufficient to explain the evolution of income at the middle in OECD countries in recent decades (Thewissen et al., 2018[29]).

Are middle incomes being hollowed out? On average across OECD countries, the share of people in middle-income households fell from 64% to 61% between the mid-1980s and mid-2010s. The decline was spread evenly across time, roughly 1 percentage point per decade and accompanied by corresponding expansions of the lower- and upper-income classes. Albeit generally modest, changes in the size of the middle-income class have been significant in a number of OECD countries, exceeding 4.5 percentage points in Israel, Germany, Luxembourg, Canada, the United States, Finland and Sweden.

Figure 1.1. Most people consider themselves as part of the middle class
Share of population in the middle-income group and considering themselves as “middle class”, 2017 or most recent year
Figure 1.1. Most people consider themselves as part of the middle class

Note: The “middle-income” group is defined by population share with household disposable income between 75% and 200% of the national median. Incomes are disposable incomes, corrected for household size.

Source: Chapter 2.


In real terms, middle-income thresholds (75%-200% of the median income) vary a lot across OECD countries. It takes an annual income of between USD 3 800 and USD 10 000 in Mexico and between USD 26 500 and USD 70 600 in Luxembourg to be part of the middle class for a single person (Table 1.1). Still most middle-income households from one OECD country would also be considered as middle income in another. With an income of USD 23 000, for instance, people would be in the middle-income class in 25 of the 35 OECD countries for which data are available.

Table 1.1. Middle-income thresholds vary a lot across OECD countries
Lower and upper middle-income thresholds in USD 2010 PPPs, 2016 or latest year available


Lower threshold

Upper threshold


26 482

70 620

United States

23 416

62 442


19 735

52 626


18 173

48 462

United Kingdom

15 856

42 283


12 206

32 549


8 707

23 219


3 757

10 019

Note: Middle-incomes classes and median incomes defined based on equivalised household disposable income, corrected for household size. The middle-income class comprises individuals in households with income between 75% and 200% of the median. Values in USD, adjusted for international differences in 2010 purchasing power parities.

Source: Chapter 2.


1.4. The economic influence of the middle class is weakening

How has the middle class faired economically? Over the past three decades, the middle-income class has experienced dismal income growth or even stagnation in some countries. Since the mid-1980s, middle incomes have grown significantly less than higher incomes, and the global financial crisis exacerbated this trend further. Between 2007 and 2016, the annual growth rate of real median incomes was 0.3% on average across OECD countries, compared to 1% between the mid-1980s and mid-1990s, and 1.6% between the mid-1990s and mid-2000s – a period when income growth was strongest among all income groups. Households with 20 to 40% of the median income fared even worse, especially since the early 2000s (Figure 1.2). Across the OECD area, except for a few countries, middle incomes are barely higher today than they were ten years ago.

Figure 1.2. Median incomes grew more slowly than top incomes
Real disposable income growth by income position, average for 17 OECD countries, 1985-2016 (1985 = 100%)
Figure 1.2. Median incomes grew more slowly than top incomes

Note: Unweighted average for 17 countries for which long-term data are available: Canada, Germany, Denmark, Finland, France, United Kingdom, Greece, Israel, Italy, Japan, Luxembourg, Mexico, Netherlands, Norway, New Zealand, Sweden and United States. Incomes are disposable incomes, corrected for household size.

Source: Chapter 2, OECD calculations from OECD Income Distribution Database (


Overall, over the past 30 years, median incomes increased a third less than the average income of the richest 10%. Moreover, in some countries the share of incomes at the very top has surged; in the United States, for example, the share of top 1% on total income has almost doubled from about 11% to 20% over the past three decades and almost half of all income growth over this period accrued to this group (Förster et al., 2014[30]; Saez, 2018[31]). Therefore, the economic influence of the middle class and its role as “centre of economic gravity” has weakened. Three decades ago, the aggregate income of all middle-income households was four times the aggregate income of upper income households, i.e. those with incomes above two times the national median; today, this ratio is less than three (Figure 1.3).

Figure 1.3. Middle incomes are losing economic influence
Aggregate income share ratio between middle and upper income households, OECD average mid-1980s to mid-2010s
Figure 1.3. Middle incomes are losing economic influence

Note: Data refer to the aggregate income of all middle-income households as a ratio of the aggregate income of all upper-income households. “Middle income” households defined as households with income between 75% and two times the national median. “Upper income” households defined as households with income above two times the national median. OECD average comprises available data from Australia, Canada, Switzerland, Czech Republic, Denmark, Finland, France, Germany, United Kingdom, Hungary, Ireland, Italy, Luxembourg, Netherlands, Norway, Spain, Sweden, and United States. Incomes are disposable incomes, corrected for household size.

Source: OECD Secretariat calculations based on data from EU-SILC (Europe), SLID and CIS (Canada), CPS March Supplement (United States) and LIS Data Center.


At the same time, the growth rate of middle incomes in OECD countries fell behind the income growth of the global middle-income group as well as that of the global top 1%. This is the pattern behind the so-called “Elephant curve” (Milanović, 2016[32]; Lakner and Milanovic, 2016[33]). Estimates of household income growth at the world scale between 1988 and 2011 show that income growth was strongest around the 40th to 60th percentile of the world distribution5, while it was weakest around the 85th to 90th percentile, which includes many of the households in the middle-income groups in OECD countries (Figure 1.4). This pattern, which became even stronger with the global financial crisis, has been used to describe the contrast between the fortunes of the two middle classes, the global middle-income class, mainly in Asian countries, and the OECD middle-income class (Milanović, 2019[34])

Figure 1.4. Real income growth of OECD middle income groups fell behind that of global middle-incomes and the global top 1% incomes
Real income growth over 1988-2008 and 1988-2011 (based on 2011 PPPs)
Figure 1.4. Real income growth of OECD middle income groups fell behind that of global middle-incomes and the global top 1% incomes

Note: Y-axis displays the growth rate in average income of the fractile group (in 2011 PPP USD). Population-weighted. Growth incidence evaluated at ventile groups (e.g., bottom 5%); top ventile is split into top 1% and 4% between P95 and P99.

Source: (Milanović, 2019[34]), Figure 1.3.


1.5. The cost of living has been rising

This sluggish income growth would already draw a bleak picture of the situation; but, in parallel, the cost of the middle class lifestyle has increased. Prior to the global financial crisis, consumption expenditure of middle-income households6 increased faster than their income. In several countries, the fall in consumption in the aftermath of the crisis was severe and persistent. Similar to other recessions, the decline was stronger for durables, but actually it was a fall in non-durables (especially services) that stood out from previous recessions (De Nardi, French and Benson, 2012[35]; Crossley, Low and O’Dea, 2013[36]).

Why has consumption grown more than income for the middle class? Middle-class lifestyle is typically associated with certain goods and services and certain living conditions, such as decent housing, good education and good and accessible health services. However, the prices of core consumption goods and services such as health, education and housing have risen well above inflation (Figure 1.5), while middle incomes have been lagging behind. In particular, ageing and new medical technologies have driven up the cost of health services; the race for diplomas is pressing parents to invest more and more in education while, at the same time, education services became more costly in a number of countries; the geographical polarisation of jobs is pushing up housing prices in large urban areas, precisely where most rewarding jobs are available.

Figure 1.5. Prices of housing, education and health increased faster than overall inflation
Average evolution of nominal prices, OECD average
Figure 1.5. Prices of housing, education and health increased faster than overall inflation

Notes: HICP: Harmonised Indexs of Consumer Prices. OECD average includes Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Luxembourg, Netherlands, Norway, Poland, Portugal, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, United Kingdom.

Source: Data from OECD.stat, Harmonised Indices of Consumer Prices (HICPs) by COICOP divisions.


Achieving a middle class lifestyle has thus become more difficult than in the past because of the strong rise in the prices of housing and other large middle-class consumption items (Blank, 2010[37]). Housing, in particular, is key: at around one-third of disposable income, it constitutes the largest expenditure item for middle-income households – up from around a quarter in the 1990s. Despite large within-country variations, house prices have been growing three times faster than household median income over the last two decades. Housing is more than just a standard consumption good: in many countries, being middle class is traditionally associated with owning a home, so soaring house prices have touched on the very meaning of being part of the middle class. Rising house prices also hinder labour mobility towards the most dynamic urban areas and, more generally, upward social mobility as current and future generations are less able to purchase property than their parents (Figure 1.6).

Figure 1.6. The cost of buying a flat in large cities has increased considerably for middle-income families
Number of years of annual income needed to buy a 60 square meter flat in the country's capital city or financial centre, for a median income couple with two children
Figure 1.6. The cost of buying a flat in large cities has increased considerably for middle-income families

Note: Households composed of a couple and two children with disposable median income. OECD average includes Australia, Canada, Denmark, Finland, France, Germany, Greece, Hungary, Israel, Italy, Netherlands, Norway, Sweden, Switzerland, United Kingdom, United States.

Source: OECD Secretariat calculations based on data from EU-SILC (Europe), SLID and CIS (Canada), CPS March Supplement (United States) and LIS Data Center: Global Property Guide; OECD Residential Property Prices Indices (RPPIs).


The rising expenditure on housing and other goods and services reduces the ability to save and squeezes the finances of middle-income households. Today, more than one-in-five middle-income households spend more than they earn which bears a risk of over-indebtedness for many of them. Over-indebtedness (defined as having a debt-to-asset ratio of over 75%) is higher for middle-income than for both low-income and high-income households and concerns around 11% of these households, on average across OECD countries (OECD, 2015[38]).

On average across the 24 European OECD countries with available data, one out of two middle-income households now report difficulties to make ends meet, though this ranges from one out of five or less in the Nordic countries and the Netherlands to two out of three or more in some Southern and Eastern European countries (Chapter 4). Furthermore, almost 40% of middle-income households are financially vulnerable: i.e. they are on arrears or would not be able to absorb unexpected expenses or a sudden income fall.

As a result, many are less optimistic about being part of the middle class. Concerns have grown even further following the global financial crisis. While the economic sentiment has bounced back with the economic recovery, in many OECD countries large groups of people remain pessimistic about the economic prospects of the next generation and nostalgic about the economic conditions in the past (Pew Research Center, 2018[39]). A growing number of people in OECD countries think that parents’ fortune and advantages play a major factor for children’s opportunities and success. 36% of people believe that having well-educated parents is essential or very important to get ahead – up from 31% in the 1990s (OECD, 2018[3]). In a recent OECD survey across 21 countries (OECD, 2018[40]; OECD, 2019[41]), 60% of parents list the risk that their children will not achieve the level of status and comfort that they have as one of the top-three greatest social and economic long-term risks. In several countries, including Austria, France, Greece, Italy, and Slovenia, this share rises as high as 70% or more.

1.6. Chances to reach the middle class have decreased for younger generations and those with lower and medium skills

It has become more difficult for younger generations to make it to the middle class. This is because older generations are often better protected from changes in the labour market and from low-income risks than the newer generations. Since the baby boomer generation, the middle-income group has grown smaller with each successive generation (Figure 1.7). The baby boom generation enjoyed more stable jobs during their working life than younger generations as well as well-developed pension systems.

Figure 1.7. Since the baby boomers generation, each new generation has seen its chances of belonging to the middle-income class fall
Share of population in middle-income households by generation and stage at the life cycle
Figure 1.7. Since the baby boomers generation, each new generation has seen its chances of belonging to the middle-income class fall

Note: Silent generation: born before 1943, Baby boomers: born 1943-64, Generation X: born 1965-82, Millennials: born 1983-2002; Generation Z: born since 2003. The middle-income class comprises individuals in households with income between 75% and 200% of the median. OECD average based on data from OECD average comprises available data from Canada, Denmark, Finland, France, United Kingdom, Ireland, Italy, Luxembourg, Mexico, Netherlands, Norway, Spain, Sweden, and United States.

Source: OECD Secretariat calculations based on data from LIS Data Center, expect EU-SILC for France (2014), Ireland (2014), Iceland (2014) and Sweden (2015).


The skills profile needed to make it to the middle income group has risen. Having only middle skills is no longer sufficient to belong to the middle-income group, as it used to be in the past. While the skill bar has moved up more generally, it did so particularly for the middle-income class. Today, almost half of middle-income workers are in high-skill occupations, compared to one-third two decades ago (Figure 1.8).7 At the same time, the share of middle-income workers in middle-skill jobs declined in all countries except Mexico and the Slovak Republic. Hence, some jobs are increasingly failing to deliver the same income status they gave in the past. This changing association between work skills and income status can help explaining some of the social frustration, which has been at the centre of the debate about “the squeezed middle”.

Nowadays a family households needs two earners to make it to the middle-income class, while in the past one earner with a high-skill job would often suffice. However, even the presence of two earners per se is becoming less effective in achieving middle-income status if not at least one of the partners is highly skilled.

Figure 1.8. The skills profile needed to reach the middle of the income distribution has increased
Shares of middle class workers who hold jobs in different skill groups, mid-1990s/mid-2000s
Figure 1.8. The skills profile needed to reach the middle of the income distribution has increased

Note: OECD average comprises available data from Austria, Belgium, Canada, Czech Republic, Denmark, Finland, Germany, Iceland, Ireland, Israel, Italy, Korea, Luxembourg, Netherlands, Norway, Portugal, Slovak Republic, Sweden, and United States.

Source: OECD Secretariat calculations based on data from LIS Data Center and EU-SILC.


1.7. A squeezed middle class raises economic, social and political risks

Among middle-class households, there is now a growing discontent with economic conditions. In this context, the stagnation of middle-class living standards in OECD countries has been accompanied in recent years by the emergence of new forms of nationalism, isolationism, populism and protectionism. Nationalistic and anti-globalization sentiments can arise because a shrinking middle class produces disillusionment and damages political engagement, or turns voters towards anti-establishment and protectionist policies (Bettiza, 2010[42]; Stiglitz, 2012[43]). Political instability is an important channel through which a squeezed middle class may upset economic investment and growth (Alesina and Perotti, 1996[44]).

A rising sense of vulnerability, uncertainty, and anxiety has translated into increasing distrust towards global integration and public institutions (OECD, 2017[45]). Trust is not only essential for a well-functioning society; is also an important element for economic growth. Higher social trust reduces transaction costs and increases innovation (Gould and Hijzen, 2016[46]; Madland, 2015[47]). Societies with a strong middle class experience higher levels of social trust but also better educational outcomes, lower crime incidence, better health outcomes and higher life satisfaction (Kelly, 2000[48]; Lynch and Kaplan, 1997[49]; Graves, 2017[50]; Thorson, 2014[51]).

The middle class champions political stability and good governance. It prevents political polarization and promotes greater compromise within government (Madland, 2015[47]). Middle classes also provide a solid basis to build a democratic state, not only by financing it through taxes but also demanding regulations, enforcement of contracts and the rule of law (Birdsall, 2016[52]).

1.8. Policies need to be geared to the new challenges facing middle-class households

Overall, this anxiety and feeling of uncertainty among middle-class households is fuelled by three main challenges, which need to be addressed by policy makers:

  1. 1. The middle class feels that the current socio-economic system is unfair: middle incomes have been growing much more slowly than higher incomes for more than three decades now.8 More and more people, especially among the lower end of middle-income households, feel that they contribute much more to the welfare state with taxes than they receive in return in form of benefits and services.

  2. 2. The middle-class lifestyle is increasingly expensive: the cost of a “typical” middle-class lifestyle has increased faster than income. As explained, this was driven in particular by prices for housing, health and education increasing faster than inflation; but also possibly by consumption patterns which tend to imitate behaviour of richer peers, leading to a cascade of unaffordable consumption 9. Worryingly, these trends have also lead to an increasing debt burden on middle-class families, which is not sustainable in the longer-run.

  3. 3. The middle class income prospects are increasingly uncertain: people in middle-class households are becoming anxious about the fact that the digital revolution may be destroying their jobs. As a matter of fact, one-in-six current middle-income workers are in jobs that are at high risk of automation, a risk closer to that of low-income (one-in-five) than that of high-income workers (one-in-ten) (Figure 1.9).

Figure 1.9. One-in-six current middle-income jobs face high risk of automation
Share of workers in occupations with high risk of automation by income class
Figure 1.9. One-in-six current middle-income jobs face high risk of automation

Note: “Lower income” households defined as households with income below 75% the national median. “Middle income” households are households with income between 75% and two times the national median. “Upper income” households are households with income above two times the national median. The risk of automation is calculated as the average of the risks of automation by occupation, weighted by the share of each occupation in the income class.

Source: OECD secretariat calculations based on LIS and PIAAC.


Lower middle-income households are more susceptible to suffer from these growingly unstable conditions. Therefore, policy options should concentrate on them in priority. Addressing rising concerns of the lower middle-class, while also addressing the challenges of the even poorer strata of the population through anti-poverty tools, are foundation elements of a policy strategy to promote inclusive growth, along the lines of the OECD Framework for Policy Action on Inclusive Growth (OECD, 2018[2]). Supporting the middle-class requires giving them access to the opportunities to succeed along their life.

1.8.1. Tackling unfairness

The main tool to foster fairness is the tax and benefit system. Public debate often focuses on taxes when referring to the rich and to benefits when referring to the poor. The middle class is in fact the key stakeholder of both instruments – contributing two thirds of direct tax revenues, and receiving 60% of public spending on cash benefits. The middle class is also more affected by tax progressivity than other groups: lower income groups are often exempted, while higher income groups have other sources of income and can exploit loopholes or exemptions. OECD countries have seen a general reduction in their top statutory personal income tax rates which fell from 66% in 1981 to 41% in 2008, and then slightly increased, being at 43% today (OECD, 2019[53]). Other taxes that mainly concern top earners were also lowered in past decades, notably corporate income tax rates (from 47% to 25%) or taxes on dividend income for distributions of domestic source profits (from 75% to 42%) (OECD, 2014[54]). Moreover, households at the top of the income and wealth distribution are significantly more likely to hide assets offshore than the middle classes (Alstadsaeter, Johannesen and Zucman, 2017[55]).

Therefore, policies should consider effectively lowering the net tax burden on middle-income households while maintaining the sustainability of public finances. In many countries, the income tax system could be made more progressive, in particular for top income earners, and fairer for the middle class. Eliminating the “bracket creep” – an inflation-induced increase in tax rates which affects middle-class tax payers in particular – could be considered. More generally, the tax burden should be shifted from labour to broader bases, including income from capital and capital gains, property and inheritance. Efforts to fight tax avoidance and tax exemptions generally benefitting wealthy individuals and corporations should be intensified.

A number of recent initiatives, such as the adoption of the automatic exchange of taxpayer financial account information (AEOI) to reduce tax evasion; and the OECD/G20 Base Erosion and Profit Shifting (BEPS) project to address the tax avoidance of multinational enterprises have provided governments with the tools necessary to broaden their tax bases and strengthen their tax systems against abuse. (OECD, 2019[56]) (OECD, 2019[57]). Additional revenues derived from these measures could be channelled to develop and support a stronger agenda for inclusive growth. Besides, measures that strengthen work incentives, such as lowering the effective tax rates for second earners, enhancing tax credits and developing in-work benefits, would contribute to improve fairness from the point of view of lower middle-class families.

Fairness concerns also exist with regard to social transfers, as 58% of middle-income households in OECD countries consider that they do not receive a fair share of public benefits given the taxes and social security contributions that they pay (OECD, 2018[3]; OECD, 2019[41]). When limiting the attention only to cash transfers and direct income taxes, it turns out that middle-income households are neither net contributors nor net recipients – they receive almost as much in social benefits as they pay in taxes. That said, working-age people with middle incomes tend to be net payers while elderly middle-income households are net beneficiaries. Redistribution is not only about cash transfers. Countries spend just as much on social transfers in kind such as publicly provided healthcare and education. Policy makers should therefore improve access to high-quality public social services and ensure better social protection coverage.

1.8.2. Tackling cost of living issues

Decent, affordable and accessible housing including home ownership is a key element of the middle-class status. Policies need to address shortages in the supply while helping households through demand-side support. On the supply side, measures are needed to encourage supply of affordable housing. A first set of solutions includes facilitating private construction to raise the responsiveness of the housing supply to market prices (i.e. to increase its price elasticity). Well-designed inclusionary zoning policies can also help tackle residential segregation and undesired urban sprawling. In many countries, public social housing that reach far beyond low-income households also serves as an efficient policy instrument to promote affordability and inclusiveness (OECD, 2016[58]). Many countries also attempt to influence housing costs directly through rent controls, or ceilings.

Demand-side policy measures first and foremost concern housing allowances or benefits. Additional measures can support lower-middle income households to purchase a home, e.g. through grants, financial assistance for loans or tax relief for home buyers. These policies need to be sufficiently targeted, and avoid measures that benefit high-income home owners more than others (e.g. mortgage tax relief). A number of European countries offer first-time buyers, specifically young people, support with the deposit needed to access a mortgage, for instance in the form of state guarantees, interest subsidies or grants and saving schemes. In countries with acute levels of housing-related debt, mortgage relief measures may be needed to help overburdened households get back on track.

Policies that contain the costs of education and health care can also have a strong impact on middle-income households. Publicly subsidised child care provision for those in need and price regulation are the most effective tools to limit the costs of child care for households with small children (Adema, W. et al., 2016[59]). Another option is to reimburse parents through a direct refund via cash benefits, or to offer tax credits or deductions. Policies to support students from middle-income families for tertiary education include tuition fee loans but also means-tested assistance for non-tuition costs such as housing, transport or books which are crucial for many children from lower middle-income families. Comprehensive school-to-work transition programmes are needed to prevent negative long-term consequences of early-life unemployment or inactivity (OECD, 2016[60]). In countries with universal health coverage, an expansion or deepening of healthcare cost coverage could provide substantial relief for low- and middle-income households. In those where health coverage is not universal, extending means-tested insurance coverage would help lower middle-income groups better manage their health costs.

1.8.3. Tackling labour market vulnerability

Policies should aim at tackling labour market vulnerability at the root. For this reason, education and training systems should offer a wider range of learning opportunities at various ages. However, equipping middle-income workers with the skills needed in a changing world of work is challenging and will require a thorough modernisation of these systems. In particular, developing vocational education and training (VET) systems, notably in the most dynamic sectors of the economy, is essential. Making VET attractive in the context of rising skill standards requires reforming training curricula, developing post-secondary pathways and engaging more with employers.

Ensuring access to adult learning is another major challenge. Currently, training participation amongst middle-income adults without tertiary education, and especially among older age groups is notably lower than for the younger and more educated. More and innovative adult learning programmes should be targeted at people employed in mid-skill jobs, or industries and regions particularly at risk from labour market changes.

Results from skills assessment and anticipation exercises should be used to identify individuals who lack the skills required by the labour market and to support them in retraining. To improve coverage, training opportunities should be de-coupled from job and employment status and individuals be given financial support for training investments, e.g. via personal training accounts that allow individuals to accumulate and transfer entitlement to training hours.

Beyond active labour market and skills policies, tackling labour market vulnerability entails many more measures, starting from labour market policies that combine flexibility and security, social protection as well as effective social dialogue to increase the leverage of labour. In light of the changing world of work, temporary or unstable jobs, often characterised by lower wages and weaker job security, will likely become more widespread as traditional standard middle-class jobs disappear. Therefore, efforts to extend social insurance as well as collective bargaining coverage to non-standard workers will also be of particular relevance in the future to secure middle-income households. One approach for improving social protection coverage is to link entitlements to individuals rather than to jobs (OECD, 2017[61]; OECD, 2019[62]). This would make entitlements to, for instance training or employment protection, “portable” as workers move from one job to another or combine different forms of employment.

1.9. Conclusion

A strong and prosperous middle class is important for the economy and society as a whole, and notably to sustain consumption and investment in education, health and housing. The taxes they pay are essential to properly fund social protection. Societies with a strong middle class also experience higher levels of social trust, lower crime incidence, and higher life satisfaction. The evidence shows that the middle class champions political stability and good governance.

Today, however, middle-class households became increasingly anxious about their economic situation. Firstly, many middle-class households consider our socio-economic systems as unfair, given that middle incomes have not benefitted from economic growth as much as upper incomes, or the top 10%. Secondly, the middle-class lifestyle is increasingly expensive, notably when it comes to housing, a good education and healthcare. Thirdly, the labour market prospects for many in the middle class are uncertain, with one in six middle-income workers being in jobs that are at high risk of automation.

A comprehensive action plan is needed to address the risks identified in this report and support the middle class, notably the more vulnerable. This plan should follow a holistic approach that involves different policy areas, paying attention to the prevention of risks and the promotion of opportunities over the lifecycle. Mitigating the impact of adverse situations when they arise is essential, but improving prospects in the labour market for middle-class workers and their children may be even more important.

Governments do have a large set of policy tools available to address these challenges of unfairness, unaffordable expenses and employment vulnerability. Targeted reforms are needed in the fields of labour market, education and training systems, as well as fiscal and social policies. A successful strategy will require a whole-of-government approach should involve all stakeholders including civil society and the social partners.


[59] Adema, W. et al. (2016), Who uses childcare? Background brief on inequalities in the use of formal early childhood education and care (ECEC) among very young children, OECD Publishing, Paris,

[44] Alesina, A. and R. Perotti (1996), “Income distribution, political instability, and investment”, European Economic Review, Vol. 40/996, pp. 1203-1228, (accessed on 10 November 2017).

[55] Alstadsaeter, A., N. Johannesen and G. Zucman (2017), “Tax Evasion and Inequality”, No. 23772, NBER Working Paper, Cambridge, MA,

[6] Autor, D. (2015), “Why Are There Still So Many Jobs? The History and Future of Workplace Automation”, Journal of Economic Perspectives, Vol. 29/3, pp. 3-30,

[17] Bassanini, A. and S. Scarpetta (2002), “Does human capital matter for growth in OECD countries? A pooled mean-group approach”, Economics Letters, Vol. 74/3, pp. 399-405,

[42] Bettiza, G. (2010), Tea Time in Washington: The Tea Party in International Perspective, LSE IDEAS Today, no. 6 (2010), LSE IDEAS, (accessed on 10 November 2017).

[28] Bird, R. and F. Newport (2017), “What Determines How Americans Perceive Their Social Class?”, Gallup, Polling Matters,

[52] Birdsall, N. (2016), “Middle-Class Heroes”, Foreign Affairs, (accessed on 20 October 2017).

[37] Blank, R. (2010), Middle class in America, U.S. Department of Commerce, Economics and Statistics Administration,

[15] Brown, D. and W. Hunter (2004), “Democracy and Human Capital Formation”, Comparative Political Studies, Vol. 37/7, pp. 842-864,

[18] Brueckner, M. et al. (2017), The Rise of the Middle Class and Economic Growth in ASEAN, The World Bank,

[72] Bussolo, M. et al. (2018), Toward a New Social Contract: Taking on Distributional Tensions in Europe and Central Asia, The World Bank,

[36] Crossley, T., H. Low and C. O’Dea (2013), “Household Consumption through Recent Recessions*”, Fiscal Studies, Vol. 34/2, pp. 203-229,

[67] Curtis, J. (2013), “Middle Class Identity in the Modern World: How Politics and Economics Matter”, Canadian Review of Sociology/Revue canadienne de sociologie, Vol. 50/2, pp. 203-226,

[35] De Nardi, M., E. French and D. Benson (2012), Consumption and the Great Recession, (accessed on 3 December 2017).

[63] Easterly, W. (2001), “The Middle Class Consensus and Economic Development”, Journal of Economic Growth, Vol. 6/4, pp. 317-335,

[70] European Commission (2017), Designing Europe’s future: Trust in institutions Globalisation Support for the euro, opinions about free trade and solidarity, Special Eurobarometer 461 – Wave EB87.2 – TNS opinion & social, (accessed on 20 November 2017).

[65] Evans, G. and J. Mellon (2016), “Social Class: Identity, awareness and political attitudes: why are we still working class?”, British Social Attitudes, Vol. 33, (accessed on 20 November 2017).

[22] Evans, M. (2004), “Subjective Social Location: Data From 21 Nations”, International Journal of Public Opinion Research, Vol. 16/1, pp. 3-38,

[68] Ferreira, F. et al. (2013), Economic mobility and the rise of the Latin American middle class, World Bank Publications,

[30] Förster et al. (2014), Trends in Top Incomes and their Taxation in OECD Countries, OECD Publishing, Paris,

[69] Frank, R., A. Levine and O. Dijk (2014), “Expenditure Cascades”, Review of Behavioral Economics, Vol. 1/1-2, pp. 55-73,

[25] Goldthorpe, J. (2016), “Social class mobility in modern Britain: changing structure, constant process”, Journal of the British Academy, Vol. 4, pp. 89-111,

[24] Gornick, J. and M. Jäntii (2013), Income Inequality : Economic Disparities and the Middle Class in Affluent Countries, Stanford University Press.

[46] Gould, E. and A. Hijzen (2016), “Growing Apart, Losing Trust? The Impact of Inequality on Social Capital Growing Apart, Losing Trust? The Impact of Inequality on Social Capital”, (accessed on 12 November 2017).

[50] Graves, F. (2017), Understanding the Shifting Meaning of the Middle Class, (accessed on 21 August 2017).

[64] Hacker, J. and P. Pierson (2010), Winner-take-all politics : how Washington made the rich richer-and turned its back on the middle class.

[71] ILO (2016), Global Wage Report 2016 / 17 Wage inequality in the workplace, International Labour Office, Geneva,

[48] Kelly, M. (2000), “Inequality and Crime”, Review of Economics and Statistics, Vol. 82/4, pp. 530-539,

[33] Lakner, C. and B. Milanovic (2016), “Global Income Distribution: From the Fall of the Berlin Wall to the Great Recession”, The World Bank Economic Review, Vol. 30/2, pp. 203-232,

[49] Lynch, J. and G. Kaplan (1997), “Understanding How Inequality in the Distribution of Income Affects Health”, Journal of Health Psychology, Vol. 2/3, pp. 297-314,

[47] Madland, D. (2015), Hollowed out : why the economy doesn’t work without a strong middle class, University of California Press, (accessed on 12 November 2017).

[34] Milanović, B. (2019), Inégalités mondiales : le destin des classes moyennes, les ultra-riches et l’égalité des chances, (accessed on 6 March 2019).

[32] Milanović, B. (2016), Global inequality : a new approach for the age of globalization, Belknap Harvard University Press, (accessed on 21 November 2017).

[7] Nedelkoska, L. and G. Quintini (2018), “Automation, skills use and training”, OECD Social, Employment and Migration Working Papers, OECD Publishing, Paris,

[56] OECD (2019), Base erosion and profit shifting - OECD,

[1] OECD (2019), Inclusive Growth - Economic growth that is distributed fairly across society, (accessed on 6 March 2019).

[57] OECD (2019), OECD delivers implementation guidance for collection of value-added taxes (VAT/GST) on cross-border sales - OECD, (accessed on 7 March 2019).

[62] OECD (2019), OECD Employment Outlook 2019: The Future of Work, OECD Publishing, Paris,

[53] OECD (2019), OECD Tax Database, (accessed on 6 March 2019).

[41] OECD (2019), The OECD Risks That Matter Survey 2018,

[3] OECD (2018), A Broken Social Elevator? How to Promote Social Mobility, OECD Publishing, Paris,

[4] OECD (2018), A Broken Social Elevator? How to Promote Social Mobility, OECD Publishing, Paris,

[10] OECD (2018), Equity in Education: Breaking Down Barriers to Social Mobility, PISA, OECD Publishing, Paris,

[40] OECD (2018), Risks that Matter Early Results from the 2018 OECD Cross-National Survey on Social and Economic Risks, OECD Publishing, Paris, (accessed on 3 March 2019).

[2] OECD (2018), The framework for policy action on inclusive growth, OECD Publishing, (accessed on 7 February 2019).

[11] OECD (2017), Educational Opportunity for All: Overcoming Inequality throughout the Life Course, Educational Research and Innovation, OECD Publishing, Paris,

[5] OECD (2017), “How technology and globalisation are transforming the labour market”, in OECD Employment Outlook 2017, OECD Publishing, Paris,

[13] OECD (2017), How’s life? 2017 : measuring well-being., OECD Publishing, Paris, (accessed on 26 February 2019).

[45] OECD (2017), Meeting of the OECD Council at Ministerial Level: Key Issues Paper, OECD, Paris, (accessed on 11 August 2017).

[61] OECD (2017), OECD Employment Outlook 2017, OECD Publishing, Paris,

[14] OECD (2017), Preventing Ageing Unequally, OECD Publishing, Paris,

[12] OECD (2017), The Pursuit of Gender Equality: An Uphill Battle, OECD Publishing, Paris,

[58] OECD (2016), OECD Affordable Housing Database. PH4.1 Public spending on support to social rental housing, OECD, Paris.

[60] OECD (2016), Society at a Glance 2016: OECD Social Indicators, OECD Publishing, Paris,

[9] OECD (2015), In It Together: Why Less Inequality Benefits All, OECD Publishing, Paris,

[38] OECD (2015), In It Together: Why Less Inequality Benefits All, OECD Publishing, Paris,

[54] OECD (2014), Focus on Top Incomes and Taxation in OECD Countries: Was the crisis a game changer?, OECD Publishing, Paris,

[8] OECD (2011), Divided We Stand: Why Inequality Keeps Rising - OECD, OECD Publishing, Paris,

[21] OECD (2010), “The Emerging Middle Class in Developing Countries”, OECD, (accessed on 13 December 2018).

[19] OECD (2003), The Sources of Economic Growth in OECD Countries, OECD Publishing, Paris, (accessed on 1 March 2019).

[39] Pew Research Center (2018), A Decade After the Financial Crisis, Economic Confidence Rebounds in Many Countries, (accessed on 8 January 2019).

[16] Pressman, S. (2007), “The Decline of the Middle Class: An International Perspective”, Journal of Economic Issues, Vol. 41, pp. 181-200,

[66] Reeves, R. (2017), Dream hoarders : how the American upper middle class is leaving everyone else in the dust, why that is a problem, and what to do about it.

[23] Reeves, R., K. Guyot and E. Krause (2018), Defining the middle class: Cash, credentials, or culture, Brookings, https:// (accessed on 3 August 2018).

[31] Saez, E. (2018), Striking it Richer: The Evolution of Top Incomes in the United States (Updated with 2017 preliminary estimates), UC Berkeley, Berkeley,

[27] Savage, M. (2016), Poorest Made Poorer? Decomposing income losses at the bottom of the income distribution during the Great Recession, Economic and Social Research Institute (ESRI),

[26] Savage, M. et al. (2013), “A New Model of Social Class? Findings from the BBC’s Great British Class Survey Experiment”, Sociology, Vol. 47/2, pp. 219-250,

[43] Stiglitz, J. (2012), The Price of Inequality: How Today’s Divided Society Endangers Our Future, W. W. Norton & Company, (accessed on 12 November 2017).

[29] Thewissen, S. et al. (2018), “Rising Income Inequality and Living Standards in OECD Countries: How Does the Middle Fare?”, Journal of Income Distribution, Vol. 27/2,

[51] Thorson, G. (2014), The Rise of Inequality, the Decline of the Middle Class, and Educational Outcomes, Segovia, Spain, (accessed on 12 November 2017).

[20] van Stel, et al. (2005), “The Effect of Entrepreneurial Activity on National Economic Growth”, Small Business Economics, Vol. 24, pp. 311–321,


← 1. See Box 1.2 for how this report defines and measures the middle class.

← 2. Despite a decline in the number of workers in occupations associated with the working class, the proportion of British people feeling that they are middle class or working class has not changed much between 1983 and 2015 (Evans and Mellon, 2016[65]).

← 3. An opposite phenomenon occurs only in Portugal, the United Kingdom, and a handful of other countries including Hungary and Slovenia. While not observed in other Eastern European countries, results for the latter may be related to a “lingering effect” of former state socialist legacy that makes people more likely than others to identify as belonging to a low social class (Curtis, 2013[67]).

← 4. Evidence from survey data and the OECD “Compare your income web tool” ( confirm this pattern (Balestra and Cohen, forthcoming[108]). Part of this phenomenon has also been described as the “Me? I’m not rich!” issue, whereby the point where someone becomes “rich” tends to be at a higher income level than one’s own (Reeves, 2017[66]).

← 5. The rise of the middle of the world income distribution was driven mainly by very strong economic growth in large countries such as China, India and Indonesia. In Latin America, the rise of the global middle class has been associated with an increase in years of schooling, labour market formalisation, female labour participation and family and demographic dynamics (Ferreira et al., 2013[68]).

← 6. With almost two thirds of total consumption, the middle-income households are by far the income group with the highest consumption share, ranging from 50% or below in the United States, Southern European and Latin American countries to 70% or above in Nordic countries and Slovakia.

← 7. This also implies that individual skills-related characteristics such as the level of education, age or tenure cannot sufficiently explain trends in wages and wage distribution (ILO, 2016[71]).

← 8. A vast majority of EU citizens believe that inequality levels are too high: 84% thought income differences were too great, and 81% agreed that national governments should take measures to reduce these differences (2018 Eurobarometer survey on perceptions of fairness in Europe). These views are confirmed by the 2018 OECD Risks That Matter survey which indicates that 70% of respondents believe that government should be doing more to ensure their economic and social security. Moreover, users of the OECD Compare your Income web-tool show a preference for lower inequalities: 70% of them would like to see a reduction in the share of income held by the top 10%. The call for a more equal income distribution is stronger in countries that have been severely hit by the crisis, such as Greece and Spain. See also (Bussolo et al., 2018[72]).

← 9. An increase in expenditures by an income group may prompt a rise in the expectations of living standards (and subsequently the spending patterns) of immediately lower income groups. The drifting away of income among top earners in recent decades may have stimulated “a cascade of additional expenditure by those with lower earnings” (Frank, Levine and Dijk, 2014[69]).

End of the section – Back to iLibrary publication page