Under Pressure: The Squeezed Middle Class

Chapter 5. Policies for a prosperous middle class

Abstract

This chapter presents policies that could address the main challenges with which middle-income households contend. Because middle-income households are so diverse, the proposed policy solutions often focus on certain subgroups, such as families who struggle to make ends meet, parents who are concerned about how to pay for their children’s education, and middle-aged workers exposed to globalisation and technological change. Section 5.2 looks at policies for boosting middle-class disposable incomes and supporting wealth accumulation. Section 5.3 discusses policies that could help middle-income households cope with the rising costs of housing, education and health and long-term care. Section 5.4 proposes ways to help equip middle-class workers with the skills they need to succeed in a changing workplace, with special emphasis on vocational education and training and adult learning.

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.

5.1. Introduction

The previous chapters in this report sought to give a comprehensive account of the growing pressure on households in the middle-class across much of the OECD. The middle-income group has lost some of its economic strength as its relative size has declined from one generation to the next (Chapter 2). The archetypal middle-class family – the middle-skilled, single-breadwinner couple with children – often no longer makes it into the middle-income group as levels of educational attainment and female labour force participation have risen in both the group itself and society at large. Many traditional middle-class jobs – notably in manufacturing – are disappearing, to be replaced by lower-quality service sector jobs or highly skilled positions. Projected transformations in the labour market – particularly the rise of automation – will affect many middle-class occupations, and such transformations are already causing growing anxiety (Chapter 3). Meanwhile, many middle-income households struggle to cope with the rising costs of housing, education and health care and have little capacity to save. Some are even running up debt in order to maintain their standard of living (Chapter 4). These trends call for targeted measures to secure middle-class living standards and promote future prosperity, as a healthy middle class is important for economic growth, political stability and social cohesion (Chapter 1).

This chapter presents policies that address the main challenges facing middle-class households. As the middle class is diverse, not all middle-class household face the same challenges or have the same interests. For example, lower and upper middle-income households have seen their earnings and disposable incomes develop in very different ways, homeowners and tenants are unlikely to have the same idea of what successful housing policy is, and a senior manufacturing worker and a young entrepreneur may not necessarily share the same views when it comes to the opportunities and risks of digitalisation. This chapter focuses particularly on lower middle-income households – those who are the most economically insecure. It proposes policies to ensure that younger generations still have the prospect of becoming part of the middle class, and it looks at ways of supporting increasingly vulnerable lower middle-class households to prevent them from slipping out of the middle class.

However, a policy dilemma is evident throughout this chapter when considering ways to support the middle class. Because they represent such a large share of the population (around 60% OECD-wide, see Figure 2.1), the middle class also account for almost two-thirds of direct tax revenue (see Figure 2.14). Therefore, any policy that seeks to reduce the middle-income group’s tax burden or to extend benefits to them may require a trade-off. More precisely, substantially reducing the tax burden on middle-income households would likely undermine the funding of some public services that they heavily rely on, or it would require more precise targeting of public expenditures to lower-income groups to reduce overall spending. Similarly, extending targeted support to lower middle-income households would require an increase in the tax burden borne by other middle-class households to generate the additional public revenue required. If greater support for the lower middle should not come at the expense of the most vulnerable in society, it will require greater contributions from the better-off.

The structure of this chapter mirrors that of the report’s analytical chapters, 2 to 4. Section 5.2 looks at ways of boosting middle-class disposable incomes and supporting wealth accumulation. Section 5.3 discusses policies that help middle-income households cope with rising costs, particularly for housing, education and health and long-term care. Section 5.4 proposes policies that help equip middle-class workers with the skills needed to succeed in a changing world of work, focusing on vocational education and training and adult learning. In all sections, the proposed solutions cut across policy domains, covering labour market and employment, education and skills, taxes and benefits and social policies.

5.2. Boosting middle-class incomes and wealth

Weak income growth over the past decade has been a key factor in the growing pressure on middle-class living standards. Between 2007 and 2016, median household disposable incomes rose, on average, by a meagre 0.3% per year in 19 OECD countries – less than one-fifth of the rate in the previous decade (see Figure 2.3), and much lower than the 0.8% increase for the top-10% income households. Such income stagnation cannot necessarily be attributed to losses incurred during the Great Recession, which were comparable to those experienced by top-income households and much milder than among those at the bottom. It is rather that income gains during the post‑2010 recovery have been lower for households in the middle than for those at the top of the distribution.

Slow wage growth is the prime reason for stagnant middle-class incomes. The last two decades have seen a gradual decoupling of wages from productivity, with real median wages growing at a much slower pace than labour productivity. This trend has occurred together with growing wage inequality since the early 2000s, with a notable widening in the gap between the lower half and top 10% of the distribution (OECD, 2018[1]; 2018[2]). The global financial crisis significantly slowed wage growth, which has remained sluggish during the recovery chiefly because of weak productivity growth and low inflation expectations (OECD, 2018[3]). Since earnings from work are by far the largest income source of middle-income households – at 97% of market income – low wage growth has translated directly into lower income growth.

The disappointing growth in earnings and incomes is of greater concern given that many feel that middle-income households pay too much of their gross income in taxes, such is the sentiment in around half of the population in 24 OECD countries.1 Analysis in Chapter 2 calls for a somewhat more nuanced view, however. Working-age middle-income households indeed contribute more in income taxes than they get out of the redistributive system in benefits. However, the imbalance is largely offset by the high net receipt of benefits among elderly middle-income households, primarily in the form of pensions.2 In other words, the overall net tax burden of middle-income households is close to zero, but there are substantial intergenerational transfers within the middle-income group, from working-age to elderly households. Notably, these calculations consider only cash benefits and direct taxes, while not accounting for in-kind benefits received by middle-income households in the form of education, health care and other public services.

Recent labour market trends have also produced a growing sense of vulnerability among middle-income households. The manufacturing sector, which long provided stable, well‑paid employment, has been shrinking in many OECD countries, a trend that intensified during the global economic crisis. And job growth since then has been primarily in lower-quality service-sector occupations (OECD, 2016[4]). In the years leading up to the crisis and during the recovery, a number of countries carried out reforms to their social protection systems, tightening eligibility criteria, shortening the maximum duration of unemployment benefits, and stepping up the means-testing of minimum-income benefits. These measures have heightened the sense of insecurity among middle-income workers: those who may have contributed to the social security system for decades suddenly face a real risk of quickly losing a significant proportion of their income and assets if they lose their job. Indeed, the risk of downward mobility among lower middle-income households has increased over the past two decades (OECD, 2018[5]) and this trend is likely to persist. Projections suggest rapid technological change will reduce job quality and earnings stability for middle-income workers, albeit to a lesser extent than among those in low-skilled, low-wage employment (see Chapter 3).

Another consequence of slow income growth is that middle-class households – in particular those with modest incomes – find it very difficult to accumulate wealth. Many therefore do not have the benefits of wealth as a means of insurance against income shocks in times of job insecurity and earnings volatility, but also as a source of future capital income. A majority of (lower) middle-income households find themselves in the lower half of the wealth distribution, which in most OECD countries accounts for only a small share of total net household wealth (Balestra and Tonkin, 2018[6]). Furthermore, middle-class wealth is largely made up of housing assets, which are insufficiently liquid to help smooth consumption effectively in the event of income shocks.3

This section discusses the best policies for securing and boosting the resources of middle-income households. Such policies include ensuring fair wages, increasing female labour force participation, reducing the net tax burden on middle-income households, effectively protecting vulnerable middle-income households from negative income shocks, and enabling and encouraging the accumulation of wealth.

5.2.1. Ensuring fair wages for low and medium earners

The decoupling of labour productivity from wages, and the resulting slow wage growth in the bottom and middle the distribution, demonstrates that productivity growth alone is not enough to secure middle-income households better living standards. In addition to raising the earnings potential of middle-income workers through education and skills policies (see Section 5.4), governments need to design labour market institutions that ensure that productivity gains are shared widely and translate into higher wages and better working conditions (OECD, 2018[2]).

Collective bargaining and social dialogue can play a central role in ensuring good wages and working conditions. Indeed, in the public debate, the weakening of trade unions over the last decades has been presented as a contributory factor behind the sluggish wage growth of middle-income households. Strengthening collective bargaining has therefore been suggested as one way to support middle-class incomes (Atkinson, 2015[7]).

However, empirical evidence of the link between the shrinking collective-bargaining coverage of middle-income workers and low wage growth is surprisingly scarce.4 What is clear is that union membership OECD-wide has almost halved since the mid‑1980s. In most countries, collective bargaining agreements now cover less than half of the workforce and, in some, much less than half (OECD, 2017[8]; Marcadent, 2018[9]). In several OECD countries, the overall drop in union membership broadly coincided with the decline in the population share of middle-income households. Moreover, unionisation rates were traditionally highest in sectors, such as resource extraction and manufacturing, that long provided “typical” middle-class jobs and are now suffering from structural decline. For the United States, a recent study has suggested that higher levels of union membership had a positive effect on the incomes of middle-income households, and that the long-run decline in membership contributed to rising income inequality (Farber et al., 2018[10]). Generally speaking, in countries with wider collective bargaining coverage there is less wage dispersion and workers covered by collective agreements benefit from wage premiums (OECD, 2018[3]).5

Since collective bargaining systems differ widely from country to country in their institutional structure and coverage (OECD, 2017[8]), policies to strengthen them depend on national contexts.6 As the world of work changes, efforts to extend collective bargaining coverage to workers in non-standard jobs are of particular relevance for middle-income households. Non-standard jobs are widely characterised by lower wages and job security, and are likely to become more widespread as standard middle-class occupations disappear (OECD, 2018[11]). Unions can and do reach out to workers in such jobs, and social dialogue has a role to play in helping regulate new forms of work (OECD, forthcoming[12]). One example is the Danish trade union 3F, which recently signed a collective bargaining agreement with a company that operates an online platform for domestic work in private homes. The agreement introduces a new category of domestic workers in the platform company, giving them employee status. At the same time, it maintains existing freelance arrangements (De Stefano, 2018[13]).7 Antitrust regulations may bar certain types of non-salaried workers from collective bargaining (OECD, forthcoming[12]). In such cases, existing regulations may need to be adjusted to allow for exemptions for some forms of self-employment and certain sectors or occupations. Such exemptions often already exist for self-employed actors and freelance artists. Alternative forms of worker organisation can complement these efforts, such as non-profit worker centres in the United States or the SMart co‑operative that supports the self-employed in Belgium and eight other European countries (OECD, 2018[14]; forthcoming[12]).

In countries or sectors with no collective agreements or low coverage, statutory minimum wages can help ensure fair pay and raise wages higher up the distribution. There is a relatively broad consensus that minimum wages are not very effective at reducing poverty, because poor households frequently have no-one in work, while many minimum-wage earners live in non-poor households (OECD, 2018[2]). However, this also implies that middle-income households, particularly those on lower middle-incomes, directly benefit from minimum-wage increases.8 In some countries, moreover, the minimum wage has been shown to have a so-called “ripple” or “knock-on” effect, with companies raising wages further up in the distribution to maintain wage differences between lower- and higher-paid workers (OECD, 2015[15]; 2015[16]).9

Levels of gross minimum wage vary substantially from country to country in the OECD.10 Scope for increases depend on a range of country-specific factors, like the initial relative level of the minimum wage, the behavioural response of employers, the degree of competition in product and labour markets, and how the minimum wage interacts with other policies, in particular taxes and benefits (OECD, 2018[2]). In Australia, industry- or occupation-wide regulations, so-called Modern Awards, set industry-specific wage floors that vary by skill level. And while some 36% of employees are covered by collective agreements, another 23% are covered by Awards only. Awards are set by the Fair Work Commission, whose members are chosen by the government and selected among employer bodies, unions, lawyers and public officials (OECD, 2018[3]).

In some countries, businesses pay so-called “living wages”. They are higher than the minimum wage and calculated to cover the cost of living either for a full-time worker or relative to the poverty line. Companies may pay a living wage as part of a voluntary commitment (in Canada, Ireland, New Zealand and the United Kingdom11, for example) or in order to qualify for public contracts, as in some cities and counties in the United States (Eurofound, 2018[17]).

5.2.2. Increasing female labour force participation and earnings

As a result of rising female employment – and the associated increase in disposable incomes across the distribution – many couples who perpetuate the once traditional single-breadwinner model fall short of the middle of the income distribution (Figure 3.7). One-and-a-half earner couples, by contrast, make up a growing share of (lower) middle-income households, with the female partner often combining paid part-time work with unpaid household tasks and/or care. As for couples with two full earners, they account for a disproportionate share of upper-middle-income and high-income households. Clearly, there is potential for boosting middle-class incomes by further increasing women’s labour force participation, hours worked and earnings.

Measures to lower effective tax rates for second earners deserve particular attention as second earners are highly responsive to work incentives (OECD, 2011[18]; 2012[19]).12 In 24 of the 34 OECD countries, second earners pay a higher average tax rate than single earners at the same earnings level, and the number of countries rises to 32 if there are children in the household (Thomas and O’Reilly, 2016[20]). The gap in average tax rates between primary and secondary earners is often particularly wide in countries that operate family-based systems of labour income taxation, such as France, Germany, Luxembourg and Switzerland.13 In individual-based systems, effective tax rates for second earners who start work or increase hours worked (so-called “marginal” tax rates) may be high if tax allowances and credits are withdrawn based on family income as in Italy, there are dependent-spouse allowances for primary earners, as in Slovenia and the Slovak Republic, or if basic credits can be transferred between a non-working and working spouse, as in Canada and the Netherlands). In recent years, a number of countries have attempted to narrow the gap between before- and after-tax pay for second earners – the so-called “tax wedge”. Examples are Australia, which abolished the Dependent Spouse Tax Offset, and the Netherlands, which reformed its tax credit system (OECD, 2016[21]).

Besides tax policy, policies that help middle-income families strike a better work-life balance, including by sharing responsibility for household tasks between men and women more fairly, have an important role to play.14

Well-designed paid-leave provisions can raise female labour force participation, as well as exerting positive effects on maternal health and well-being and even child health and development outcomes (Adema, Clarke and Frey, 2015[22]). They reward women who work and build up leave entitlements prior to giving birth, giving them job security and easing them back into work post-birth. And, if fathers take up a significant share of their leave entitlements, paid leave also helps parents share caring responsibilities more equally (OECD, 2017[23]). As well as paid-leave duration, payment rates are an important determinant of the attractiveness of paid leave to middle-income families and of their income situation after childbirth.15 Household net incomes typically fall by an OECD-wide average of 19% for a couple on average earnings when parents take maternity and paternity leave (OECD, 2017[24]). However, replacement rates for medium- and high-earners tend to be much lower when benefits are paid at a flat rate, as in many English-speaking countries.16 Moreover, payments often decline much more substantially over the following months as parents take parental leave.17 In Belgium, for example, the net equivalised household income of a family with two middle-earners drops from 88% in the first month after childbirth to only 60% in the sixth month. Raising replacement rates in countries where they are low – either generally or specifically for medium and high earners – would increase women’s incentive to work before having children and boost family incomes, both before and after childbirth.

To encourage men to spend more time at home caring for their children, a growing number of countries have introduced “fathers-only” paid leave, e.g. paternity leave and individual parental-leave entitlements available on a “use-it-or-lose-it” basis only. Most common are mother and father quotas, where specific portions of an overall parental leave period are reserved for each parent, as in the Nordic countries. Other options include “bonus periods” – where a couple may qualify for extra weeks of paid leave if both parents use a certain amount of shareable leave, as in Germany – or the provision of paid parental leave as an individual entitlement for each parent (OECD, 2017[23]; 2017[25]). Japan and Korea grant both mothers and fathers around one year of non-transferable paid parental leave each, though take-up among fathers tends to be very low. In Korea, only 8.5% of people who took paid parental leave in 2016 were men (OECD, 2019[26]). Once again, higher payment rates could help improve matters and encourage couples to share leave more evenly. Currently it may make economic sense for a couple to agree that the mother take more paid parental leave, as the father is usually the higher earner. If leave benefits were to replace the bulk of earnings, then couples may well be more willing to share parental leave.

If mothers are to successfully resume work after leave, access to affordable good-quality childcare is important. Indeed, the expansion of publicly provided or subsidised childcare has been one driver of rising female labour force participation (Thévenon, 2013[27]). Comparatively large shares of children from middle-income families are, in fact, in early childhood education and care (ECEC). There is scope, though, for improved attendance among the under‑3s: on average about 37% from the middle tertile of the income distribution were in ECEC OECD-wide in 2016, compared to 27% in the lower tertile and 44% in the upper tertile (OECD, 2018[28]). Two countries that have made significant progress in this respect are Korea and Germany, both of which have heavily increased public expenditure on ECEC. Korea increased subsidies for childcare, while Germany introduced a legal childcare entitlement for all children older than one.18

The quality of childcare is of as much concern as its cost (see Section 5.3) for middle-income families. If formal childcare is to be attractive, it must be as good as alternative informal solutions, such as family-provided education and care. Trained and qualified teachers who enjoy proper working conditions and low child-to-staff ratios are essential requirements in high-standard childcare (OECD, 2011[29]; 2018[30]). While good-quality ECEC has repeatedly been shown to deliver beneficial cognitive and non-cognitive outcomes for low-income children, the fast growth in childcare provision more broadly yields mixed results. In Quebec, for example, the overly rapid expansion of compulsory early childhood education increased female labour force participation, but seems to have had lasting negative effects on non-cognitive child development outcomes (Baker, Gruber and Milligan, 2008[31]; 2015[32]).

5.2.3. Reducing the tax burden on middle-income households

Reducing taxes or expanding benefits immediately boosts middle-class disposable incomes and can help alleviate the sentiment among middle-income households that they pay too much tax, especially relative to the benefits they receive.

After years of rising labour taxes in the wake of the economic crisis, a number of OECD countries have lowered the personal income tax (PIT) rates for low- and middle-income earners. Examples include the elimination of the second PIT bracket in Belgium, the introduction of two new tax brackets in Portugal, a reduction in the lower rates of the Universal Social Charge in Ireland, and tax rate cuts for middle-income earners in Canada, Luxembourg and Slovenia. Several countries, including Belgium, Germany, the Netherlands, Norway, Slovenia and the United Kingdom, increased basic allowances and general tax credits (OECD, 2017[33]; 2018[34]).19 Both reductions in PIT rates in the lower parts of the tax schedule and more generous basic PIT provisions primarily benefit (lower) middle-income households, as those on very low incomes often do not earn enough to pay PIT.

Reducing social security contributions can also be an effective means of raising the take-home income of middle-income households. Indeed, in the vast majority of OECD countries, social security contributions, rather than income taxes, account for most of the direct tax burden for average-wage earners. As a result, social security contributions have little redistributive effect, and can even be regressive, in a number of OECD countries (Causa and Hermansen, 2017[35]; Immervoll and Richardson, 2013[36]). Partially shifting the funding of social protection systems away from social security contributions towards PIT may be an option notably for benefits that are only weakly linked to labour market behaviour – such as health insurance, certain types of pensions, and family allowances (OECD, 2018[2]).

A recurring cause of frustration for middle-class taxpayers in some countries (such as Austria and Germany) is the gradual inflation-induced increase in tax rates because tax rules are defined in nominal terms. It is referred to as “bracket creep” and particularly affects low- and middle-income earners for whom marginal tax rates rise most quickly with additional income if the tax schedule is progressive. Simulations indicate that its effect on income tax revenue may indeed be considerable, even at relatively low inflation rates unless if tax bands are frequently inflation-adjusted (Immervoll, 2005[37]; OECD, 2008[38]; Parlament der Republik Österreich, 2018[39]). While the issue can be easily addressed by, for example, indexing tax rates and allowances to inflation (as in Switzerland) or nominal wage growth (as in Sweden), governments may appreciate the budgetary flexibility they gain by being able to correct for inflation at their own discretion (Bach, 2012[40]).

However, any measures to cut middle-income households’ tax burden put pressure on the sustainability of public finances. Significant tax cuts for middle-income groups would almost certainly lead to substantial shortfalls in public revenue, as middle-income households generate nearly two-thirds of overall direct tax revenue (see Chapter 2). The funding of social protection systems and public services, of which middle-class households are major beneficiaries, would undoubtedly be affected.

Larger reductions in the middle-class tax burden may therefore call for changes in the tax structure. In particular, tax reductions for the lower middle-income group would likely require better-off households – possibly including upper middle-income households – to make a greater contribution to financing public expenditures:

• A more effective taxation of personal capital income could allow the tax burden on labour to be lowered, providing tax relief for lower middle-income households and encouraging greater employment. Capital ownership is more highly concentrated than earnings, and in a number of OECD countries, the wealthiest often pay lower effective tax rates than the middle class. This is because a greater share of their income comes from returns on their financial or non-financial assets, and capital gains made on these assets are taxed at considerably lower rates than labour income.20 Indeed, lower taxation of capital incomes has been one driver of the decline in the efficacy of income redistribution policies across the OECD since the mid-1990s (Causa, Vindics and Akgun, 2018[41]). More recently, Iceland raised its capital income tax rate from 20% to 22% in 2018 to narrow the gap between the taxation of labour and capital income, while in the Netherlands, the government has announced an increase in the tax rate for income from substantial shareholdings from 25% to 28.5% as from 2021 onwards.

• Well-designed inheritance and gift taxes could also be part of the solution. Inheritance and gift taxes, paid by heirs and recipients, are still applied relatively widely across the OECD. Many countries, however, have narrowed their base since the mid‑1990s with the introduction of numerous exemptions and deductions, increasing opportunities for evasion through tax planning schemes affordable almost exclusively to higher-income groups. Belgium and France, however, levy substantial revenues from inheritance tax – 0.7% and 0.6% of GDP in 2016, respectively. The amounts collected suggest that other countries, too, could further exploit the distributional impact of inheritance and gift taxes (OECD, 2018[42]). Japan, for example, reformed its taxation of inherited assets and gifts inter vivos to good effect in 2015, reducing the tax-free threshold on the inheritance tax by 40% for certain higher-income groups. Appropriate allowances for households with low to medium wealth could ensure that inheritance taxes do not discourage the middle classes from accumulating wealth.

• Higher marginal tax rates for high-income earners could help finance some of the reductions in the tax burden of lower middle-income households. Proponents of higher rates emphasise that they have little to no negative effect on economic growth, and that there is a historic precedent for much higher top marginal tax rates. Up until the 1980s, the average top statutory tax rate in OECD countries was over 65%, compared to just over 43% in 2016 (OECD, 2018[34]). Israel, Korea, Luxembourg and Norway all increased their top PIT rates in 2017, with Korea applying the largest rise, from 40% to 42%. However, the capacity to generate substantial additional tax revenue from high-income earners is limited if one wishes to avoid also increasing the tax burden on the middle – the implied top marginal tax rates would otherwise have to be very high.

• Greater international coordination on closing tax loopholes and fighting tax fraud by individuals and corporations can help governments broaden their tax bases and hence raise additional funding. Estimates suggest that over 70% of the world’s wealth held in tax havens originates from the richest households in OECD countries (Alstadsæter, Johannesen and Zucman, 2018[43]; 2017[44]). Country-level implementation of the wide-ranging G20/OECD Base Erosion and Profit Shifting (BEPS) Project has already had an impact on reducing corporate tax avoidance, with evidence emerging that some multinationals have already changed their tax arrangements to better align with their business operations.21 The widespread adoption of the OECD/G20 Standard for Automatic Exchange of Financial Account Information in Tax Matters has also seen a significant increase in cooperation between countries on the financial accounts held by individuals in tax jurisdictions where they are not residents.22

Net wealth taxes have also been promoted recently as a way to raise tax revenues. However, recent OECD research has shown that when broad-based personal capital income taxes and well-designed inheritance and gift taxes are in place, there is a limited case for net wealth taxes, be it from an efficiency or equity perspective (OECD, 2018[42]).23

On the benefits side, in-work benefits can help boost disposable incomes lower down the income distribution by topping up modest earnings from work and encouraging employment (OECD, 2018[2]; Immervoll and Pearson, 2009[45]). The best-known employment-conditional income support programme is perhaps the United States’ Earned Income Tax Credit (EITC), introduced in 1975 to support working families on low earnings. Those whose income tax burdens are lower than the credit they would be entitled to may receive the difference in cash – the tax credit is “non-wastable”. The EITC has repeatedly demonstrated positive employment and income effects, notably for single parents. While it is primarily an anti-poverty measure, with its effect being greatest among households on incomes of between 75% and 150% of the federal poverty line, more modest benefits accrue even to households on incomes that are as high as 250% of the poverty line (Hoynes and Patel, 2015[46]). In France, the Prime d’Activité is a direct cash benefit for working households with earnings below certain thresholds. The scheme yields up to EUR 300 per month for a single employee or self-employed worker earning around half of the full-time minimum wage. After extensions in 2018 and early 2019, it now benefits single-person households earning up to 1.5 times the minimum wage (Carcillo, Hijzen and Thewissen, forthcoming[47]).24

The affordability of extending in-work benefits to lower middle-income households varies from country to country. Recent evidence from the United States suggests that the EITC largely funds itself, with 87% of the programme’s costs being recovered through savings in public assistance payments and increased tax revenues (Bastian and Jones, 2018[48]). However, designing and financing in-work benefits tends to be more difficult in countries where earnings at the bottom of the distribution are compressed. They have to phase out benefits more quickly as earnings rise. As a result, it is difficult to target the working households entitled to benefits with precision, the working households entitled to benefits, and there is a heightened risk that recipients may deliberately reduce their earnings to qualify (OECD, 2018[2]).

5.2.4. Providing effective income protection for workers with unstable employment and earnings trajectories

Globalisation and digitalisation are profoundly transforming OECD labour markets. They increase the need for well-designed social protection systems while simultaneously throwing up challenges to the design of such systems. Greater labour market polarisation has heightened the risk of job displacement for mid-earning workers, and increased the need for effective out-of-work assistance to bolster incomes in the event of job loss and support re-employment. Many newly created jobs are of low quality – unstable and poorly paid – so raising concerns over inadequate coverage and funding of existing social security systems. Indeed, most of these systems were originally designed on regular, full-time, open-ended contracts with a single employer (OECD, 2017[8]). Already today, fewer than one-in-three jobseekers OECD-wide is covered by unemployment benefits (OECD, 2018[3]).

One approach to improving social protection coverage in times of increasingly fragmented working careers is to link entitlements to individuals rather than to jobs (OECD, 2017[8]; OECD, 2018[49]). In most OECD countries, employees build up unemployment benefit entitlements and pension rights independently of the employer for whom they work. By contrast, entitlements to severance pay in the event of dismissal typically lie with a specific employer, with a change of employers bringing about a loss of these rights. In some cases, health insurance coverage is also tied to the employment contract, so that employees who lose their job or transition into self-employment may also risk losing their coverage.

Austria offers an interesting policy lesson with its measure to increase the “portability” of entitlements. In 2003, the country replaced its severance pay scheme with personal, company-based pension accounts to which employers contribute a fixed rate of individual earnings. In the event of dismissal employees with at least three years of tenure can chose between receiving their severance payment from the account or saving it to boost their future pension entitlement. Employees who quit voluntarily or lose their job after a tenure of less than three years do not qualify for severance payment, but may carry over their entitlements to their next employer (OECD, 2013[50]; OECD, 2018[49]). The scheme was extended to independent contractors in 2008, and has been shown to increase the job mobility of workers in struggling firms (Kettemann, Kramarz and Zweimüller, 2016[51]). France introduced personal training accounts (compte personnel de formation) in 2015 allowing employees to build up and use training credits that are maintained if they change employer or become unemployed. A personal training account is also in place in Iceland (Brandt, 2015[52]; OECD, 2019[53]; 2019[54]; forthcoming[55]).

There is substantial scope for improving benefit coverage for the self-employed, who have more limited or no access to insurance-based benefits in most OECD countries (OECD, 2015[56]; Spasova et al., 2017[57]). A recent EU study estimated that 55% of the self-employed were unlikely to be entitled to unemployment benefits in 2014, and nearly 40% when it came to sickness benefits (Matsaganis et al., 2016[58]). The share of middle-income households headed by a person in self-employment – about 10% OECD-wide – is currently lower than among families on low or high incomes (Figure 3.11). However, the projected decline in regular employment in the middle of the skills distribution may give rise to greater independent employment among middle-income households.

A positive side effect of wider unemployment benefit coverage is that jobseekers stay in touch with public employment services, and so benefit from active labour market support. A number of countries have introduced voluntary schemes for the self-employed, e.g. unemployment insurance in Austria and Spain and family and sickness benefits in Canada. A weakness of such voluntary schemes, though, is that they tend to attract the most risk-prone people, which may jeopardise their financial sustainability (OECD, 2018[49]; forthcoming[12]).

An alternative approach to preventing gaps in social protection coverage could be to loosen its ties with the employment relationship. Granting benefit entitlements based on residency criteria or needs assessment, rather than on employment history, would be one way to do so. Benefits could be either contribution-financed – such as the Dutch basic pension scheme and the Swiss public earnings-related pension scheme – or entirely tax-financed, as with most social assistance schemes. Indeed, some social benefits – such as health insurance and maternity or parental leave – are already universal in a number of OECD countries. The Australian model may offer useful guidance for social benefit provision. Most of the country’s public benefits are flat-rate entitlements financed from general government revenue, with no explicit social security contributions required as a condition of eligibility. Income support payments are means-tested, though the criteria applied are typically more generous than those for social assistance schemes in many OECD countries (OECD, 2018[11]; OECD, 2018[11]).

Among middle-income households threatened by income loss, one group that is likely grow in size is mid-career workers displaced by structural economic change and forced to switch industry or occupation. Recent OECD case studies (OECD, 2018[11]; 2018[3]) highlight the obstacles to re-employment with which they have to contend – e.g. obsolete skills, no recent job search experience and possibly unrealistic pay aspirations. However, the studies also point to their strengths – notably histories of stable employment, strong labour force attachment and high motivation. Early counselling on the adjustment process and the development of realistic re‑employment strategies can be very helpful. An interesting example in this respect are Sweden’s so-called “job security councils”. They are established through collective agreements and financed by employers, providing support both to employers and displaced employees in the event of restructuring (OECD, 2015[59]).

Some jobseekers are hard to re‑employ. The focus should be on ensuring that they use their out-of-work periods efficiently in order to develop the skills required in changing labour markets (Section 5.4). Wage insurance schemes provide income protection for displaced workers who take up new jobs at lower wages by covering part of the earnings gap, albeit for a set period. There is only limited evidence to date on the effectiveness of such schemes, however. Evaluations of two small-scale programmes in Canada and the United States find that they indeed reduced the decline in the net incomes of displaced workers. They did not, though, have any significant effect on how fast they found new jobs or on their post-displacement wages (Bloom et al., 2001[60]; Wandner, 2016[61]; OECD, 2018[3]).

5.2.5. Enabling and encouraging middle-class households to accumulate wealth

The policy options described in the previous subsections can help boost the disposable incomes of middle-class households and, thereby, their capacity to save. To save, however, households need investment opportunities and incentives. One challenge that low- and middle-income households face is that most financial instruments available to those with small savings yield lower returns than those for large savers (Atkinson, 2015[7]).

A simple solution could be to introduce national savings bonds with a guaranteed positive real interest rate and a maximum holding per person, so affording households with limited savings capacity the opportunity of low-risk wealth accumulation. The United Kingdom, for example, regularly issues government-backed Index-linked Savings Certificates, most recently at an annual real interest rate of 1%.25 Each issue comes with an investment ceiling of GBP 15 000 (EUR 16 700) per person. Returns are free of income tax, which also makes them attractive to higher-income earners. Ireland has a similar instrument in place with its Savings Bonds and Certificates, though they are not inflation-protected (Atkinson, 2015[7]).

A more far-reaching approach would be to provide all young adults with a capital endowment – or “minimum inheritance”.26 They could access it on turning 18 or 21 and use it for purposes like funding their education and training, as down payment on a mortgage, or as starting capital for starting a company. Proponents of minimum inheritance have suggested financing it through an annual personal wealth tax or a lifetime capital receipts tax and that should be paid back at death.

Child development accounts could serve a similar purpose. Current proposals by researchers in the United States call on governments and charities to make payments into so-called “baby bonds” for all children from families whose net worth is below the national median. The bonds would become accessible to the children when they turn 18 years old (Hamilton and Darity Jr., 2010[62]).27

A small step in a similar direction was the Child Trust Fund in the United Kingdom, a long-term tax-free savings account for children introduced in 2005 but discontinued in 2011.28 It was designed to ensure that all British children had savings when turning 18 and to foster good saving habits. Evaluations suggest that the number of people putting aside monthly savings for their children increased threefold during the six-year period that the programme was active, whilst the amount saved rose by 60% over the same period (United Kingdom Parliament, 2010[63]). Similar policies have been proposed to incentivise smaller-scale wealth accumulation for a specific purpose. For example, children’s savings accounts (CSAs) have been advocated as a way of building up sufficient assets to pay for higher education. In the United States, Elliott and Lewis (2018[64]) suggest that public institutions or philanthropic sources should be encouraged to place deposits in CSAs at birth, while low-wealth families would be given incentives for additional regular saving.29

Reducing variance in the tax treatment of different capital assets could also be a step towards encouraging wealth accumulation of lower middle-income households while improving tax equity. Recent OECD research has found that the large disparities within countries’ tax treatment of different assets play a significant role in determining which assets people save their money in (OECD, 2018[65]), generally favouring more affluent savers. Private pension funds tend to be the most tax-favoured form of saving, with owner-occupied residential property also significantly tax-favoured. In contrast, rental property is often subject to relatively high marginal effective tax rates, as are bank accounts. Curbing the variance in the tax treatment of different assets – in other words, improving “tax neutrality” – by curtailing capital gains tax exemptions for example, could help increase the returns on smaller savings (Aghion et al., 2017[66]). Preferential tax treatment for private pension remains important for the middle class, however, given the need to encourage middle-income households to accumulate wealth for their retirement.

Inheritances are the source of a significant part of household wealth, including for the middle class (Balestra and Tonkin, 2018[6]). Indeed, a considerable share of the lower middle-income group hold relatively significant wealth despite their lower incomes, often because they have inherited their house. Inheritance taxes can generate significant revenue (see above) and promote equality of opportunity by reducing overall wealth inequality (OECD, forthcoming[67]), but must be carefully designed to avoid overburdening the middle class. For example, a substantial tax bill for lower middle-income households could force them into selling their inherited property to pay tax due. Inheritance taxes should be progressive, at increasingly higher rates for those with higher incomes and those who receive a higher-value inheritance (Causa and Woloszko, forthcoming[68]). Tax-free allowances ensure that those receiving small inheritances are spared. For lower middle-income households facing liquidity constraints, taxes on larger inheritances could be repaid in instalments or, for those occupying the inherited property, tax payments could be deferred until the property is sold.

5.3. Helping households cope with the rising costs of housing, education, health and long-term care

Middle-income groups’ spending increased faster than their incomes between 1995 and 2015 (Chapter 4). Alongside a change in consumption patterns, there were rises in the prices of certain goods and services that were once traditional staples of middle-income households. As a result, many of them struggle to save, spending almost all that they earn, with many running up significant debts. Around 40% of middle-income households in EU OECD countries state that they have troubles coping with any unexpected expenses or sudden falls in income. Indeed, over-indebtedness has become commonplace in a number of countries (see Figures 4.10 and 4.11).

Rising housing costs have been central to middle-income households’ struggles (see Figure 4.5).30 Housing is their largest expenditure item, accounting for approximately one-third of disposable income OECD-wide. In the last two decades, house prices have been growing three times faster than household median income. This trend reflects widespread shortages in the supply of housing, particularly in booming urban areas, where construction often fails to keep pace with rising demand (Barbosa et al., 2017[69]). Moreover, housing is arguably more than just a standard consumption good. In many countries, being middle class has traditionally been associated with owning a home. Soaring house prices and high rents, which may put homeownership out of reach for both lower middle-income households as well as younger generations, throw into doubt the middle-class’s self-perception and the promise of upward mobility. As a result, it is becoming an increasingly unrealistic prospect for many young people to accumulate the necessary wealth to one day acquire property. In a wide range of countries, they are far less likely to purchase property than their parents were.31

The cost of education has also risen considerably across OECD countries in the last 30 years. Middle-income households in EU OECD countries spent a 32% higher share of their disposable income on education in 2015 than in 1988 (Eurostat, 2018[70]), with the greatest increases seen in spending on pre-primary and tertiary education. Such increases in costs are obviously a strain on the budgets of middle-income families. However, they can also jeopardise upward mobility by reducing participation rates in early childhood education32 – simultaneously creating additional hurdles for second earners with caring responsibilities seeking to enter the labour market, – and discouraging young people from pursuing higher education. Social class remains one of the most significant predictors of a child’s educational outcomes (OECD, 2018[5]).

In some countries, health care costs have also become a major concern. On average, lower middle-income households in EU OECD countries spent a 28% greater share of their budgets on health care in 2015 than a decade earlier (Eurostat, 2018[70]). This figure masks significant variations between countries, however. Spending was particularly high in Greece and Latvia at respectively 6.8% and 7.2% of middle-income budgets compared to 1.4% and 1.9% in the Netherlands and Sweden. Social-care costs are forecast to rise further as life expectancy continues to increase – 13% of over‑65s are already receiving long-term care (LTC) in OECD countries (OECD, 2017[71]). And those figures do not include the social care services that relatives commonly provide at home, and which often place significant financial, mental and physical pressure on families, and particularly on women.

This section looks at ways of supporting middle-income households who have to contend with rising costs. Policy solutions seek to help families live in affordable housing, access quality education for their children and reduce their expenditure on caring for their own health and that of their loved ones.

5.3.1. Ensuring affordable housing

Policy changes to make housing more affordable can reduce the strain on middle-income budgets considerably. To achieve sustainable improvements in housing affordability, policy makers need to address shortages in the supply of housing while providing demand-side support to bring accommodation within the financial means of households. While many housing policies in OECD countries have traditionally favoured homeowners, more could be done to promote a robust and affordable rental market.

A first set of solutions focuses on facilitating private construction to make the housing supply more responsive to market prices – increase its price elasticity, in other words. A critical assessment of existing land-use policies could be a first step in the many countries where unnecessarily strict regulations prevent or slow construction, thus driving up prices and rents in densely populated areas (Kok, Monkkonen and Quigley, 2014[72]; Saks, 2008[73]; Quigley and Rosenthal, 2005[74]). Examples of such regulations include minimum lot sizes, off-street parking requirements, single-family rules, lengthy permit application processes, and restrictions on accessory dwelling units.

Relaxing land-use regulations might also to make it easier to build higher-density housing projects that lower- and middle-income groups can afford, lessening residential segregation and urban sprawl (Reeves, 2017[75]). Tokyo used that approach in the early 2000s, allowing the construction of new housing clusters to increase the supply of affordable housing and achieved one of the lowest levels of social segregation for a city of this size by relaxing land use restrictions.33 There are, however, concerns that higher-density housing projects would encourage high-rise luxury developments over affordable housing. One response is inclusionary zoning, which requires property developers to set aside a certain share of the units that they build as affordable accommodation below market rates for low- and middle-income households.34 However, policies to lower house prices may meet opposition from current homeowners – often more affluent middle-income households.

As a complement to the provision of private-sector housing, some OECD countries operate social (subsidised) housing schemes that cater not only to low-income households, but also to those higher up the income distribution. Australia, Austria and Korea, for example, implement large-scale social rental housing programmes, in which they invest between 0.3% and 0.6% of GDP (Salvi del Pero et al., 2016[76]). In France, nearly one in six people lived in social housing in 2016 (Delance, 2018[77]). The most notable example is Austria, where Vienna’s municipal housing subsidiary programme finances 80% of the dwellings constructed in the city, thereby meeting demand from parts of the middle class. The resulting plentiful supply of accommodation, together with rent-control regulations that apply to all subsidised buildings, generally keeps rents down, improves housing quality and reduces income segregation (City of Vienna, 2017[78]; Institute for Urban Strategies, 2018[79]). Across the OECD, 22 out of 29 OECD countries with available data support subsidised housing as a way of promoting affordability and inclusiveness. Such housing is intended mainly – though not exclusively – for low-income households (OECD, 2016[80]).

The length of time it takes for newly built housing – whether built privately or publicly – to translate into affordable homes has led many countries to control or cap rents. Rent controls usually entail determining, first, the highest permissible initial rent for a tenancy contract and, second, the rate of rent increases acceptable during the contract. The Netherlands and Germany have been increasingly active in protecting tenants from rent increases (Davies et al., 2017[81]).35 Altogether, 15 OECD countries practice rent control (OECD, 2016[82]), though there is ongoing debate over its merits.

Critics argue that it can be counterproductive, deterring property owners from investing in housing or offering their homes for rent, constraining supply and pushing up house prices and rents. Indeed, a recent study on San Francisco found that rent control reduced the supply of rented accommodation in the city by 15% and, in the long run, promoted gentrification (Diamond, Mcquade and Qian, 2018[83]).36 Proponents of rent control point to the disparity in rents as a share of household budgets in countries that control and those that do not. For example, the average household spends 25% of its income on rent in Germany, which practices rent control, compared to 40% in England, which does not (Davies et al., 2017[81]).

Most OECD countries also offer demand-side support to households who buy or rent a home. At least 29 OECD countries facilitate homeownership through grants, financial loan assistance and tax relief for buyers.37 While slightly less common than social housing and income-related housing allowances, financial assistance for homebuyers tends to represent a much larger share of public spending on housing (Salvi del Pero et al., 2016[76]), and it is particularly relevant to middle-class households. In a typical OECD country, a significant reduction in the down payment required to access a mortgage increases the probability of owning a home for middle-income households (Andrews and Sánchez, 2011[84]).38

However, such policies come with important downsides. Preferential tax treatment of residential property, particularly mortgage interest tax relief and lower tax rates for owner-occupied housing, can have highly regressive effects, disproportionately benefiting more affluent middle-income and high-income homeowners (Salvi del Pero et al., 2016[76]). Support for homeownership may also encourage the purchase of more and larger properties by those who can already afford to buy, diverting investment from other savings assets towards housing, and reducing the supply of housing available for rent. As a result, opportunities to increase high-density housing are curtailed and further upward pressure is put on house prices, limiting access to homeownership for first-time buyers (Andrews and Sánchez, 2011[84]). Yet, scrapping all distortive tax advantages for residential property would significantly increase the tax burden on middle-class homeowners, who hold significantly higher proportions of their total assets in housing than any other income group (Balestra and Tonkin, 2018[6]). Applying a progressive tax schedule on recurrent taxes on immovable property is one policy recommendation to tax housing assets more effectively. This would need to be based on up-to-date property prices, which is often not the case in many OECD countries. Any tax should also account for liquidity issues in tax payments, such as enabling households to pay in instalments. In Denmark, a property tax reform was approved in 2017 and will become fully effective in 2021. It includes a new system for housing valuation and replaces a nominal freeze of property taxes with proportional taxation, maintaining a progressive element for the most valuable homes (OECD, forthcoming[67]).

A possible response is means-tested homeownership support schemes. Chile, for example, invests nearly 0.5% of GDP in support for low- and middle-income homebuyers through the Solidarity Fund for Housing Choices and the Integrated Housing Subsidy System (OECD, 2016[85]). Alternatively, mortgage interest tax deductions, as they exist in the United States, could be gradually replaced by one-time refundable tax credits for first‑time house buyers, which have much more favourable distributional properties (Gale, 2019[86]). A number of countries offer support to young first-time buyers with the deposit needed to access a mortgage coming in the form of state guarantees (the Housing Loan Guarantee in Estonia, for example), interest subsidies (the ASP saving and loan scheme in Finland) or grants and saving schemes (“A Dwelling for the Young” in Poland).

Such measures can be of valuable help to middle-income households with limited savings that were disproportionately affected by the increased deposit requirements widely introduced to tighten up mortgage regulations in the wake of the 2008 financial crisis (Whitehead and Williams, 2017[87]).39 In countries with an acute level of housing-related debt, mortgage relief measures can help overburdened households get back on track40. They usually entail refinancing mortgages by postponing or subsidising mortgage payments and debts. Examples include Norway’s start-up loans, Hungary’s Debt Management Service and Ireland’s Mortgage to Rent scheme (OECD, 2016[85]).

5.3.2. Helping families pay for their children’s education

Publicly subsidised, price-regulated ECEC is an effective way of limiting costs for households with small children without compromising on quality. A case in point is the provision in some Nordic countries, which does not require parents to pay much from their own pockets. Denmark, for example, conducts means-tests, providing ECEC services for free to low-income families and at reduced rates to those on moderate incomes.41 Iceland and Sweden also stand out as countries where participation rates are high and vary little with household income. In these countries, a combination of heavily subsidised, income-tested fees and sufficient supply means that very young children are likely to participate in formal ECEC regardless of family income (OECD, 2017[23]).42

An alternative option for helping families with high out-of-pocket expenditure is to reimburse them through the tax-benefit system. Such an approach can take the form of a direct refund through cash benefits, as seen in child-care benefits in the Netherlands, which is determined by the price of ECEC and parents’ income. Other countries offer tax credits or deductions to parents who use formal services. In Portugal, for example, 30% of the costs of formal ECEC are tax-deductible up to a joint income limit of 160% of the national minimum wage (Adema, Clarke and Thévenon, 2016[88]).

However, means-tested childcare support needs to be carefully designed. Targeting too narrowly may leave lower middle- and middle-income parents with little support even though they may need it. Loose targeting, by contrast, can be very expensive, particularly where the income distribution is narrow, potentially taking resources away from lower-income families who may be in urgent need. Financial assistance can be phased out gradually as incomes rise to prevent ECEC becoming unaffordable for lower middle-income families and to maintain an incentive for parents to increase earnings from work. Some degree of price regulation may also be desirable to prevent childcare providers from capturing part of the financial support by raising prices (Rodgers, 2018[89]).

There are significant differences in the provision and funding of higher education in OECD countries. Accordingly, policy solutions for reducing the cost of higher education should also vary to meet local needs. Some countries with high tuition fees operate financial‑support systems that offer all students loans with income-contingent repayments, combined with means-tested grants. Such an approach can be an effective way of widening access and promoting equity, while sharing the costs of education between taxpayers and students (OECD, 2016[90]). In an attempt to offset increases in university tuition fees in 2006 and 2012, England and Wales passed reforms that led to all students qualifying for tuition fee loans, regardless of economic status, available from the existing government-backed student loans company. The reforms also introduced means-tested grants and loans in addition to maintenance loans. Tuition fee loans are repayable only once a graduate’s annual salary exceeds a given threshold, which is currently a little below the median income. Once a graduate’s earnings are above a second, higher, income level of approximately double the median income, the maximum real interest rate charged is the inflation rate plus 3%.

However, there is concern over both the incentives and sustainability of high tuition fees, particularly when accrued under mortgage-style lending systems. High debt levels may deter children from lower middle-income backgrounds from entering higher education (Callender and Mason, 2017[91]; Perna, 2008[92]), and contracting such a high student debt may make these young adults reluctant to take out loans later in life – to buy a home, for example – potentially curbing their social mobility (Whitehead and Williams, 2017[87]). In the United States, certain states have now entirely abolished tuition fees for families earning below a certain threshold. New York, for example, has scrapped fees at state universities for households earning less than USD 125 000, which is just above the mean gross income of U.S. upper middle-income households.

In countries where tuition fees are low – such as Austria, Belgium, France, Italy and Switzerland – it is crucial to provide financial support for costs not directly related to tuition. Means-tested assistance for non-tuition costs serves distributional purposes most effectively when it is granted according to financial need rather than academic achievement (Heller and Marin, 2002[93]) and can also ensure costs remain manageable for governments. Loans and grants may cover tuition, but where necessary they may also be needed for attendance costs, i.e. money for books, housing and transport (Marcucci, 2013[94]). Providing students with support for accommodation and travel is of particular importance in the light of rapid price increases as for many children from lower middle-income families, the cost of housing and transport can discourage higher education attendance (Broton and Goldrick-Rab, 2016[95]). Poland uses a combination of low tuition fees, scholarships and loans to ensure that young people enjoy wide access to higher education. In 2011, it widened its system of grants by awarding them on the grounds of income rather than merit, as was the previous practice. It also raised the income threshold for grant eligibility by 30% to cover attendance costs, such as accommodation and transport.

5.3.3. Supporting households with high health care and long-term care costs

Large out-of-pocket payments for health and long-term care can be a significant burden on the budgets of middle-income families. The value and frequency of these expenditures, and thus inequalities in the access and use of services, are closely tied to the health care funding model in place (Devaux and de Looper, 2012[96]). Major reforms to health and long-term care systems are challenging for a number of reasons – the variety of stakeholders involved, the sheer size of systems, the need to maintain operating levels, their cost and the time required. However, even limited improvements to existing systems may yield significant positive effects.

The vast majority of OECD member countries offer universal health care coverage. It may be funded by taxes and publicly provided, as in the United Kingdom, or through compulsory public or private health insurance, often with regulated premiums, as in France and the Netherlands. The share of health care costs covered for particular services – known as the height of cost coverage – can vary notably, however. Significantly lower coverage for some outpatient specialist care, such as dentistry and optical services, and for the purchase of pharmaceuticals (Pearson et al., 2016[97]). Out-of-pocket spending on such services may be substantial, therefore, regardless of the health care financing model (OECD, 2018[98]).

Increasing the share of health care costs covered could provide substantial relief for middle-income households.43 Countries with co-payment systems could apply payment ceilings, either for specific services or for total co-payments. This instrument is in place in many EU countries, with the ceiling defined either as a nominal amount, as in Denmark and Sweden, or as a share of annual income, as in Austria, Germany and Luxembourg (Paris et al., 2016[80]). Voluntary health insurance can also serve to reduce the impact of out-of-pocket payments on access and financial hardship, such as in France and Slovenia, but this requires widespread access to private coverage and, in particular, measures to ensure that lower-income groups have access. More specifically, greater coverage of pharmaceutical spending would help to reduce financial hardship caused by out-of-pocket health spending (OECD, forthcoming[99]).44 Under insurance-based financing systems, governments have a number of options to support lower middle-income households including subsidising premium payments, and making sure that coverage is not lost if people default on premium payments.

In middle-income countries without universal health coverage, moving towards comprehensive coverage, regardless of employment status or income, would be the best solution for narrowing health inequalities and easing strains on household budgets. Indeed, in some of these countries, middle-income households spend a higher share of their budget on health care costs than poorer and richer households (Eurostat, 2018[70]; Komisar, 2013[100]).45 Extending health insurance subsidies to poorer middle-income households, particularly those with incomes marginally greater than current qualification thresholds, could help to reduce household budget pressures. Moreover, coverage gaps can be reduced to prevent geographical location or employment status determining the affordability of insurance.

Ageing populations and other societal changes across OECD countries have increased the demand for LTC for those who are not able to look after themselves.46 Most OECD governments have set up some form of collectively financed schemes for personal health care and nursing costs, though benefit provision is often tightly means-tested or linked to income, in order to control costs.47 For many middle-income households, LTC costs can equate to the majority of their income if they are not supported. And for some, this may require selling of their assets, such as their house, to pay for care costs, instead of being able to pass them on to their children.

Because it is hard to predict when and for how long people might need LTC services, the most efficient way of avoiding large out-of-pocket payments by the sick, elderly and their families is to pool LTC-related financial risks through targeted universal coverage (Colombo et al., 2011[101]).48 That is, within universal health systems, support can be directed to where needs are greatest, ensuring both fairness and value for money. However, only one-third of OECD countries offer a form of targeted universal coverage, whether through tax-funded social-care systems (as in the Nordics), dedicated social insurance schemes (in Germany, Japan, Korea, Luxembourg and the Netherlands) or via the general health care system (in Belgium).

In the absence of targeted LTC coverage, well-structured support can go a long way in helping informal carers, who are in most cases women caring for a family member or friend. One-fifth of OECD countries provide no financial support to carers (Colombo et al., 2011[101]), while others, which means-test, could be more generous. In the Nordic countries of Denmark, Finland and Norway, municipalities manage LTC and pay caregivers directly. Payments for caregivers vary from municipality to municipality and depend on the needs of the care recipient, but some countries apply a minimum lower threshold to ensure caregivers receive at least some compensation. The cost of administering these LTC programmes can be high, but they take pressure off local care institutions where access to care is often very restricted.

Aside from cash payments, countries deploy a range of measures to support informal carers – for example, training, respite care, pension credits and leaves of absence. Access to such support can be determined through needs assessments for both care recipients and caregivers, as is common in Australia, Sweden and the United Kingdom. However, these assessments are often complicated by the challenge of determining who can be considered as an informal carer. A number of countries prioritise financial compensation for carers who have had to give up their job or cut down on their participation in the labour market (Ansah et al., 2016[102]).49 Other support can include basic care training for the family members concerned, work reconciliation measures such as flexible work arrangements, and respite care. When implemented together these measures can collectively form holistic LTC plans.

5.4. Equipping the middle class with the skills needed in a changing workplace

Labour markets in OECD countries are undergoing profound transformation driven by globalisation, digitalisation, demographic trends and migration. These trends are changing the types and quality of jobs available and the skills sets in demand (Chapter 3 and OECD (2017[8]; forthcoming[12]; forthcoming[103])), thereby affecting the employment prospects, job security and earnings of middle-class workers. Production processes are increasingly integrated, with many mid-skilled, middle-pay jobs being offshored to lower-income countries. Manufacturing sectors, as the traditional backbone of a middle-class economy, are exposed to growing competition through the international trade in final goods (OECD, 2017[104]). The rapid digitalisation and automation of OECD economies is projected to reinforce labour market polarisation: an estimated one-in-six middle-income workers are currently employed in occupations at great risk of being automated (Figure 3.12). The risk is particularly high in occupations that do not require advanced cognitive skills or complex social interaction, such as clerical or production work (Nedelkoska and Quintini, 2018[105]). And while OECD analysis (forthcoming[12]) suggests that the ongoing transformations are unlikely to cause a net job destruction in OECD economies, they certainly give rise to significant anxiety. Nearly three-quarters of people in EU OECD countries are worried that robots and artificial intelligence will “steal people’s jobs” (European Commission, 2017[106]), and a similar share in the United States believe that artificial intelligence will destroy more jobs than it will create (Gallup, 2018[107]).

The growing demand for higher, non-routine skills puts pressure on countries to rethink and reform their education and training systems. The most widespread response to the growing skill demand has been to expand and encourage university education, with nearly one-in-two young adults OECD-wide completing a university-level degree (OECD, 2018[108]).50 Meanwhile, vocational education and training (VET) is less and less able to fulfil its traditional function of providing a pathway into middle-income groups. Indeed, the decline in the share of young people in the middle-income group (Figure 2.11) has been even more pronounced among those in low- and mid-skilled occupations.51 This is a concern, because VET has also historically provided learning pathways for those from less‑educated, lower-income backgrounds giving them the opportunity to develop the foundation skills and practical knowledge needed to find skilled work. If VET is to remain a catalyst for upward mobility into the middle class, many OECD countries will need to modernise their vocational programmes.

Strong provision of adult learning can help people develop the complex skills demanded by labour markets. Adult learning should be part of a wider move to move from a front-loaded education system – which often ends when students finish secondary or higher education – to lifelong learning. Adult learning programmes can be particularly useful to people employed in mid-skilled jobs and in industries and regions particularly at risk from labour market changes. They enable them to acquire the competencies they need to keep their jobs or to transition into better-quality employment (OECD, 2017_[8]). However, attendance rates and duration amongst older middle-income adults without tertiary education are notably lower than among those who are younger and better educated. Another obstacle is that employers tend to underinvest in employees whom they perceive to be less attached to the company – e.g. contract workers or women with caring responsibilities (OECD, 2019_[44]) – many of whom belong to the middle-income group.

5.4.1. Offering attractive vocational pathways

If vocational programmes are to steer learners into increasingly skill-intensive middle-class jobs, they should set out well-defined pathways that go further than upper-secondary level to post-secondary and higher education. A common reason why, in many OECD countries, vocational education appears unattractive to more able or ambitious learners is that VET qualifications are often seen as dead ends, affording little opportunity to deepen and update skills or transition to academic programmes (OECD, forthcoming[109]). This need not be the case, however, and a number of OECD countries have developed pathways to higher-level professional qualifications that are both attractive to learners and demanded by employers.

• In Austria, higher vocational colleges (Berufsbildende höhere Schulen, BHS) offer popular five-year upper- and post-secondary programmes that combine vocational and general education. Students acquire a higher vocational qualification and may choose to obtain a Matura university admission certificate. BHS graduates with a Matura certificate attain foundation skill levels comparable to those of general high school graduates (OECD, 2014[110]), and more of them continue higher education than take up work (Statistik Austria, 2018[111]; 2019[112]).

• Japan’s kōsen, prestigious colleges of technology, give students three years of specialised upper-secondary education followed by two years of post-secondary training in technical subjects, particularly engineering (OECD, 2017[113]). Since their introduction in the 1960s, the kōsen have proven highly responsive to the skill needs of the Japanese industry. They have also earned a reputation for giving excellent vocational training to students from lower socio-economic backgrounds. Kōsen graduates qualify for admission to higher education and nearly 40% of them do in fact go on to university.

While such post-secondary vocational programmes will likely play an increasingly important role in providing pathways towards high-skilled occupations, still too few countries systematically integrate work-based learning into their programmes as a quality-assured, credit-bearing element (OECD, 2014[114]). Work-based learning should be a core component of good-quality vocational programmes, because it provides trainees with an attractive learning environment and employers with influence over the training quality and contents. In Denmark, all academy profession programmes include a minimum of three months of work-based learning. Learners’ work placement outcomes are systematically assessed, and the links between vocational programmes and employers form part of the accreditation process. In Sweden, mandatory workplace training constitutes as much as one-quarter of the two-year professional programmes.

Many countries carry out skills assessment and anticipation exercises to align VET policy and social partner actions with the needs of rapidly changing labour markets (OECD, 2016[115]). These exercises can inform curriculum development and career guidance, and help identify the number of student places on courses:

• Germany’s Federal Institute for Vocational Education and Training produces short-term econometric forecasts of the supply and demand for apprenticeship places for the following year.

• Australia and Northern Ireland use skill needs assessments to promote apprenticeships in occupations and industries with greater demand for skilled labour. Australia relies on this information to direct funding to training organisations that offer programmes suited to address the identified skill needs. Northern Ireland extends the funds available for apprenticeships in these sectors to candidates of all ages, not just young people.

• The Austrian government uses skills assessment and anticipation exercises to determine the provision of post-secondary education. In order to receive accreditation for new programmes, Austrian legislation requires that Universities of Applied Sciences (Fachhochschulen) complete a demand and acceptance survey that evaluates the projected demand for each qualification seeking accreditation.

However, relatively few countries have used skills anticipation exercises specifically to predict possible skill imbalances arising from the digitalisation of the economy, and those who do have usually not systematically used those to change the design of skills policies and programmes (OECD, 2016[115]).

As workers’ careers are becoming more fragmented, equipping people with skills that are transferrable as they move between jobs, occupations or employment types is increasingly important. VET programmes should provide occupation-specific skills that are general enough to be portable across employers. More importantly, they need to build a set of broader cognitive skills (literacy, numeracy, problem-solving) and social and emotional skills (teamwork, communication, flexibility and the capacity to learn new skills) that will be of growing demand. Specifically, such skills will be needed as graduates working as professionals and technicians will be confronted with increasingly complicated tasks (OECD, 2015[116]). They are also more malleable and thus provide learners with greater flexibility in the job choice and with a tool for further learning. Results from the OECD Survey of Adult Skills indicate that, currently, students from short-cycle (i.e. less than bachelor’s level) professional training programmes, who later enter jobs involving higher-level technical and professional skills, partly lack basic skills, particularly numeracy (OECD, 2014[114]).

Making vocational education a more attractive prospect requires improved funding. In almost all countries, per-student expenditure on education increases with each educational level. The sole exception is post-secondary, non-tertiary education, where the average per-student expenditure is around that in primary education (OECD, 2018[108]).

5.4.2. Promoting adult learning – on the job and out of work

Adult learning programmes that target the middle class should seek principally to equip those at risk of losing their jobs with the skills needed for tomorrow’s economy. An effective adult learning provision that is relevant to the needs of the labour market is challenging. It involves meeting evolving demands and co‑ordinating stakeholder’s needs, including those of businesses, governments and social partners, to develop coherent programmes. Nevertheless, policies should focus on providing tailored training for the older and less educated and encouraging their take-up, given these groups have traditionally had low participation rates.

To ensure that adult education and training are well aligned with the skill needs of the economy, many OECD countries rely on the skills assessment and anticipation exercises described above, steering investment in adult learning towards programmes targeting in-demand skills (OECD, 2017[117]). Norway’s national Strategy for Skills Policy, for example, was jointly developed by 15 different partners.52 The co‑ordination required to put it in place helped to align the actors’ priorities for adult learning programmes, i.e. the sectors, skills and people that should be targeted (OECD, 2014[118]). Where adult learning systems need expanding in the future, financial co‑operation between actors is expected to ensure they receive sufficient resources.

Findings from skills assessment and anticipation exercises can also help to identify individuals who lack the skills required by the labour market and support them in retraining. Training incentives may target workers and firms in sectors where demand is declining, that are at high risk of automation or offshoring, or grappling with significant restructuring in how work is organised. Australia and Austria have both implemented retraining schemes in response to these changes:

• Australia targets employees in sectors where future employment opportunities are expected to be low or where large-scale closures are likely to affect the local labour market. In 2018, the Stronger Transitions Package was introduced to support individuals in five regions affected by structural change in transitioning to new jobs and training for the future. It begins by conducting comprehensive skills assessments. The findings they yield enable tailored support for job search preparation, resilience training, language, literacy and numeracy support, digital literacy training, financial management information, exploring self-employment options, health and wellbeing support, and industry awareness experience (OECD, 2019[53]).

• In Austria, social partners introduced Outplacement Labour Foundations (Arbeitsstiftung) to support workers in industries undergoing structural change. These foundations provide funding to cover the costs of training and related expenses, active job search assistance and career guidance. Employers can form a foundation for their employees by themselves or as part of a group of companies – the latter of which has been apparent at a regional and sectoral level in response to large-scale job cuts. Foundations are co‑funded by local labour actors, who include the public employment service and the employers affected (OECD, 2018[3]).

To encourage workers to reskill, learning opportunities often need to be more accessible and flexible and include targeted career advice and guidance to help individuals make informed decisions about investing and committing to further skills development. One way to improve access to training is to make it available to all, not just those in employment. France, for example, introduced personal training accounts in 2015, which allow individuals to accumulate entitlement to training hours. The accrued entitlement is transferable between jobs and employment statuses. Iceland uses a similar personal training account model. Designing training programmes in modular formats can help provide flexibility so that training can be undertaken around busy work schedules and family responsibilities (OECD, forthcoming[12]). And to lighten the training burden, particularly for older workers, the formal recognition of skills acquired through experience should be strengthened.

To encourage uptake and provision, public financial support should be made available for training. This would reduce training costs for employers, therefore incentivising greater investment in training for at-risk employees. In the Netherlands, workers aged 45 and over can benefit from subsidised career development guidance (Ontwikkeladvies), while in Germany, the public employment service funds the training of lower-skilled and older workers in small and medium-sized enterprises through the WeGebAU programme. The programme gives companies a 75% subsidy towards the cost of training workers aged 45 and older (Dauth, 2017[119]). Another example is from Austria, where the public employment service covers course and course-related costs (e.g. learning materials, special clothing, and accommodation) for jobseekers and low-income employees so that financial barriers do not deter jobseekers and low-income employees from taking up training.

References

[124] Ackerman, B. and A. Alstott (1999), The stakeholder society, Yale University Press, New Haven & London.

[22] Adema, W., C. Clarke and V. Frey (2015), “Paid Parental Leave: Lessons from OECD Countries and Selected U.S. States”, OECD Social, Employment and Migration Working Papers, No. 172, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jrqgvqqb4vb-en.

[88] Adema, W., C. Clarke and O. Thévenon (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, https://www.oecd.org/els/family/Who_uses_childcare-Backgrounder_inequalities_formal_ECEC.pdf.

[66] Aghion, P. et al. (2017), “Tax Simplicity and Heterogeneous Learning”, Working Paper Series, No. 24049, NBER, http://www.nber.org/papers/w24049.

[43] Alstadsæter, A., N. Johannesen and G. Zucman (2018), “Who owns the wealth in tax havens? Macro evidence and implications for global inequality”, Journal of Public Economics, Vol. 162, pp. 89-100, http://dx.doi.org/10.1016/j.jpubeco.2018.01.008.

[44] Alstadsaeter, A., N. Johannesen and G. Zucman (2017), “Tax Evasion and Inequality”, No. 23772, NBER Working Paper, Cambridge, MA, https://www.nber.org/papers/w23772.pdf.

[84] Andrews, D. and A. Sánchez (2011), “The Evolution of Homeownership Rates in Selected OECD Countries: Demographic and Public Policy Influences”, OECD Journal: Economic Studies, Vol. 2011, http://dx.doi.org/10.1787/19952856.

[102] Ansah, J. et al. (2016), “Projecting the effects of long-term care policy on the labor market participation of primary informal family caregivers of elderly with disability: insights from a dynamic simulation model”, BMC Geriatrics, Vol. 16/1, http://dx.doi.org/10.1186/s12877-016-0243-0.

[7] Atkinson, A. (2015), Inequality: what can be done?, Harvard University Press, Cambridge, MA.

[40] Bach, S. (2012), “Abbau der kalten Progression: Nicht die einzige Herausforderung beim Einkommensteuertarif”, DIW Wochenbericht, No. 12.2012, German Institute for Economic Research DIW, Berlin, https://www.diw.de/documents/publikationen/73/diw_01.c.395418.de/12-12.pdf.

[32] Baker, M., J. Gruber and K. Milligan (2015), “Non-cognitive deficits and young adult outcomes: the long-run impacts of a universal child care program”, NBER Working Paper, No. 21571, National Bureau of Economic Research, http://www.nber.org/papers/w21571.

[31] Baker, M., J. Gruber and K. Milligan (2008), “Universal Child Care, Maternal Labor Supply, and Family Well‐Being”, Journal of Political Economy, Vol. 116/4, pp. 709-745, http://dx.doi.org/10.1086/591908.

[6] Balestra, C. and R. Tonkin (2018), “Inequalities in household wealth across OECD countries: Evidence from the OECD Wealth Distribution Database”, Statistics and Data Working Paper, No. 2018/01, OECD, Paris, https://doi.org/10.1787/7e1bf673-en.

[69] Barbosa, F. et al. (2017), Reinventing construction: A route to higher productivity, McKinsey Global Institute, http://www.mckinsey.com/mgi.

[48] Bastian, J. and M. Jones (2018), “Do EITC Expansions Pay for Themselves? Effects on Tax Revenue and Public Assistance Spending”, https://drive.google.com/file/d/1GbBeeQzfGH9fF9Y1u5rS55Sn3eStBWE7/view.

[121] Bick, A. et al. (2018), “Long-term changes in married couples’ labor supply and taxes: evidence from the US and Europe since the 1980s”, NBER Working Paper, No. 24995, National Bureau of Economic Research, Cambridge, MA, http://dx.doi.org/10.3386/w24995.

[60] Bloom, H. et al. (2001), “Testing a Financial Incentive to Promote Re-employment among Displaced Workers: The Canadian Earnings Supplement Project (ESP)”, Journal of Policy Analysis and Management, Vol. 20/3, pp. 505-523, http://dx.doi.org/10.1002/pam.1005.

[52] Brandt, N. (2015), “Vocational training and adult learning for better skills in France”, OECD Economics Department Working Papers, No. 1260, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jrw21kjthln-en.

[95] Broton, K. and S. Goldrick-Rab (2016), “The Dark Side of College (Un)Affordability: Food and Housing Insecurity in Higher Education”, Change: The Magazine of Higher Learning, Vol. 48/1, pp. 16-25, http://dx.doi.org/10.1080/00091383.2016.1121081.

[131] Calavita, N. and A. Mallach (eds.) (2010), Inclusionary Housing in International Perspective: Affordable Housing, Social Inclusion, and Land Value Recapture, Lincoln Institute of Land Policy.

[91] Callender, C. and G. Mason (2017), “Does Student Loan Debt Deter Higher Education Participation? New Evidence from England”, The ANNALS of the American Academy of Political and Social Science, Vol. 671/1, pp. 20-48, http://dx.doi.org/10.1177/0002716217696041.

[120] Campbell, J., N. Ikegami and M. Gibson (2010), “Lessons From Public Long-Term Care Insurance In Germany And Japan”, Health Affairs, Vol. 29/1, pp. 87-95, http://dx.doi.org/10.1377/hlthaff.2009.0548.

[47] Carcillo, S., A. Hijzen and S. Thewissen (forthcoming), “Implementing the OECD Jobs Strategy in France: Assessing recent reforms and policy directions”, OECD Social, Employment and Migration Working Papers, OECD, Paris.

[35] Causa, O. and M. Hermansen (2017), “Income redistribution through taxes and transfers across OECD countries”, OECD Economics Department Working Papers, No. 1453, OECD Publishing, Paris, https://dx.doi.org/10.1787/bc7569c6-en.

[41] Causa, O., A. Vindics and O. Akgun (2018), “An empirical investigation on the drivers of income redistribution across OECD countries”, OECD Economics Department Working Papers, No. 1488, OECD Publishing, Paris, https://dx.doi.org/10.1787/5cb47f33-en.

[68] Causa, O. and N. Woloszko (forthcoming), “Housing, Wealth Accumulation and Wealth Distribution: Evidence and Stylized Facts”, OECD Economics Department Working Papers, OECD Publishing, Paris.

[78] City of Vienna (2017), Municipal Housing in Vienna. History, facts & figures, https://www.wienerwohnen.at/dms/workspace/SpacesStore/aa75756e-2836-4e77-8cfd-f37cc15e2756/1.0Wiener-Gemeindebau-engl.pdf.

[101] Colombo, F. et al. (2011), Help Wanted?: Providing and Paying for Long-Term Care, OECD Health Policy Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264097759-en.

[132] Corlett, A. and L. Judge (2017), Home Affront: housing across the generations, Resolution Foundation, UK, https://www.resolutionfoundation.org/publications/home-affront-housing-across-the-generations/.

[119] Dauth, C. (2017), “Weiterbildung Geringqualifizierter und Beschäftigter älterer Arbeitnehmer in Unternehmen (WeGebAU)”, in Moller, J. and U. Walwei (eds.), Arbeitsmarkt Kompakt: Analysen, Daten, Fakten, Institut für Arbeitsmarkt und Berufsforschung der Bundesagentur für Arbeit (IAB) / Bertelsmann Verlag, Nürnberg/ Bielefeld, http://amk.iab.de/content/home/iab-arbeitsmarkt-kompakt.pdf.

[81] Davies, B. et al. (2017), Lessons from Germany. Tenant power in the rental market, Institute for Public Policy Research, London, https://www.ippr.org/files/publications/pdf/lessons-from-germany-jan17.pdf.

[13] De Stefano, V. (2018), Collective bargaining of platform workers: domestic work leads the way, Regulating for Globalization - Trade, Labor and EU Law Perspectives, http://regulatingforglobalization.com/2018/12/10/collective-bargaining-of-platform-workers-domestic-work-leads-the-way/.

[77] Delance, P. (2018), “11 millions de personnes sont locataires d’un logement social”, Insee Première, No. 1715, Institut National de la statistique et des études économiques (Insee), Paris, https://www.insee.fr/fr/statistiques/3635547.

[96] Devaux, M. and M. de Looper (2012), “Income-Related Inequalities in Health Service Utilisation in 19 OECD Countries, 2008-2009”, OECD Health Working Papers, No. 58, OECD Publishing, Paris, https://dx.doi.org/10.1787/5k95xd6stnxt-en.

[83] Diamond, R., T. Mcquade and F. Qian (2018), The Effects of Rent Control Expansion on Tenants, Landlords, and Inequality: Evidence from San Francisco *, National Bureau of Economic Research, http://dx.doi.org/10.3386/w24181.

[64] Elliott, W. and M. Lewis (2018), Making education work for the poor: the potential of children’s savings accounts, Oxford University Press, Oxford.

[17] Eurofound (2018), Concept and practice of a living wage, Publications Office of the European Union, Luxembourg, http://dx.doi.org/10.2806/024749.

[106] European Commission (2017), Special Eurobarometer 460 - Attitudes towards the impact of digitisation and automation on daily life, European Union, http://ec.europa.eu/commfrontoffice/publicopinion/index.cfm/ResultDoc/download/DocumentKy/78998.

[70] Eurostat (2018), Final Consumption Expenditure of Households, http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=nama_10_co3_p3&lang=en.

[10] Farber, H. et al. (2018), “Unions and inequality over the twentieth century: new evidence from survey data”, NBER Working Paper, No. 24587, National Bureau of Economic Research, Cambridge, MA, http://www.nber.org/papers/w24587.

[133] Fitzsimons, J. (2014), The German private rented sector – a holistic approach, The Knowledge Centre for Housing Economics, https://www.bvc.dk/media/1332/the_german_private_rented_sector_web.pdf.

[86] Gale, W. (2019), Fiscal Therapy : balancing today’s needs with tomorrow’s obligations., Oxford University Press.

[107] Gallup (2018), Optimism and Anxiety: Views on the Impact of Artificial Intelligence and Higher Education’s Response, Gallup Inc, https://news.gallup.com/file/reports/226475/Gallup-Northeastern_Report_Artificial_Intelligence_2018.pdf.

[62] Hamilton, D. and W. Darity Jr. (2010), “Can ’Baby Bonds’ Eliminate the Racial Wealth Gap in Putative Post-Racial America?”, The Review of Black Political Economy, Vol. 37, pp. 207-216, http://dx.doi.org/10.1007/s12114-010-9063-1.

[93] Heller, D. and P. Marin (2002), Who Should We Help? The Negative Social Consequences of Merit Scholarships, Harvard University, Cambrige, MA.

[46] Hoynes, H. and A. Patel (2015), “Effective Policy for Reducing Inequality? The Earned Income Tax Credit and the Distribution of Income”, NBER Working Paper, No. 21340, National Bureau of Economic Research, Cambridge, MA, http://dx.doi.org/10.3386/w21340.

[37] Immervoll, H. (2005), “Falling up the stairs: The effects of “bracket creep” on household incomes”, Review of Income and Wealth, Vol. 51/1, pp. 37-62, http://dx.doi.org/10.1111/j.1475-4991.2005.00144.x.

[45] Immervoll, H. and M. Pearson (2009), “A Good Time for Making Work Pay? Taking Stock of In-Work Benefits and Related Measures across the OECD”, OECD Social, Employment and Migration Working Papers, No. 81, OECD Publishing, Paris, https://dx.doi.org/10.1787/225442803245.

[36] Immervoll, H. and L. Richardson (2013), “Redistribution Policy in Europe and the United States: Is the Great Recession a ’Game Changer’ for Working-age Families?”, OECD Social, Employment and Migration Working Papers, No. 150, OECD Publishing, Paris, https://dx.doi.org/10.1787/5k44xwtc0txp-en.

[79] Institute for Urban Strategies (2018), Global Power City Index, The Mori Memorial Foundation, Tokyo, http://mori-m-foundation.or.jp/pdf/GPCI2018_summary.pdf.

[130] Kesselheim, A. et al. (2015), “Prescription Drug Insurance Coverage and Patient Health Outcomes: A Systematic Review”, American Journal of Public Health, Vol. 105/2, pp. e17-e30, http://dx.doi.org/10.2105/ajph.2014.302240.

[51] Kettemann, A., F. Kramarz and J. Zweimüller (2016), “Beyond Severance Pay: Labor Market Responses to the Introduction of Occupational Pensions in Austria”, mimeo, University of Zurich, https://www.econ.uzh.ch/dam/jcr:13c720d8-130e-4d2e-bc91-07e9881ddbbc/Kettemann.pdf.

[128] Kohl, S. (2018), “More Mortgages, More Homes? The Effect of Housing Financialization on Homeownership in Historical Perspective”, Politics & Society, Vol. 46/2, pp. 177-203, http://dx.doi.org/10.1177/0032329218755750.

[72] Kok, N., P. Monkkonen and J. Quigley (2014), “Land use regulations and the value of land and housing: An intra-metropolitan analysis”, Journal of Urban Economics, Vol. 81, pp. 136-148, http://dx.doi.org/10.1016/j.jue.2014.03.004.

[100] Komisar, H. (2013), The Effects of Rising Health Care Costs on Middle-Class Economic Security, AARP Public Policy Institute, Washington, DC.

[9] Marcadent, P. (2018), Wage dynamics and labour market institutions, European Central Bank, Sintra, Portugal, https://www.ecb.europa.eu/pub/pdf/sintra/ecb.forumcentbank201810.en.pdf.

[94] Marcucci, P. (2013), “The politics of student funding policies from a comparative perspective”, in Heller, D. and C. Callender (eds.), Student Financing of Higher Education: A comparative perspective, Routledge, Abingdon.

[58] Matsaganis, M. et al. (2016), “Non-standard employment and access to social security benefits”, Social Situation Monitor, Research Note, No. 8/2015, European Commission, http://ec.europa.eu/social/BlobServlet?docId=15687&langId=en.

[105] Nedelkoska, L. and G. Quintini (2018), “Automation, skills use and training”, OECD Social, Employment and Migration Working Papers, No. 202, OECD Publishing, Paris, https://dx.doi.org/10.1787/2e2f4eea-en.

[53] OECD (2019), Getting Skills Right: Future-Ready Adult Learning Systems, Getting Skills Right, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264311756-en.

[26] OECD (2019), PF2.2: Parents’ use of childbirth-related leave, http://www.oecd.org/els/family/PF2-2-Use-childbirth-leave.pdf.

[54] OECD (2019), Policy Responses to New Forms of Work, OECD Publishing, Paris, https://dx.doi.org/10.1787/0763f1b7-en.

[5] OECD (2018), A Broken Social Elevator? How to Promote Social Mobility, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264301085-en.

[108] OECD (2018), Education at a Glance 2018: OECD Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/eag-2018-en.

[30] OECD (2018), Engaging Young Children: Lessons from Research about Quality in Early Childhood Education and Care, Starting Strong, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264085145-en.

[2] OECD (2018), Good Jobs for All in a Changing World of Work: The OECD Jobs Strategy, OECD Publishing, Paris, https://doi.org/10.1787/9789264308817-en.

[98] OECD (2018), Health Expenditure and Financing, https://stats.oecd.org/index.aspx?DataSetCode=SHA.

[14] OECD (2018), “Issues note: The emergence of new forms of work and their implications for labour relations”, Paper presented at the 1st Meeting of the G20 Employment Working Group, http://www.oecd.org/g20/topics/employment-education-and-social-policies/OECD-Note-on-The-emergence-of-new-forms-of-work.pdf.

[1] OECD (2018), OECD Economic Outlook, Volume 2018 Issue 2: Preliminary version, OECD Publishing, Paris, https://dx.doi.org/10.1787/eco_outlook-v2018-2-en.

[3] OECD (2018), OECD Employment Outlook 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/empl_outlook-2018-en.

[28] OECD (2018), OECD Family Database PF3.2: Enrolment in childcare and pre-school, http://www.oecd.org/els/soc/PF3_2_Enrolment_childcare_preschool.pdf.

[135] OECD (2018), Risks that Matter Early Results from the 2018 OECD Cross-National Survey on Social and Economic Risks, OECD, Paris, http://www.oecd.org/social/risks-that-matter.htm.

[34] OECD (2018), Tax Policy Reforms 2018: OECD and Selected Partner Economies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264304468-en.

[65] OECD (2018), Taxation of Household Savings, OECD Tax Policy Studies, No. 25, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264289536-en.

[49] OECD (2018), “The Future of Social Protection: What works for non-standard workers?”, Policy Brief on the Future of Work, https://www.oecd.org/employment/future-of-social-protection.pdf.

[11] OECD (2018), The Future of Social Protection: What Works for Non-standard Workers?, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264306943-en.

[42] OECD (2018), The Role and Design of Net Wealth Taxes in the OECD, OECD Tax Policy Studies, No. 26, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264290303-en.

[25] OECD (2017), Dare to Share: Germany’s Experience Promoting Equal Partnership in Families, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264259157-en.

[117] OECD (2017), Getting Skills Right: Skills for Jobs Indicators, Getting Skills Right, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264277878-en.

[71] OECD (2017), Health at a Glance 2017: OECD Indicators, OECD Publishing, Paris, https://dx.doi.org/10.1787/health_glance-2017-en.

[113] OECD (2017), Investing in Youth: Japan, Investing in Youth, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264275898-en.

[8] OECD (2017), OECD Employment Outlook 2017, OECD Publishing, Paris, https://dx.doi.org/10.1787/empl_outlook-2017-en.

[122] OECD (2017), OECD Family database - PF2.1: Key characteristics of parental leave systems, http://www.oecd.org/els/family/database.htm.

[24] OECD (2017), OECD Family Database - PF2.4: Parental leave replacement rates, http://www.oecd.org/els/family/database.htm.

[104] OECD (2017), OECD Skills Outlook 2017: Skills and Global Value Chains, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264273351-en.

[123] OECD (2017), Starting Strong 2017: Key OECD Indicators on Early Childhood Education and Care, Starting Strong, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264276116-en.

[33] OECD (2017), Tax Policy Reforms 2017: OECD and Selected Partner Economies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264279919-en.

[23] OECD (2017), The Pursuit of Gender Equality: An Uphill Battle, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264281318-en.

[115] OECD (2016), Getting Skills Right: Assessing and Anticipating Changing Skill Needs, Getting Skills Right, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264252073-en.

[90] OECD (2016), “How much do tertiary students pay and what public support do they receive?”, Education Indicators in Focus, No. 41, OECD Publishing, Paris, https://dx.doi.org/10.1787/5jlz9zk830hf-en.

[85] OECD (2016), OECD Affordable Housing Database. PH2.1 Public spending on financial support to home buyers, OECD, Paris, https://www.oecd.org/els/family/PH2-1-Public-spending-support-to-home-buyers.pdf.

[80] OECD (2016), OECD Affordable Housing Database. PH4.1 Public spending on support to social rental housing, OECD, Paris, https://www.oecd.org/els/family/PH4-1-Public-spending-social-rental-housing.pdf.

[82] OECD (2016), OECD Affordable Housing Database. PH6.1. Rental Regulation, OECD, Paris, https://www.oecd.org/els/family/PH6-1-Rental-regulation.pdf.

[4] OECD (2016), OECD Employment Outlook 2016, OECD Publishing, Paris, https://dx.doi.org/10.1787/empl_outlook-2016-en.

[21] OECD (2016), Taxing Wages 2016, OECD Publishing, Paris, http://dx.doi.org/10.1787/tax_wages-2016-en.

[59] OECD (2015), Back to Work: Sweden: Improving the Re-employment Prospects of Displaced Workers, Back to Work, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264246812-en.

[129] OECD (2015), Fiscal Sustainability of Health Systems: Bridging Health and Finance Perspectives, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264233386-en.

[56] OECD (2015), In It Together: Why Less Inequality Benefits All, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264235120-en.

[15] OECD (2015), Minimum wages after the crisis: Making them pay, OECD, Paris, http://www.oecd.org/social/Focus-on-Minimum-Wages-after-the-crisis-2015.pdf.

[16] OECD (2015), OECD Employment Outlook 2015, OECD Publishing, Paris, https://dx.doi.org/10.1787/empl_outlook-2015-en.

[116] OECD (2015), OECD Skills Outlook 2015: Youth, Skills and Employability, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264234178-en.

[110] OECD (2014), OECD Skills Strategy Diagnostic Report: Austria 2014, OECD Skills Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264300255-en.

[118] OECD (2014), OECD Skills Strategy Diagnostic Report: Norway 2014, OECD Skills Studies, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264298781-en.

[114] OECD (2014), Skills beyond School: Synthesis Report, OECD Reviews of Vocational Education and Training, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264214682-en.

[50] OECD (2013), OECD Indicators of Employment Protection, http://www.oecd.org/employment/emp/oecdindicatorsofemploymentprotection.htm.

[19] OECD (2012), Closing the Gender Gap: Act Now, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264179370-en.

[126] OECD (2011), Divided We Stand: Why Inequality Keeps Rising - OECD, OECD Publishing, Paris, http://www.oecd.org/els/soc/dividedwestandwhyinequalitykeepsrising.htm.

[29] OECD (2011), Starting Strong III: A Quality Toolbox for Early Childhood Education and Care, Starting Strong, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264123564-en.

[18] OECD (2011), Taxation and Employment, OECD Tax Policy Studies, No. 21, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264120808-en.

[38] OECD (2008), Taxing Wages, 2006/2007: Special feature: Tax reforms and tax burdens, OECD, Paris.

[67] OECD (forthcoming), Economic Policy Reforms 2019: Going for Growth Report, OECD Publishing, Paris.

[99] OECD (forthcoming), Health for Everyone? Social Inequalities in Health and Health Systems, OECD Health Policy Studies, OECD Publishing.

[55] OECD (forthcoming), OECD Economic Surveys: France 2019, OECD Publishing, Paris.

[12] OECD (forthcoming), OECD Employment Outlook 2019, OECD Publishing, Paris.

[103] OECD (forthcoming), “OECD Skills Outlook 2019: Thriving in a Digital World”, OECD Publishing, Paris.

[109] OECD (forthcoming), OECD Skills Strategy 2019: Skills to shape a better future, OECD Publishing, Paris.

[125] Paine, T. (1797), “Agrarian Justice”, Social Insurance History, The United States Social Security Administration, https://www.ssa.gov/history/paine4.html.

[39] Parlament der Republik Österreich (2018), Budgetäre Auswirkungen der kalten Progression, Budgetdienst, Vienna, https://www.parlament.gv.at/ZUSD/BUDGET/2018/BD_-_Anfragebeantwortung_zu_budgetaeren_Auswirkungen_der_Kalten_Progression.pdf.

[97] Pearson, M. et al. (2016), Universal Health Coverage and Health Outcomes, OECD, Paris, https://www.oecd.org/els/health-systems/Universal-Health-Coverage-and-Health-Outcomes-OECD-G7-Health-Ministerial-2016.pdf.

[92] Perna, L. (2008), “Understanding High School Students’ Willingness to Borrow to Pay College Prices”, Research in Higher Education, Vol. 49/7, pp. 589-606, http://dx.doi.org/10.1007/s11162-008-9095-6.

[74] Quigley, J. and L. Rosenthal (2005), “The Effects of Land Use Regulation on the Price of Housing: What Do We Know? What Can We Learn?”, Cityscape: A Journal of Policy Development and Research, Vol. 8/1.

[75] 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, Brookings Institution Press.

[89] Rodgers, L. (2018), “Give credit where? The incidence of child care tax credits”, Journal of Urban Economics, Vol. 108, pp. 51-71, http://dx.doi.org/10.1016/j.jue.2018.10.002.

[73] Saks, R. (2008), “Job creation and housing construction: Constraints on metropolitan area employment growth”, Journal of Urban Economics, Vol. 64/1, pp. 178-195, http://dx.doi.org/10.1016/J.JUE.2007.12.003.

[76] Salvi del Pero, A. et al. (2016), “Policies to promote access to good-quality affordable housing in OECD countries”, OECD Social, Employment and Migration Working Papers, No. 176, OECD Publishing, Paris, http://dx.doi.org/10.1787/5jm3p5gl4djd-en.

[57] Spasova, S. et al. (2017), “Access to social protection for people working on non-standard contracts and as self-employed in Europe - A study of national policies”, European Social Policy Network (ESPN), European Commission, Brussels, http://dx.doi.org/dx.doi.org/10.2767/700791.

[112] Statistik Austria (2019), Bildungsbezogenes Erwerbskarrierenmonitoring – BibEr, http://www.statistik.at/web_de/statistiken/menschen_und_gesellschaft/bildung_und_kultur/bildungsbezogenes_erwerbskarrierenmonitoring_biber/index.html.

[111] Statistik Austria (2018), Bildung in Zahlen 2016/17 - Schlüsselindikatoren und Analysen, Statistik Austria, Vienna, http://www.statistik.at/web_de/services/publikationen/5/index.html?includePage=detailedView§ionName=Bildung%2C+Kultur&pubId=508.

[134] Sturtevant, L. (2016), Separating Fact from Fiction to Design Effective Inclusionary Housing Programs, Center for Housing Policy, National Housing Conference, https://ihiusa.org/wp-content/uploads/Seperating-Fact-from-Fiction.pdf.

[27] Thévenon, O. (2013), “Drivers of Female Labour Force Participation in the OECD”, OECD Social, Employment and Migration Working Papers, No. 145, OECD Publishing, Paris, https://dx.doi.org/10.1787/5k46cvrgnms6-en.

[127] Thévenon, O. and A. Solaz (2013), “Labour Market Effects of Parental Leave Policies in OECD Countries”, OECD Social, Employment and Migration Working Papers, No. 141, OECD Publishing, Paris, https://dx.doi.org/10.1787/5k8xb6hw1wjf-en.

[20] Thomas, A. and P. O’Reilly (2016), “The Impact of Tax and Benefit Systems on the Workforce Participation Incentives of Women”, OECD Taxation Working Papers, No. 29, OECD, Paris, https://doi.org/10.1787/d950acfc-en.

[63] United Kingdom Parliament (2010), House of Commons General Committee: Draft Child Trust Funds (Amendment No. 3) Regulations, https://publications.parliament.uk/pa/cm201011/cmgeneral/deleg5/100720/100720s01.htm.

[61] Wandner, S. (2016), Wage Insurance as a Policy Option in the United States, W.E. Upjohn Institute, http://dx.doi.org/10.17848/wp16-250.

[87] Whitehead, C. and P. Williams (2017), “Changes in the regulation and control of mortgage markets and access to owner-occupation among younger households”, OECD Social, Employment and Migration Working Papers, No. 196, OECD Publishing, Paris, https://dx.doi.org/10.1787/e16ab00e-en.

Notes

← 1.  In comparison, two out of three find taxes too high for low-income households, while only about one-quarter find them too high for high-income households (OECD calculations based on the International Social Survey Programme 2016: Role of Government). Those numbers are consistent with recent results from the OECD Risks that Matter Survey, which indicate that, on average, 58% of middle-income households in OECD countries consider that they do not receive a fair share of public benefits for the taxes and social security contributions that they pay (OECD, 2018[135]).

← 2.  Among middle-income households of working age, the imbalance between personal direct taxes paid and benefit received averages 17% of disposable income OECD-wide (Figure 2.16). Among middle-income households with an elderly head of household, benefits exceed taxes by 60% of disposable income. Middle-income households’ average net tax burden has not significantly increased over the last decade in OECD countries.

← 3.  Real-estate wealth represents the largest share of the gross assets of households in the middle quintile – 77% on average across the OECD (Balestra and Tonkin, 2018[6]).

← 4.  Data on trends in collective bargaining coverage by type of occupation are limited, and the available information on occupational patterns of middle-income workers (Figure 3.3) is too coarse to reliably estimate coverage rates.

← 5.  This holds true at least in systems where bargaining takes place at firm level. The work environment also tends to be of better quality in firms with a recognised form of employee representation, such as a trade union or works council. They also benefit from a better work environment, largely because of lower work intensity, more training options and better prospects for career advancement.

← 6.  See OECD (2018[2]; 2018[3]) for detailed discussions.

← 7.  Freelancers can apply to become employees and so be covered by the collective agreement. They automatically obtain employee status after 100 hours of work unless if they choose to opt out.

← 8.  Estimating the exact share of workers on the minimum wage in middle-income households is very difficult, because data with precise information on both incomes and hourly wages (or earnings and hours worked) are lacking in most countries.

← 9.  Higher minimum wages are thus associated with lower wage inequality, both between and within countries: a 10% increase in the minimum-to-median wage ratio reduces earnings inequality by 3% (2011[126]).

← 10.  Gross minimum wages average 50% of the median wage across OECD countries, ranging from below 40% in the Czech Republic, Mexico, the United States, Estonia and Japan to 60% and over in Turkey, Chile, France and Slovenia.

← 11.  The UK government started significantly raising the statutory minimum wage for workers aged 25 years and over from 2016, referring to it as the National Living Wage. The level of the National Living Wage is not based on any estimated basic cost of living, however, and it is separate from the (higher) living wages paid on a voluntary basis by some UK companies.

← 12.  Recent cross-country evidence for seven European countries and the United States indicates that changes in labour income taxes have accounted for a large share of the changes in employment rates and hours worked by married women since the 1980s (Bick et al., 2018[121]).

← 13.  In such systems, second earners effectively pay tax at a higher part of the income tax rate schedule (i.e., usually a higher tax bracket) than they would under individual-based taxation, because the primary earner is already gaining the full benefit from the lower part of the tax rate schedule (OECD, 2016[21]). The gap in average tax rates between a primary and a secondary earner is largest in Belgium, which operates a partial-splitting system that has very similar effects. In Germany, family-based taxation is optional but usually beneficial for couples with large earnings differences.

← 14.  Childbirth and parenting responsibilities are often associated with a break in women’s careers. The gender gap in labour force participation, earnings and hours worked is much greater for mothers than for women without children (OECD, 2018[3]). Raising women’s lifetime earnings would also increase their pension entitlements, so reducing concerns about old-age poverty, which tends to be higher for women than for men.

← 15.  Longer paid-leave durations are associated with a modest increase in female employment rates, the female-to-male employment ratio, and the relative number of hours worked by women compared to men. However, effects turn negative when paid leave is longer than two years. Greater public spending during the paid leave is associated with higher full-time employment rates (Thévenon, 2013[27]; Thévenon and Solaz, 2013[127]).

← 16.  In the United Kingdom and the United States, for instance, the equivalised net income for an average-earnings, dual-earner, two-child couple during the first month after the birth of the second child was less than half of what it was prior to birth in 2014.

← 17.  Maternity or paternity leave is the employment-related leave of absence for employed mothers or fathers at or around the time of childbirth. Parental leave is often in addition to specific maternity or paternity leave and frequently follows the maternity leave period. Entitlement to the parental leave period is often individual (i.e. each parent has their own entitlement), while entitlement to public income support is frequently family-based (OECD, 2017[122]).

← 18.  Enrolment of the under‑3s in formal ECEC has increased fourfold in Korea and nearly threefold in Germany since 2005 (OECD, 2017[123]).

← 19.  Moreover, several countries increased tax allowances for families with children (Germany and Israel), single parents (Luxembourg) or home carers (Ireland) (OECD, 2017[33]; OECD, 2018[34]).

← 20.  Also, capital gains are usually taxed only after the sale of assets so that borrowing against these assets enables the wealthy to further minimise their tax burden.

← 21.  The new International Value Added Tax/Goods and Services Tax Guidelines is estimated to have increased EU tax revenues by over EUR 3 billion.

← 22.  In 2018, 4 500 bilateral exchanges of information were made between 86 sending and partner jurisdictions concerning assets held by non-tax residents.

← 23.  Net wealth taxes tend to be more distortionary and less equitable than capital gains taxes, as they are generally levied irrespective of the actual returns that taxpayers earn on their assets. But where a country’s overall tax burden – on capital, inheritance and gifts inter vivos – is low, or their implementation unfeasible, net wealth taxes can be an imperfect policy substitute. They may be useful in countries with dual income tax systems that tax capital income at flat, often low, rates. Only four OECD countries levied recurrent taxes on individuals’ net wealth in 2017, down from 12 in 1990.

← 24.  OECD calculations for a single person without children using the OECD tax-benefit model, www.oecd.org/social/benefits-and-wages.

← 25.  Interest rate for bonds with a three‑year term. The instrument was withdrawn from general sale in 2011 because of high investment levels but will be reissued in May 2019.

← 26.  The philosopher Tom Paine is believed to be the first to have proposed a minimum inheritance, in the late eighteenth century (1797[125]). Ackerman and Alstott (1999[124]) reignited debate on the topic in the 1990s, suggesting giving USD 80 000 to all US citizens upon turning 21. More recently, the late Anthony Atkinson was a vocal proponent of its implementation (2015[7]).

← 27.  Children in the lowest wealth quartile would receive up to USD 50 000 or USD 60 000.

← 28.  The CTF provided new parents with a GBP 250 voucher which they could invest, tax-free, for 18 years in a shares or cash-based account for every child born after 1 September, 2002. Children born in low-income households received grants that were twice as large. A further GBP 250 were provided when the child turned seven, while in the meantime the account could be topped-up with an additional GBP 1 200 per year. For middle-income families a well managed fully topped up fund could yield a maximum of GBP 35 000.

← 29.  In 2017, 382 000 children used one of the 54 CSA programmes in 32 US states.

← 30.  On average, nearly 9% of mortgaged middle-class homeowners in OECD countries are defined as being overburdened by housing costs, as they spend at least 40% of their monthly disposable income on mortgage payments. In 2016, this figure reached 20% in Ireland, Greece and Sweden (Salvi del Pero et al., 2016[76]).

← 31.  This trend is striking in the United Kingdom, where millennials in their 30s have only half the chance of owning their home than the baby boomers at the same age (Corlett and Judge, 2017[132]).

← 32.  Amongst EU OECD countries, 45% of lower middle-income families with a child under five years old report making little or no use of formal childcare. They explain that it is due to high costs. Out-of-pocket childcare costs for families with moderate earnings are particularly high in Ireland, Switzerland and the United Kingdom. In these countries, children aged three or younger from middle-income backgrounds are one-quarter to one-third less likely to use formal ECEC than children from high-income backgrounds (Adema, Clarke and Thévenon, 2016[88]).

← 33.  Besides the aforementioned intra-urban aspects, cities and national governments could review the actual regional regulations that restrict the expansion of dense, sustainable housing.

← 34.  Research shows that inclusionary zoning initiatives can have a positive, though modest, affect on the number of housing units. In the United States, inclusionary zoning policies produced between 129 000 and 150 000 affordable units over a 30‑year period (Calavita and Mallach, 2010[131]). California produced at least 29 000 affordable units between 1999 and mid‑2006 (Sturtevant, 2016[134]). However, the success of such initiatives depends on whether governments and developers can establish viable financial schemes to close the funding gap created by providing homes at below the market price.

← 35.  In the Netherlands, rent controls cover around 90% of the rental sector. In Germany, regulated rents apply to all rental units except for new or modernised buildings (OECD, 2016[82]); rents cannot be increased by more than 20% over a three-year period; and landlords may only raise the rent once every 12 months (Fitzsimons, 2014[133]).

← 36.  Property owners reduced the rental supply of small multi-family housing by either living in the house they own or by converting them into more expensive, high-end rented accommodation. In the long run, changes as the housing supply raised prices, attracted higher-income residents and increased gentrification (Diamond, Mcquade and Qian, 2018[83]).

← 37.  There is a wide variation in the different combination of policy instruments across countries. Chile and Mexico spend more on grants to home buyers and Luxembourg, Poland, France, Spain, Canada, the Netherlands and Japan give more support to mortgage borrowers (OECD, 2016[85]).

← 38.  A ten percentage-point increase in the housing loan-to-value ratio, such as a reduced down-payment, is associated with a 2‑percentage point rise in the homeownership rate of middle-income households – and twice as much for young households (Andrews and Sánchez, 2011[84]).

← 39.  Regulating the minimum size of deposits may be beneficial for middle-income households in the long run, however, as restricting the expansion of mortgage credit can limit house price rises and improve affordability (Kohl, 2018[128]).

← 40.  Household debt-to-income ratios are significantly above 100% in most countries and exceed 200% in the Netherlands, Portugal and Spain. Households in the lower quintiles of the income distribution are particularly vulnerable, with debt-to-income ratios exceeding the conventional at-risk threshold value of 300% (Causa and Woloszko, forthcoming[68]).

← 41.  Fees are adjusted in line with household income up to a threshold of approximately 1.25 times the average wage.

← 42.  Public expenditure on ECEC as a share of GDP in the three countries are 60 to 100% higher than in the OECD on average, at 1.2 to 1.8% (the OECD average in 2015/16 was 0.76% of GDP, OECD (2018[108])).

← 43.  Extending cost coverage would likely be expensive. However, greater spending could be financed by efficiency improvements, lower spending on other areas of or broadening the contribution base in many countries. See OECD (2015[129]) for a range of options.

← 44.  Therapy using pharmaceutical drugs plays a key role in both the primary and secondary prevention of many diseases. Several studies have shown that financial barriers to accessing these necessary medicines are strongly correlated not only with poorer health outcomes, but also increased use and cost of other health services (Kesselheim et al., 2015[130]).

← 45.  However, when using the concept of catastrophic spending, the poor are more likely to face financial hardship than the middle class due to direct health care costs.

← 46.  Including the labour market participation of women and changing living arrangements in old age.

← 47.  LTC assessment criteria in Germany and Korea, for example, are more demanding than in Japan (Campbell, Ikegami and Gibson, 2010[120]).

← 48.  Within the confines of universalism, there are many ways to target or direct support where needs are greatest, so ensuring both fairness and value for money. This idea is thus known as “targeted universalism”.

← 49.  Cash benefits should be part of wider care plans that include basic training for the family member concerned, work reconciliation measures such as flexible work arrangements, and other forms of support to carers, including respite care.

← 50.  Among 25‑to‑34 year-olds, 44.5% had completed a tertiary degree in 2017, compared to 26.4% in 2000.

← 51.  Besides young adults in elementary occupations, service and sales people, crafts and trades workers and plant and machine operators have experienced the largest declines in the chances of reaching the middle-income group since the mid-1990s. Meanwhile, the chances have remained more or less stable for technicians and associated professionals and professionals. OECD calculations for 25‑to‑34 year-olds using data from the ECHP/EU-SILC and LIS data for 18 countries.

← 52.  The Strategy partners include five different ministries, the Sami Parliament, the Norwegian Association of Local and Regional Authorities, three employer associations, four trade union associations and one organisation representing civil society.