A better America

Lane Kenworthy, The Good Society
October 2017
Preliminary and incomplete

America could do better by changing some of its policies and institutions and by adding some new ones.1 Here’s what would help:

After outlining the details for each of these, I turn to how much it will cost and how to pay for it.


In a rich nation such as the United States, everyone should have health insurance. We also should do better at controlling healthcare costs; while we won’t go bankrupt spending 18% of our GDP on health, or even more, the fact that every other rich democratic country achieves equivalent or better health outcomes while spending far less suggests that we have considerable room for improvement.2 How can we achieve these two goals?

The most straightforward path would be to expand coverage through Medicare, Medicaid, and a “public option”: lower the age at which Americans can get Medicare, raise the income limit for Medicaid eligibility, and add a Medicare-like program that individuals and families can purchase on health insurance exchanges and that firms can purchase for their employees.3 Eventually a large portion of the population would be covered by these three programs. At that point they could be merged and coverage extended to the full population. The US would then have a single-payer healthcare system. This would achieve universal coverage, and the government, as the dominant insurer, would be in a strong position to control healthcare costs.

Canada’s experience suggests that this type of arrangement can function quite effectively. Every Canadian has health insurance, and as figure 1 indicates, over the past half century life expectancy has increased more rapidly than in the US despite a far smaller rise in healthcare expenditures.

Figure 1. Life expectancy by health expenditures
The data points are years, from 1970 to 2013. The lines are loess curves. Life expectancy: years at birth. Health expenditures: public plus private, as a share of GDP. Data source: OECD.

Such a system wouldn’t eliminate private insurers. There would be a market among the rich for insurance plans better than the one(s) offered by the government. And employers and individuals might choose to supplement the basic health insurance plan with an additional one, just as many elderly Americans with Medicare currently do.

Over time, government has gradually increased its role in promoting access to health insurance. The Veterans Administration (VA) was created in 1865 and significantly reformed in 1930 and 1994. In the 1940s and 1950s the federal government created and expanded a tax deduction for firms that contribute to health insurance for their employees. Medicare was created in 1965 and extended to cover prescription drugs in 2004. Medicaid too was created in 1965, and the share of the population it covers was expanded in the 1980s, in 1999 with the S-CHIP program and in 2010 via the Affordable Care Act (ACA). Figure 2 shows the rise in the share of Americans covered by Medicare and Medicaid since the mid-1960s. Together, these two programs cover about 40% of the US population. The 2010 ACA also requires that medium-size and large firms offer health insurance to their employees, it establishes an individual mandate to have health insurance, it provides subsidies for persons and families with modest incomes, it requires that health insurers allow people to remain on their parents’ plan through age 25, and it forbids insurers from denying insurance to persons with preexisting conditions.

Figure 2. Health insurance via Medicare and Medicaid
Share of the US population. Data source: Centers for Medicare and Medicaid Services, “Medicare and Medicaid Statistical Supplement,” tables 2.1 and 13.4.

Why not expand employer-based health insurance? America’s employer-centered health insurance system was a historical accident.4 It originated during World War II, when wage controls made it difficult for firms to offer higher pay in order to attract and retain good employees. Some decided to offer health insurance instead. After the war, encouraged by a new tax break, this practice proliferated, and it has remained in place ever since. But in a society where people switch jobs frequently, it makes little sense for insurance against a potentially major and very costly risk to be tied to one’s employer. Moreover, providing health insurance is expensive for firms, putting them at a disadvantage relative to small firms and foreign competitors. And it likely acts as a brake on wage increases.

Why does employer-based health insurance work well in some other countries, such as Germany and Japan? The reason is that if people quit or lose their job, they’re automatically switched into a government (“community”) health insurance plan. And the cost of health care is contained, so it’s less of a burden for employers. This happens in part because health insurance firms and funds aren’t for-profit, so they aren’t inserting additional costs into the system, and partly via cost controls set by centralized agreements between insurers and providers, with government stepping in if that fails.5

Do Americans like government health insurance? Most say they do. Figure 3 shows that about two-thirds of Americans think Medicare and Medicaid are working well for the groups they serve. In 2015, Gallup asked a representative sample of US adults “Are you satisfied or dissatisfied with how the healthcare system is working for you?” As figure 4 shows, satisfaction was higher among those getting their health insurance via the military, the Veterans Administration, Medicare, or Medicaid than among those getting it via an employer or purchasing it directly themselves.

Figure 3. Medicare and Medicaid are working well
Share of US adults. Questions: “Would you say the current Medicare program is working well for most seniors, or not?” “Would you say the current Medicaid program is working well for most low-income people covered by the program, or not?” Response options: working well, not working well, don’t know. The lines show the share responding working well, with don’t know responses excluded. Data source: Kaiser Family Foundation, “Medicare and Medicaid at 50,” 2015.

Figure 4. Satisfaction with how the healthcare system is working by insurance source
Share of US adults who say they are satisfied. The data are for 2015. Question: “Are you satisfied or dissatisfied with how the healthcare system is working for you?” Data source: Rebecca Riffkin, “Americans with Government Health Plans Most Satisfied,” Gallup, 2015, gallup.com.

Should government not only pay for health insurance and oversee it but also be the provider? That’s how countries such as the United Kingdom, Sweden, Finland, and some others do it, and it tends to work well. Indeed, the UK got top ranking in a recent assessment of healthcare quality in 11 affluent nations by the Commonwealth Fund.6 It’s extremely unlikely, however, that the United States will replace its existing array of private for-profit and nonprofit medical providers with a fully government-run physician and hospital system.

How much would a single-payer healthcare system cost, and where would the money come from? In 2015, the US spent $3.2 trillion, 18% of the country’s GDP, on health care. The government’s share is a little less than half of this total. The tax benefit to employers costs about $250 billion, Medicare $650 billion, Medicaid $560 billion, health care for veterans $65 billion, and health care for current military personnel and their families $40 billion.

Medicare and Medicaid limit the amount they will pay to healthcare providers, and they have relatively low administrative costs. As figure 5 shows, even though they’ve been covering more and more of the population (figure 3), the share of GDP spent on these two programs has been rising at about the same pace as the rest of the healthcare system. Their cost will continue to rise going forward, owing partly to population aging and expansion of Medicaid coverage and partly to the general rise in healthcare costs, but the projected increases are fairly small.7

Figure 5. Health expenditures
Share of GDP. Data source: Centers for Medicare and Medicaid Services, “National Health Expenditures by Type of Service and Source of Funds.”

A key obstacle facing the move to a single-payer system is that taxes would have to increase significantly in order to pay for it. But this isn’t insurmountable. A single-payer system likely would reduce total spending on health care. According to one estimate, adding coverage for the roughly 9% of Americans who now lack it and improving coverage for the 35% who currently are underinsured would increase costs by about 10%. But single-payer would reduce overall healthcare costs by approximately 18%: 7% from reduction in administrative costs, 3% from lower pharmaceutical prices, 3% from paying Medicare rates to healthcare providers, and 5% from improved service delivery (reduction in unnecessary services, inefficiently delivered services, missed prevention opportunities, and fraud).8 If correct, this estimate suggests a single-payer healthcare system would cost roughly 90% of the current spending total, or about 16% of GDP. That means government expenditures on health would rise by about 8% of GDP.

Of current health spending, 45% is by government (federal, state, and local). The other 55% is private: 27% by households, 20% by private firms, and 8% by other private sources.9 The cost of a single-payer system would need to come from taxes that replace these private expenditures. There are many possibilities, from a payroll tax paid by employers to an income tax and/or consumption tax on households. While the dollar figure will surely sound scary to some Americans, such a system won’t mean additional payments for health care; it will simply require different forms of payment (public instead of private).

So is single-payer the solution for the United States? In the long run, probably yes. In the short run, it may be more sensible to focus on making health insurance universal and making sure all Americans have insurance that is minimally adequate. The most straightforward way to do this is by expanding access to Medicaid and/or Medicare. According to one estimate, this would increase health care costs by approximately 10%, or about 1.75% of GDP.10


A 1993 law, the Family and Medical Leave Act, gives employees the right to 12 weeks of job-protected leave for the birth of a child or to care for a sick relative. But this only applies to companies with 50 or more employees. And there is no requirement that the leave be paid. Only 14% of American workers have employer-provided paid family leave.11 Consequently, many Americans in middle-and low-income households take little time off. That’s unfortunate, because outcomes for children tend to be best when they are with their parent(s) during the first year of life.12

In Sweden, parents of a newborn child have 13 months of job-protected paid leave, with the benefit level set at approximately 80% of earnings. Two of those months are “use it or lose it” for the father; if he doesn’t use them, the couple gets eleven months instead of thirteen. In addition, parents can take four months off per year to care for a sick child up to age twelve, paid at the same level as parental leave.13 As figure 6 shows, Sweden’s policy is a generous one, but not exceptionally so by the standards of other rich nations.

Figure 6. Paid parental leave
The data are for mothers. Includes both maternity and parental leave. Replacement rate is the share of wage or salary received. 2016. Data source: OECD, “Key Characteristics of Parental Leave Systems.” “Asl” is Australia; “Aus” is Austria.

The United States is the only affluent democratic country without a paid parental leave program. A few states — California, New Jersey, New York, and Rhode Island — along with Washington DC have enacted small-scale programs. Results from California’s, which has been in place since 2004, are encouraging.14

A new federal parental leave program for the US should provide a minimum of 26 weeks of paid leave per child, with an incentive for the father to take a portion of the leave. The leave should be job-protected. The replacement rate should be at least 50%. All workers meeting minimum work history requirements, including those in small firms and self-employed persons, should be eligible.15

Sweden’s policy costs about 0.75% of GDP per year.16 With a slightly less generous version and our larger per capita GDP, an American counterpart might cost around 0.5% a year.17


Many things affect children’s well-being and life chances. Money is one of them.18 An increase in family income of a mere $3,000 during a person’s first five years of life is associated with nearly 20% higher earnings later in life.19

Most other affluent countries have a universal “child benefit” or “child allowance.” In Canada, for instance, a family with two children under age six and an income below $30,000 receives an allowance of $10,000 (in US dollars). The amount is less for families with older children and/or higher income.20

The United States has a weaker version, the Child Tax Credit, which doles out a maximum of $1,000 a year per child. Families with no earnings don’t qualify. Nor do those who don’t file a federal income tax return. As a result, low-income households benefit far less than middle-income households.21 We also have an exemption on federal income taxes of $4,000 per child (“child deduction”). This too is worth little to those with low incomes, many of whom owe no federal income tax, and even for households in the middle of the income distribution it is of limited value.22

A team of researchers led by Luke Shaefer has offered a sensible proposal. The Child Tax Credit and child deduction would be replaced by a child allowance paid to all families with children. The amount would be $3,600 per year for younger children and $3,000 for older ones. It would be universal and unconditional — not contingent on employment or income or assets. It would be paid monthly via electronic transfer in order to ensure a high take-up rate. The money would be taxable, though that wouldn’t affect its level for many households with low or lower-middle income. This child allowance would reduce the child poverty rate in the United States by around 40%, and it would virtually eliminate extreme child poverty. The cost would be approximately $180 billion per year, or 1% of GDP. After subtracting the current cost of the Child Tax Credit and the child deduction (about $100 billion), the net additional cost would be around 0.5% of GDP.23


Unemployment insurance is a key policy mechanism for ensuring economic security. Our program was created in 1935 as part of the New Deal. It is paid for by the federal government, but states have considerable leeway in determining eligibility criteria and benefit levels.

Only about 40% of unemployed Americans qualify for compensation. Particularly vulnerable are persons who have low wages, work part-time or intermittently, move frequently from one employer to another, are self-employed, or interrupt employment for childbirth or family care. The average share of prior earnings replaced by unemployment benefits is just 45%.24 Two simple reforms would address these problems. The first is to federalize eligibility rules and benefit levels, as part of the problem is that some states are extremely stingy. The second is to adjust eligibility criteria to accommodate nonstandard workers and nonstandard employment, which are more common now than in the past and will become even more common going forward.25

We also should add a wage insurance component to the program. Some Americans who get laid off can’t find a job that pays as well and are forced to settle for less. For a year or two, wage insurance would fill half of the gap between the former pay and the new lower wage.26


The United States is the only rich nation without a public sickness insurance program.27 Though many large private-sector firms offer employees some paid sickness days and five states (California, Hawaii, New Jersey, New York, and Rhode Island, along with Washington DC) have a public program, one in three employed Americans gets zero days of paid sick leave.28

Sweden’s approach offers a useful model.29 A person with illness, disease, or injury that causes her to miss work receives 80% of her pay. The amount of the benefit is capped at about $28,000 per year.30 The benefit is taxed as ordinary income. Day 1 isn’t reimbursed. Days 2 through 14 are paid by the employer, and after that the money comes from a public sickness insurance fund. Self-employed persons are paid from the public fund. The payments can last up to a year (longer for a serious disease). A certificate from a doctor is required after seven days and a detailed medical exam after one month. Eligibility begins after three months of employment for sickness and disease and immediately for workplace injury.

We need not begin with a program nearly as generous as Sweden’s, which has typically cost about 1.5% of GDP. A program spending about a third of that amount, 0.5% of GDP, would be a good enough start.


Disability is pervasive and varied31:

“A baby is starved of oxygen during childbirth. A construction worker slices off a finger with a power saw. Another loses partial lung function after spending a year cleaning up dusty debris at Ground Zero. A retired professional football player forgets instructions he was given moments earlier. A nurse suffers back strain from helping an obese patient into bed. A young adult develops schizophrenia. An oncologist diagnoses an energy-depleted 55-year-old salesman with multiple myeloma. A soldier in Iraq suffers a traumatic brain injury when an explosive device detonates underneath a transport vehicle. A Cornell student is paralyzed for life by a prescription drug-induced stroke. Another suffers the same fate as the result of an act of drunken horseplay.

“Disability may be innate, as in the cases of those born with developmental deficiencies. It may be total, as in the case of the worst traumatic brain injuries, but more often it is partial. It may be temporary or permanent. It may lead to a shortened life span, but often it does not. It may occur on the job, but more often it happens away from work. Whatever its genesis and character, disability leaves the victim with a diminished capacity to work for a living. Through rehabilitation and retraining some can overcome the functional limitations engendered by their disabilities, but many cannot regain sufficient functioning to enter or reenter the workforce.”

About 20% of Americans are disabled. Approximately 30% will at some point in their work career experience a disability significant enough to cause them to miss 90 or more days.

One-third of Americans have private disability insurance (short-term and/or long-term) through their employer, mostly in large or middle-sized firms. A few purchase disability insurance on their own. The chief source of disability compensation for most working-aged Americans is three public programs: workers compensation, Social Security Disability Insurance (SSDI), and Supplemental Security Income (SSI).

Workers compensation covers about 85% of employed Americans. It pays out about $30 billion per year to people injured on the job, generally covering two-thirds or less of a worker’s earnings. The incidence of on-the-job injury requiring time off from work has decreased from about 3% of employees in the early 1990s to 1% as of 2010. As a result, workers compensation claims and payments have decreased.

Persons who become severely disabled and have paid Social Security taxes in five of the previous ten years may, after a five-month lag, receive Social Security Disability Insurance payments. After two years, they also receive Medicare to cover healthcare costs. The average SSDI payment is $15,000 a year. About 40% of applicants qualify for the benefit. Recipients are reviewed every two to five years to determine whether they remain eligible.

Around 10 million Americans, including former workers, spouses, and children, receive SSDI benefits. The share of the population getting SSDI payments has increased over the past generation, due to the rise in the share of people employed, an increase in the retirement age for Social Security eligibility (from 65 to 67), the aging of the population, and an expansion of eligibility criteria to include musculoskeletal maladies and mood disorders.

Disabled Americans who don’t qualify for SSDI, and who have assets of less than $2,000 excluding house and car, may be eligible to receive Supplement Security Income payments. These average $7,000 per year. The number of recipients is approximately 6 million. About one in three applicants are deemed eligible. Recipients also typically qualify for Medicaid health insurance.

Some people who are disabled temporarily don’t qualify for unemployment insurance, worker compensation, SSDI, SSI, or veterans compensation and don’t have private disability insurance. Five states provide temporary disability insurance for such circumstances. Because this group of states includes heavily-populated California and New York, more than a third of Americans have this protection. In many states, however, it is easy to fall through this crack in the system.

There are two major deficiencies in our support for disabled Americans. One is the lack of short-term disability insurance for many Americans with partial or temporary disabilities. The other is our limited commitment to vocational rehabilitation for disabled persons who might be able to return to work. Such efforts have a low success rate, even in countries that dedicate more resources than we do. Yet the signal these efforts send — about our commitment to genuine inclusion for disabled persons and about our support for employment — arguably justifies the cost.32

The United States spends about 1.5% of GDP on “incapacity” programs. Adding an additional 0.5% would put us closer to the average of 2.5% in other rich democratic nations.33


What to do about working-age households with no one employed has long been the thorniest question in American social policy.34 There is no optimal solution. If we are generous, some will cheat the system. If we are stingy, we cause avoidable suffering. Given this tradeoff, the best approach is a policy that vigorously promotes employment for those who are able, provides a decent minimum for those who aren’t, and deals on a case-by-case basis with those who can work but don’t.

This requires three modifications to what we have now. First, we should adjust our approach to caseworkers and the assistance they provide. In theory, caseworkers help Temporary Assistance for Needy Families (TANF) recipients find jobs, but in reality many are undertrained, overworked, and have limited means to provide real help.35 For some Americans at the low end of the labor market, adulthood is a series of transitions between part-time or full-time employment, off-the-books work, government benefits, romantic relationships, child rearing, drug or alcohol addiction, and time in jail.36 The best thing we can do is to provide help, support, cajoling, pushing, and the occasional threat. People who struggle to find a job after leaving school (whether at age 22, 18, or earlier) should immediately get individualized help.37 This may include temporary cash support, a push into a training program, and/or a push into counseling. Strugglers should be monitored as they move along in life. For this to be effective, we need caseworkers who are well trained, connected to local labor market needs, committed to their job, and not swamped with clients. They must be able to make realistic judgments about when clients can make it in the work force and when the best solution is simply to help them survive.

Second, government should act as an employer of last resort. (See the “Full Employment” section below.)

Third, TANF’s benefit amounts should be increased and eligibility criteria eased. Given the five-year lifetime limit on TANF receipt, a more generous benefit is unlikely to be a deterrent to employment. And in bad economic times, such as the 2008-09 recession and its aftermath, the five-year limit has proved too strict, causing needless hardship.38 We should allow more exemptions to this limit during economic downturns.

Food Stamps (SNAP) can be kept as is. It is effective, efficient, and widely appreciated.39 Current spending on TANF is about $30 billion per year. Doubling that amount would help to fund the needed changes. That would cost an additional 0.2% of GDP.


The poverty rate among elderly Americans has fallen steadily over the past half century, and the best available projections suggest that average incomes in the bottom 40% of elderly households will continue to rise in coming decades. Yet that rise is projected to be relatively slow, with incomes in old age falling farther and farther behind growth of the economy.40

Social Security benefits could be increased.41 A modest, gradual rise over time is appropriate as the economy grows, and it surely is affordable. However, this is only a partial solution. We need one or more of the other retirement income security pillars — personal savings, employer pensions, homeownership, earnings — to increase as well.

We could try to encourage more saving. But previous attempts, such as offering tax advantages (IRAs), have had little impact on the savings behavior of most ordinary Americans.42

We should shore up employment-based pensions. Rather than allow Americans to contribute to defined-contribution plans if they have a steady job and if their employer offers a plan and if they know about it and if they feel they can afford to put some of their earnings in it, we could make contributing the default option and make it available to everyone. Employers that have an existing plan could continue it, but they would have to automatically enroll all employees and deduct a portion of earnings unless the employee elects to opt out. Employees who lack access to an employer plan would be automatically enrolled in a new universal retirement fund, and those who lack an employer match would be eligible for matching contributions from the government.43

In the absence of federal government action along these lines, a few states have created their own programs. California’s Secure Choice Retirement Savings Program, the largest of these, requires firms with five or more employees to enroll them if it doesn’t offer a company-sponsored pension program. The program contributes 3% of each paycheck, unless an employee chooses to opt out. But there is no matching contribution from the employer or the state.44

We also could facilitate greater employment among the elderly. The employment rate among Americans age 65 and over dropped steadily in the 1960s, 1970s, and 1980s. Since the late 1990s it has slowly but steadily risen.45 Later retirement isn’t a good option for everyone, of course, especially those who have spent most of their working lives in stressful or physically taxing jobs. But for those who can manage it, it is doubly beneficial: it provides an additional source of income, and it allows people to delay receipt of Social Security, which in turn increases the benefit level they will receive.


Many ordinary Americans would like to live in a large city but can’t afford to. Cities are attractive for a variety of reasons: they are where many jobs are located, particularly analytical professional positions; they are diverse; they provide lots of eating and entertainment choices; and unlike a generation ago, they are relatively safe and clean. Cities also are economically productive: they concentrate lots of economic activity in a small space, and by bringing people together they generate multiplier effects. And cities are environmentally friendly: they use far fewer cars and less heat and electricity per person than do suburbs and rural areas.

But home prices and rents in some large cities — New York, Boston, Washington DC, San Francisco, San Jose, Los Angeles, San Diego, among others — exceed what many poor, working-class, and even middle-class Americans can afford. The chief cause is an inadequate supply of housing. When demand for something is high and supply is limited, the price tends to go up. In some instances, such as Manhattan and San Francisco, the inadequate supply of homes and rental units owes partly to physical constraints imposed by surrounding water, yet that can be overcome by additional vertical construction. The key obstacle is restrictions on new building stemming from zoning laws and historical preservations designations.46

We should loosen these restrictions. Local government is the ideal source of action. But where city councils and mayors are reluctant to act, state governments may have to step in. California’s did so in 2017 with passage of SB35, which stipulates that if local ordinances or decisions needlessly prevent or delay construction of new affordable housing, the state will step in and overrule the local authority.

Income in the bottom fifth of US households averages just $20,000 a year, so lower-income Americans need assistance with housing costs not only in large cities, but virtually everywhere.47 The federal government currently spends around $50 billion a year on low-income housing assistance.48 This assistance comes through a variety of programs. Since the 1930s the government has built public housing units, which are offered to low-income tenants at below-market rents. Though there has been little new public housing construction in recent decades, about one million such units remain across the country. Since the 1960s it has subsidized private construction of low-cost rental units and subsidized the rent that low-income tenants pay. Since 1986 it has provided a tax credit (the Low Income Housing Tax Credit, or LIHTC) to developers for construction or rehabilitation of rental housing in which at least 20% of tenants have incomes below half the area’s median income. And since 1974 the federal government has given “Section 8” housing vouchers to some low-income households who rent on the private market. Renters pay 30% of their income toward rent, and the voucher pays the difference between this amount and the rent amount (up to an allowable maximum).

These programs serve about five million households. Eligibility criteria have varied across the programs and over time within them. Roughly speaking, households with an income below 50% or sometimes 80% of the area median income tend to be eligible. Among eligible households, only one in four receives assistance from any of these programs. This isn’t due to lack of interest; about six million households are on waiting lists for a housing voucher and/or a public housing unit.

On average, these programs have enhanced access to housing, reduced over-crowding, improved housing quality, and increased residential mobility for their low-income recipients.49 Run-down, violence-plagued public housing projects such as Cabrini Green in Chicago have been the exception, not the rule. Research does suggest, though, that housing vouchers tend to boost housing quality more, and at lower cost, than public housing.50

We should expand housing assistance, via provision of a voucher, to the 15 million or so low-income Americans who are eligible for such assistance but don’t currently receive it. Doing so would cost about $75 billion a year.51 The federal government spends (forgoes) about $80 billion each year on the mortgage interest tax deduction. The aim of this program is to boost home ownership, but many other affluent nations have homeownership rates comparable to ours or higher without a tax incentive. Moreover, most of the mortgage interest deduction goes to households in the top fifth of incomes; few in the middle or below benefit from it.52 We could pay for the expansion of low-income housing assistance by ending this program. The additional cost to taxpayers would therefore be $0.


Universal high-quality publicly-funded early education for children aged 1 to 4 would facilitate work-family balance for Americans, and the best available evidence suggests it would enhance opportunity for children who grow up in less-advantaged homes.53

Why can’t we leave early education to the market? A good early education system will combine three features: accessibility, affordability, and quality. For Americans able and willing to pay a lot for childcare, our current market-based system typically delivers all three. But for those with low to moderate incomes, getting access to affordable care often means sacrificing quality.54 A universal system with public funding and some direct public provision could change this, ensuring good-quality care to everyone at an affordable price.

Government already pays for some early education: the federal government funds Head Start, some special education services, and tax breaks for childcare, and some state governments fund preschool for four-year-olds and subsidize childcare for poor families. Yet current funding is nowhere near sufficient to ensure that everyone has access to good-quality child care and preschool.

Should government not only pay for but also provide early education? Those who say yes contend that this is the only way to guarantee universal access to preschool and care that’s above an acceptable quality threshold. On the other hand, it isn’t necessary that government be the sole provider. Denmark and Sweden allow private providers, as long as they meet quality standards. In many districts across the US we allow private providers for publicly-funded K-12 schooling (charter schools), and we allow private doctors and hospitals to provide medical care for Medicare and Medicaid recipients. What’s the ideal mix? We don’t know. Maybe it’s 25% of kids in public early education centers, or perhaps it’s 75%. This depends largely on how many private providers can combine good quality with a reasonable rate of return.

Why not simply increase access for those with low incomes? The argument for universal early education is threefold. First, it isn’t just low-income parents who struggle to find good-quality care that’s affordable. Middle-class parents do too. Second, family structure and parents’ traits and behaviors are key sources of disadvantage, and they don’t overlap perfectly with family income. If we target low-income households, we’ll miss many children who need help. Third, development of cognitive and especially noncognitive skills is aided by peer interaction. Children from less advantaged homes gain by mixing with kids from middle-class homes, which doesn’t happen in a program that exclusively serves the poor.55

Should we encourage parents to put their kids in out-of-home early education immediately after birth? Probably not. As noted in the “Paid Parental Leave” section, research suggests children tend to fare best staying with a parent during the first year of life.

Some of the revenue needed to fund early education can come from user fees. Early education is different from police protection and health care, the kinds of services that almost no one opts to go without. Even if good early education programs were readily available, some families would choose not to use them because they prefer to provide stay-at-home parental care for their young children. And of course some American adults have no children. This argues for having parents who do want to use early education pay something — even parents with low incomes. Here too the Nordic approach is sensible; in Denmark and Sweden programs charge on a sliding scale, with the fee rising in proportion to family income, but never going above 10%.

The bill to taxpayers will depend on specific details, but a rough estimate is 1% of GDP, or $180 billion, per year.56 There are two ways to reach this number. First, suppose 75% of children aged one to four end up enrolled in early education. That’s 12 million children. If we spend $12,500 per child, the same as for K-12 schools,57 total expenditures would be around $150 billion. We’ll want the teacher-child ratio for early education to be better than for K-12, which will increase the cost a bit.58 Second, public expenditure on early education in Denmark and Sweden is about 1.5% of GDP.59 We’re likely to end up with more private provision and we have a larger per capita GDP, so 1% of our GDP might well be sufficient to create a system that approximates theirs in quality and accessibility. Government (federal, state, and local) currently spends about $30 billion per year on child care and preschool,60 so additional spending would amount to around $150 billion, or 0.8% of GDP.

When someone suggests borrowing a policy or institution from the Nordic countries, skeptics point out that these countries are very different from America — they’re small, they’re more ethnically and racially homogenous, and their cultures and histories are quite distinct from ours. What works there, in other words, won’t necessarily work here. However, this doesn’t justify blanket skepticism about policy borrowing. We need to consider the particulars of the policy in question. It’s not clear why a system of publicly-funded early education centers (schools) can function effectively only in a small homogenous country. France does this, even though it’s a pretty large nation. Belgium does too, despite its diversity.61 And the United States does a reasonably good job with kindergartens and elementary schools. Though education experts and ordinary Americans routinely profess dissatisfaction with the country’s K-12 public schools, inequality in capabilities expands when children aren’t in school (before kindergarten and during summers), while K-12 schools hold it at bay. American schools could be better, to be sure, but for less advantaged children they are, even in their current condition, far more helpful than the likely alternative.

Another objection is that we don’t know how large the impact of early education will be in boosting the capabilities of children from less advantaged families. While the expectation of a sizable effect is compelling, and we have supportive evidence from K-12 schooling, from three high-quality early education programs, and from cross-country comparison, that evidence is limited.62 Yet because equalizing opportunity is such a prized goal, even a modest improvement would make early education worthwhile. And regardless of its impact on opportunity, it will be of considerable benefit in helping parents balance work and family.

If parents have access to affordable good-quality childcare and preschool, might they be less likely to stay together or get married in the first place?63 That’s conceivable, but the historical and comparative evidence suggests reason for skepticism. Enrollment in elementary and secondary schools grew steadily in the United States from the late 1800s until around 1960, but it wasn’t until the 1960s, after the rise in school enrollment slowed sharply, that rates of divorce and out-of-wedlock birth shot up.64 And more children grow up with both parents in France and Sweden, each of which has a universal early education system, than in the United States.65

Should we worry about rent-seeking if a substantial amount of early education is publicly provided? Public-sector employees may be able to get above-market pay and benefits, increasing the cost to taxpayers.66 The evidence on this is mixed.67 Early education proponents draw a parallel with the military, police protection, fire fighting, medical care, K-12 schooling, and other public services. Most Americans judge it worthwhile to pay for these services even if there is some rent-seeking.

Finally, why not just give the money to parents and let them choose whether to use it on early education or on something else? The argument against doing so is that if early education has individual and social benefits, it makes sense to require that the money be used for that and only that. The same is true of safety (military, police), infrastructure (roads, bridges), health insurance (Medicare, Medicaid), and K-12 schooling, among others. It’s worth emphasizing that no one would be forced to enroll their children in early education; parents who prefer to stay home with their children during the first five years would still be able to do so.


Of the two-thirds of Americans who don’t get a four-year college degree, some enter the labor market directly, others get some vocational training in high school or community college, and others complete a certificate program of some sort. Some among these two-thirds fare reasonably well, but the experience of other affluent nations suggests that a more robust approach to vocational education could help.

One option is an “apprenticeship” program that students would begin around age 16 and that combines classroom and on-the-job training. The best such programs run for three or four years and are tightly integrated with employers and employer organizations to ensure that the skills being produced are needed ones rather than simply ones schools feel competent to provide.68

Federal government funding can be put to two particularly useful purposes here. One is to encourage and subsidize local or regional consortia of high schools, community colleges, universities, and employer associations with the aim of building career pathways.69 The other is a tax credit for employers who create apprenticeships. Harry Holzer, for instance, recommends a credit of $1,000 per apprenticeship with a goal of one million new apprenticeships.70 The total cost these efforts would likely be no more than 0.1% of GDP.


America’s colleges and universities are among the country’s greatest achievements. Taken as a whole, they’ve long been, and remain, the best in the world. Yet too few Americans from less-advantaged families enroll in college, too few of those who enroll end up getting a four-year degree, and our colleges and universities probably aren’t doing as well as they should in educating students. These deficiencies have multiple causes, only one of which is the cost of college.71 And yet Christopher Jencks is correct in pointing out that “Making college a lot more affordable is a challenge governments know how to meet, while making students learn a lot more is a challenge we do not currently know how to meet. Under those circumstances, starting with affordability is probably the best bet.”72

What is the best way to make college more affordable? If we want to reduce the cost mainly for students from lower-income families, we ought to focus on room and board, because many such students already receive enough financial aid and grant money to cover most or all of the cost of tuition.73 However, many American high schoolers and parents aren’t aware of this, and the process of applying for school-based financial aid and federal grants can be complicated. Making two-year and four-year public colleges tuition free for in-state students would send a clear, simple message that college is affordable for all Americans.74 Another potential benefit of zero tuition at public colleges is that it could increase pressure on private colleges to lower their prices.

A common objection is that eliminating tuition is inefficient, in that students whose families can afford to pay the tuition now wouldn’t have to. But this is true of all universal transfers and services. The key question isn’t whether there is waste. It’s whether the benefits of the program outweigh the costs. With “free college,” it seems likely they would.

Another concern has to do with the potential for funding reductions over time. The United Kingdom’s experience illustrates the problem. Prior to 1998, public universities in the UK were tuition-free. But in the 1980s and 1990s the government proved unwilling to increase funding in order to keep up with rising enrollments. Eventually it chose instead to cap the number of students colleges could admit. In 1998 the zero tuition policy was ended.75 It’s impossible to know whether a similar dynamic would play out in the US context, but this is an important cautionary tale.

What would the price tag for tuition-free public college be? About 12 million Americans attend in-state public colleges every year, a majority of them in community colleges. Average tuition is $6,000.76 So the total cost would be approximately $75 billion, or about 0.4% of GDP.

The cost of room and board averages about $10,000 for US college students, and tuition-free college won’t do away with these costs. Moreover, many Americans will continue to want to attend private colleges, which won’t be tuition-free and may in fact continue to get more expensive.77 Consequently, many American students will continue to have to take out loans. (The same is true in Sweden, where students pay no tuition and yet the average college graduate owes about $20,000 in student loans.78) In the late 1990s we created a program that allows income-based repayment of student loans: the lower the student’s income after college, the smaller the portion of their loan debt they are required to pay back. This program was enhanced in 2007 and 2010, and it could be expanded further. We also could lengthen the loan repayment period to 20 or perhaps even 30 years.79 And it would help to automatically enroll college students in this program, rather than requiring them to find out about it on their own and then wade through a complicated application process in order to utilize it.

The total cost of these two efforts to increase college affordability would be around 0.5% of GDP.


Since the late 1960s, affirmative action programs for university admissions and employment have promoted opportunity for women and for members of racial and ethnic minority groups.80 Not surprisingly, as these programs have been cut back or ended in recent decades, some of that progress has been reversed.81 Affirmative action should continue, but with family background as the focal criterion.82


Employment is front and center in the American ethos. Self-sufficiency and self-realization via paid work are at the core of the “Protestant ethic” that shaped the country in its early years and the “American dream” that has animated it since the mid-1800s.

Employment isn’t always a good thing. The need for a paycheck can trap people in careers that divert them from more productive or rewarding pursuits. Work can be physically or emotionally stressful. It can be monotonous, boring, alienating. Some jobs require a degree of indifference, meanness, or dishonesty toward customers or subordinates that eats away at one’s humanity. And work can interfere with family life.

Yet employment has significant virtues. It imposes regularity and discipline on people’s lives. It can be a source of mental stimulation. It helps to fulfill the widespread desire to contribute to, and be integrated in, the larger society. It shapes identity and can boost self-esteem. With neighborhood and family ties weakening, the office or factory can be a key site of social interaction. Lack of employment tends to be associated with feelings of social exclusion, discouragement, boredom, and unhappiness.

We also need a significant majority of people in paid work to help fund government programs. Given our commitment to pensions for their elderly, health care for most, and assorted other services and transfers, there is a need for additional government revenue. Some can come from raising tax rates, but increasing the share of the population in paid work can help. It provides an increase in tax revenues without requiring an increase in tax rates. High employment eases the fiscal crunch another way too, by reducing the number of people fully or heavily reliant on government benefits.

From the mid-1940s to 2000, the employment rate among working-age Americans rose steadily, from 60% to nearly 78%. But since 2000 we’ve moved in the opposite direction. Our employment rate currently stands at about 74%. This is partly a product of the deep 2008-09 recession. But sluggish employment growth began before the crash. And in the meantime many other rich nations have increased employment sharply, suggesting the problem isn’t something endemic to modern affluent societies.83

There are a number of things we can do the address this. One is better family-friendly policies, particularly child care and preschool, paid parental leave, and sickness insurance, as described above.

We should strive to improve schooling, from early education to K-12 to apprenticeships to college and beyond.

We should commit more resources to active labor market policy. Sweden and Denmark have used retraining and job placement assistance to help improve the efficiency of the private-sector labor market and public employment to increase demand for labor.84 These two countries have committed 1.5% to 2% of GDP to such programs.85 Swedish firms must notify their local board in advance when employees are to be laid off and when they have job openings that have lasted more than ten days. Workers who are displaced or who leave their job by choice can receive subsidized training through the employment service. Officials in local labor market boards keep in close communication with firms and with officials in other areas regarding trends in skill needs. The training programs are full-time and range in duration from two weeks to more than a year. The service then helps to place workers in new positions. If necessary, an employer subsidy may be used to encourage a private-sector employer to hire, or a public-sector job may be created. Denmark increased pursuit of active labor market programs in the mid-1990s, with apparently beneficial effects. A recent meta-analysis of research on such programs concludes that they tend to have a beneficial effect on medium-term employment outcomes.86

It helps if monetary policy authorities prioritize labor market tightness. The Federal Reserve, like independent central banks in other rich nations, is charged with maintaining both price stability and low unemployment. Its choices about which to prioritize, particularly during periods in which the unemployment rate is low and there are worries about the potential for a jump in inflation, impacts not just the business cycle but also the long-run employment rate. The longer the labor market can remain tight, the stronger the pressure on employers to increase wages and salaries, which tends to attract more people into paid work. And the boost to wages is a good thing apart from its effect on employment rates.87

Government should serve as employer of last resort. “Make-work” has a mixed history in the United States. It played a prominent role in the 1930s, and subsequent smaller-scale programs have boosted employment rates of low-end workers.88 These programs are often criticized because they have not tended to improve long-term labor market success of beneficiaries. But we should think of make-work not as a route to “real” employment, but rather as a worthwhile expense. If we believe people who are able and willing to work should be employed, the fact that make-work might not provide a ladder to a good job shouldn’t discourage us.

That doesn’t mean we should guarantee everyone a job in the place where they currently live. Doing that would prevent sensible shrinkage of some towns or cities that are no longer economically viable. What, then, can we do to boost employment in such locations? Investment in infrastructure can help, and is a good thing to do in its own right (see below). Another possibility is a temporary subsidy for wages in new private-sector jobs. A version of this proved effective in the aftermath of the deep 2008-09 recession.


The federal minimum wage in the United States is low, and it has been flat for half a century, as figure 7 shows. We should increase it to around $12 per hour and index it to inflation. States and localities with thriving economies and/or a higher cost of living could set their own minimum wage at a higher level, as many currently do.

Figure 7. Minimum wage
Federal minimum wage. 2015 dollars; inflation adjustment is via the CPI-U-RS. Data source: Economic Policy Institute, stateofworkingamerica.org/data.

The chief worry about increasing the minimum wage is that it will reduce employment. The best available evidence, however, suggests that modest increases in the statutory minimum in the past have not reduced employment. The best test, because it is closest to an experimental design, is a “differences in differences” approach.89 The fact that many of the US states have set minimum wages higher than the federal minimum, in varying degrees and at different times, is helpful for analytical purposes. In the early 1990s David Card and Alan Krueger compared changes in employment in fast food restaurants on either side of the New Jersey-Pennsylvania border after one state increased its minimum wage while the other didn’t. Arindrajit Dube and colleagues pursued this strategy for every pair of adjacent counties straddling state borders in which one increased its minimum wage between 1990 and 2006. They, like Card and Krueger, found no adverse employment effect of minimum wage increases.90 This suggests reason for optimism that raising the wage floor to $10 an hour, and perhaps to $12, is likely to cause little or no employment decline.

The real question is whether the federal minimum wage should be even higher. California, New York state, and the city of Seattle, among others, have passed legislation to raise their statutory minimum to $15 an hour by the early 2020s. Some argue that we should we do this for the nation as a whole.91 That would be a big increase, and it would affect a lot of Americans. Half of employed Americans earn less than $36,500 a year, and a person working full-time year-round at a $15 minimum wage would have annual earnings of $30,000.92

As figure 7 suggests, we have no prior historical experience with a federal minimum wage anywhere close to $15 an hour. Nor is the cross-country experience of much help. As figure 8 shows, among the 14 rich democratic nations that have a statutory minimum, none has gotten above the equivalent of $11 per hour in US dollars.

Figure 8. Minimum wage
2015 US dollars. Currencies converted using purchasing power parities. Austria, Denmark, Finland, Italy, Norway, Sweden, and Switzerland don’t have a statutory minimum wage. Data source: OECD. “Asl” is Australia.

Even some prominent advocates of a significantly higher wage floor fear that $15 an hour might cause substantial job loss, particularly in less-affluent parts of the country.93 Given this concern, coupled with the lack of empirical evidence on which to base a policy conclusion, the wise approach probably is to increase the federal minimum to something below $15 per hour, index it to inflation, monitor its impact, and then adjust as needed.94


The Earned Income Tax Credit (EITC) is a very effective program, encouraging employment while boosting the incomes of households who struggle in the labor market.95 The EITC subsidizes earnings by as much as 45%, providing up to $6,300 (for a household with three or more children). It is paid to households rather than to individuals, and the money comes in a lump-sum once a year.96 Households with at least one employed adult and earnings below $54,000 are eligible. The credit functions like a cash benefit; if it amounts to more than the household owes in federal income taxes, the household receives the difference as a cash refund.

As figure 9 shows, the amount of the subsidy increases with earnings up to a certain level, then plateaus, and then decreases with earnings.

Figure 9. Earned Income Tax Credit benefit structure
The benefit levels shown are for 2017. Source: Tax Policy Center, “Earned Income Tax Credit Parameters.”

The average amount recipient households get is $2,300 per year. The benefit level increased sharply between 1987 and 1996. Since then it has been flat. Nearly one in four Americans receives the EITC. This share rose significantly between the late 1980s and the mid-1990s and again in the 2000s, a result of changes in eligibility criteria, increases in the benefit amount, and stagnant wage levels for Americans on the lower rungs of the wage ladder.97

The United States and the United Kingdom were the first countries to introduce an EITC-type program, both in the 1970s. In recent decades many other rich longstanding democratic countries have adopted some version of it, including Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Ireland, Japan, the Netherlands, New Zealand, Portugal, South Korea, and Sweden. A number of the US states and a few cities have their own EITC; many are small, but some supplement the national EITC by as much as 75%.98

We can improve on the existing federal Earned Income Tax Credit in four respects.

First, the EITC is far too small for Americans who don’t have any children, a figure 9 makes clear. A household with one child can receive up to $3,400, but the maximum for a childless household is just $500. Childless households account for 25% of EITC recipients, but they receive only 5% of total EITC payments. The credit thus creates little employment incentive for childless adults, and it provides very little income support if they are employed.

Second, the fact that the size of the EITC credit depends on household earnings (actually pretax income) and that it has a phase-out range creates some employment disincentives for a potential second earner in a household whose earnings are likely to be relatively low.

A way to address these two problems is to give the EITC to employed persons, rather than households, and to give it to all or most such households, rather than only those with low earnings. Sweden has an EITC, first introduced in 2007, that works like this, as figure 10 shows. (Sweden’s EITC is small for earnings below $35,000 because its very strong unions mean few full-time workers have earnings below that amount.)

Figure 10. Swedish and US Earned Income Tax Credit benefit structures
The benefit levels shown are for 2010, in US dollars. United States: single unmarried adult with one child. Sweden: one earner (not contingent on the presence or number of children). PPP conversion: 1 US dollar = 9 Swedish kroner. Data sources: Karin Edmark et al, “Evaluation of the Swedish Earned Income Tax Credit,” Working Paper 2012:1, Institute for Labor Market Policy Evaluation, 2012, figure 1; Tax Policy Center.

Third, we could increase the amount of the EITC credit. The proposed EITC might look something like the following:

Figure 11. EITC benefit structure: actual and proposed
For discussion of the proposed benefit levels, see the text. The actual benefit levels shown are for 2017. Source: Tax Policy Center, “Earned Income Tax Credit Parameters.”

Fourth, the EITC could help to compensate for stagnant pay, one of the signature weaknesses of the US economy since the late 1970s.99 To ensure that incomes rise over time as the economy grows, we could index the EITC to GDP per capita, rather than to inflation. This won’t compensate fully for stagnant pay; a typical middle-class American earns about $30,000 a year, so if the EITC rises in line with the economy but earnings don’t, income (earnings plus EITC) growth will lag behind growth of the economy. It’s a partial remedy, not a full solution. But it will help.

How much would an expanded federal EITC cost? A fully universal version would give it to all 145 million employed Americans. If the credit averages $3,500 per person, the cost would be approximately 2.8% of GDP. (Sweden’s EITC costs 2.4% of GDP.) That’s a sizable increase over current expenditure on the program, which is 0.3% of GDP.100 The cost could be reduced by tapering the credit for those at the top of the distribution — the top tenth or the top fifth. The added cost might then be around 2.25% of GDP.

There are two main objections to an expanded EITC as a centerpiece of strategies to boost incomes. First, an employment-conditional earnings subsidy might cause wage levels to fall. In the presence of the subsidy, employers may offer a lower wage than they otherwise would, and workers may be willing to accept a lower wage. Also, the subsidy may increase the supply of less-educated people seeking jobs, and without an increase in employer demand for such workers, this rise in supply is likely to push wages down. Existing studies suggest that the EITC may indeed reduce wages somewhat, but the evidence is thin and the effect is likely fairly small.101 The best way to address this danger is with a moderate to high wage floor.102

Second, some will ask why taxpayers rather than employers should bear the cost of ensuring that household incomes rise. This is an understandable sentiment. But consider how we think about health insurance, pensions, unemployment insurance, and sickness insurance. Like income, these contribute to material well-being. In all affluent nations, including the United States, they are financed at least partly by taxes. Few object to the fact that firms aren’t the sole funders.


Public spaces and services matter directly for people’s lived experience. Think of roads, bridges, stoplights, enforcement of speed limits, air traffic control, sidewalks, museums, parks, sports fields, forests, campgrounds, beaches, oceans, lakes, swimming pools, zoos, phone lines, broadband, the internet, public television and radio programming, subsidization of free private TV and radio networks, libraries, festivals, and more.

Infrastructure also underpins a successful economy. America’s firms, and citizens, face significant hurdles and risks due to our failure to maintain and improve our roads, bridges, plane and rail systems, city layouts, broadband networks, and water systems. In 2013, the American Society of Civil Engineers reported that one-third of US roads are in poor or mediocre condition. According to the Federal Highway Administration, one-quarter of America’s bridges are deficient or functionally obsolete. The cost of traffic congestion in fuel and lost time is estimated to be nearly 1% of GDP. Delayed and canceled plane flights cost another 0.25% of GDP. We have an efficient freight-rail system for transporting products, but high-speed rail to move people around is nonexistent. The California drought in the mid-2010s exposed, once again, the inadequacy of our water supply systems. Four thousand dams are in need of repair. And as of 2015, 15% of Americans reported not using the internet and one-third lacked access to a high-speed connection.103

This isn’t to say that America’s infrastructure is worse than it used to be. That notion is based largely on anecdote.104 Nor is it to suggest that our infrastructure is far behind that of other affluent nations. We do lag behind some of them, according to the most recent assessment by the World Economic Forum, but the gap isn’t enormous.105 The point, rather, is that our infrastructure isn’t as good as it could and should be.

The needed amounts of money aren’t huge. One proposal, from a progressive think tank, estimates that to bring our roads, bridges, mass transit, rail, ports, airports, inland waterways, drinking water, wastewater, and energy infrastructure up to par would require additional expenditures by the federal government of about 0.5% of GDP per year.106

Investment in infrastructure doesn’t only grease the wheels of the economy. It also increases employment.107

Boosting employment is helpful in the aggregate, but it’s also vital, in the contemporary era, to the pursuit of geographical fairness. As production of food and goods has become steadily more automated and as more of it has moved abroad, smaller cities and towns across the United States have struggled economically. Places that suffer a sudden loss of a major employer experience particularly acute economic and social pain. Large job losses tend to have ripple effects, as unemployed households reduce their spending and thereby reduce employment at retail stores, restaurants, and other potential sources of substitute employment.

According to one study, counties with fewer than 100,000 residents accounted for 32% of the new businesses created during the 1992-96 economic recovery, but just 15% during in the 2002-06 recovery and 0% during the 2010-14 recovery. The pattern was similar for jobs: in the 1992-96 recovery, counties with population below 100,000 got 27% of the net increase in the nation’s jobs, while in 2010-14 they got just 9%.108 Worry about jobs was, it appears, one of the key contributors to the election of Donald Trump as president in 2016. One analysis found that across counties the strongest predictor of a shift in voting from Obama in 2012 to Trump in 2016 was a low employment rate.109 Another study found an especially strong predictor of the swing from Obama to Trump was the share of jobs that are “routine” — those in manufacturing, sales, clerical work and related occupations that are easier to automate or send offshore.110

There is no simple long-term remedy for small cities and towns, assuming recent patterns continue. But the building and repair of infrastructure can certainly help in the short-run.111 Given that a good infrastructure is directly beneficial for quality of life, and given that it can kickstart economic development, it tends to be money well spent.

In other rich democratic countries, the law requires that employers give their employees between 10 and 38 paid vacation days and holidays. The average in these nations is 27 days.112 In the United States, the comparable number is zero. Most public employees nevertheless get some, and 77% of private-sector employers offer some to their workers. Yet some employees get none, and the average number of paid days off for those who get any is just 18.113

We should make the provision of paid vacation days and holidays mandatory. And it would make sense to increase the number of days to, say, 25. That would put us near the average in comparable nations.


The additional expenditures needed to fund these various programs would total around 10% of GDP. Here is a breakdown. Details for each area or policy are provided above.

  • 2.25% of GDP: Earned Income Tax Credit
  • 1.75%: Health insurance
  • 1.0%: Full employment
  • 0.8%: Early education
  • 0.5%: Paid parental leave
  • 0.5%: Child allowance
  • 0.5%: Sickness insurance
  • 0.5%: Disability assistance
  • 0.5%: College
  • 0.5%: Infrastructure and public spaces
  • 0.25%: Pensions
  • 0.2%: Unemployment insurance and wage insurance
  • 0.2%: Social assistance
  • 0.1%: Apprenticeships
  • 0%: Housing assistance
  • 0%: Minimum wage
  • 0%: Affirmative action
  • 0%: Paid vacation days and holidays


Increasing tax revenues by 10% of GDP would be a significant change for the United States, but it wouldn’t be unprecedented. During the course of the twentieth century, revenues’ share of America’s GDP rose by about 25 percentage points. And an increase of 10 percentage points would put the US merely in the middle of the pack — not at the top — among the world’s rich countries.114

What would be the best way to do this? Broadly speaking, there are two options: “soak the rich” or “spread the burden.” The world’s rich nations tend to do the latter. They have a relatively proportional tax system, with everyone paying roughly the same share of their pretax income in taxes.115 They do this instead of soaking the rich — taxing the rich at much higher rates than everyone else — for two reasons. One is to minimize tax resistance by the rich. The other is the need to go where the money is; even if the rich have very high incomes, there aren’t that many of them, so in order to generate a lot of revenue it’s usually necessary to spread the tax burden up and down the income ladder.

America too has a relatively flat tax system. Figure 12 shows average effective tax rates in the United States at various points along the pretax income distribution (hollow circles). An “effective tax rate” is calculated as taxes paid divided by pretax income. These calculations include all types of taxes at all levels of government. The effective tax rates paid by Americans are fairly similar up and down the income ladder.

However, the distribution of pretax income is quite unequal. Households at the top get a much larger portion of the income than those in the middle or bottom.116 As a result, the distribution of tax payments is also very unequal, as we see in figure 12 (dark circles). Households in the top quintile pay about 65% of all tax dollars, the middle fifth pay about 10%, and the bottom fifth pay 2%. Each is paying a similar percentage of their income in taxes, but because the affluent have so much of the income, they end up paying a lot of the tax dollars.

Figure 12. Effective tax rates and shares of tax payments by income quintile
Includes all types of taxes (personal and corporate income, payroll, property, sales, excise, estate, other) at all levels of government (federal, state, local). The data are for 2016. Effective tax rates: taxes paid as a share of pretax income. Data source: Institute on Taxation and Economic Policy (ITEP), “Who Pays Taxes in America: 2016.”

Suppose we were to increase taxes for everyone, keeping the distribution of tax payments exactly the same as it is now while increasing revenues by 10% of GDP. What would that change look like for households at various points along the income distribution? Households in lowest fifth of incomes would account for about 2% of these added revenues, households in the middle around 10%, and households in the top quintile 65%. In dollar terms, households in the bottom fifth of incomes would pay, on average, about $1,400 more per year, those in the lower-middle fifth $3,600, those in the middle fifth $7,000, those in the upper-middle fifth $13,100, and those in the top fifth $46,200 more.117

As a presidential candidate in 2008, Barack Obama pledged to not increase taxes for households in the bottom 95% of incomes. Hillary Clinton, the Democratic nominee in 2016, made the same pledge. In the contemporary US context, there is some sense in focusing on the top in the search for more revenue. The chief rationale for progressive taxation is that those with more income can afford to pay a larger share of that income than those with less.118 The incomes of Americans in the middle and below have risen slowly over the past few decades. Meanwhile, incomes at the top have soared, so it’s reasonable to ask them to pay a larger share of those incomes.

However, there’s a limit to how much additional tax revenue we can get from those at the top. Figure 13 shows the effective tax rate on the top 5% of households going back to 1960. We have three estimates of this tax rate (two of the three include only federal taxes, not state and local). The dot for the year 2016 indicates what the effective tax rate on this group would need to have been in that year in order to increase tax revenues by 10% of GDP.119 It’s a very high rate, and one far above the actual rate at any point in the past half century. This seems neither desirable nor likely to find favor among policy makers.

Figure 13. Effective tax rate on the top 5% of incomes
Effective tax rate: tax payments as a share of pretax income. The chart has three estimates of the actual rate. The gray lines are for federal taxes. The black line is for all taxes (federal, state, and local). Data source for the top gray line: Thomas Piketty and Emmanuel Saez, data set for “How Progressive Is the U.S. Federal Tax System?,” Journal of Economic Perspectives, 2007, elsa.berkeley.edu/~saez. Data source for the lower gray line: Congressional Budget Office, “The Distribution of Household Income and Federal Taxes, 2011,” data set, alternative income definition, worksheet 13. Data source for the black line: Institute on Taxation and Economic Policy (ITEP), “Who Pays Taxes in America,” various years. Calculation of the rate needed to increase tax revenues by 10% of GDP is as follows: Get the total pretax income of the top 5% of households by multiplying this group’s average pretax income (from ITEP) by its number of households (from the Census Bureau). Then divide 10% of GDP by the group’s total pretax income.

What, then, should we do to increase government revenues by 10% of GDP? A multipronged approach might work. Here is one possibility:

  • 5.0%: add a national consumption tax (VAT) at a rate of 12%, with limited deductions or a small flat rebate
  • 1.4%: improve collection of unpaid taxes and reduce use of tax havens
  • 1.0%: return to the 2000 (pre-Bush) federal income tax rates for taxpayers with incomes below $450,000
  • 0.7%: increase the effective tax rate for the top 1% by 6 percentage points (from 34% to 40%)
  • 0.7%: add a carbon tax
  • 0.5%: add a financial transactions tax of 0.5% on trades
  • 0.3%: increase the payroll tax by 1 percentage point
  • 0.2%: increase the cap on the Social Security payroll tax so the tax covers 90% of total earnings, as it did in the early 1980s
  • 0.2%: end the real estate tax credit

Begin with a national consumption tax. As figure 14 shows, the US raises the least revenue from consumption taxes of any rich nation. Currently we collect only about 5% of GDP in consumption taxes, almost entirely at the state and local levels. Most other affluent countries collect 10% or more.120 A value-added tax (VAT) at a rate of 12%, with limited deductions, would likely bring in about 5% of GDP in revenue.121

Figure 14. Income tax revenues, payroll tax revenues, and consumption tax revenues
Share of GDP. The data are for 2013. The countries are ordered according to total tax revenues as a share of GDP. “Payroll” includes both payroll taxes proper and social security contributions. For France, the symbol for income taxes isn’t visible because income and consumption taxes each account for the same share of GDP. Data source: OECD. “Asl” is Australia; “Aus” is Austria.

Because of its regressivity, the idea of a large consumption tax has yet to be embraced by America’s left.122 The degree of regressivity can be lessened by exempting more items from the tax123; but the greater the exemptions, the less revenue the tax will bring in. A better strategy might be to offset the regressivity of a new consumption tax with other changes to the tax system.

The right tends to object to a VAT for fear it will become a “money machine” — a tax that can be steadily increased over time. But this fear is based on a misreading of the experience of other rich nations. Some countries have decreased their VAT rate, some have held it constant, and most of those that have increased it did so mainly in the 1970s and early 1980s, when high inflation made such increases less noticeable.124 Some argue that tax increases in rich countries since the 1960s have come mainly via VAT increases, but they’ve in fact come as much or more via increases in income and payroll taxes.125

By improving collection of unpaid taxes and reducing the use of tax havens, we could raise, as a conservative estimate, 1.4% of GDP.126

We could return to the pre-Bush income tax rates for taxpayers with incomes below $450,000. (This was done in 2013 for incomes above that amount.) Doing so would increase revenues by about 1% of GDP.127

We could raise income tax rates for those in the top 1% a bit more.128 This might entail increasing the tax rate on personal income or capital income, or both. The effective tax rate on the top 1% currently is around 34%.129 An increase of 6 percentage points, to a 40% effective rate, would hardly be confiscatory. Increasing the effective tax rate for this group by 6 percentage points would generate about 0.7% of GDP.130 A common worry is that raising taxes on the rich will cause them to flee, or at least to park their money elsewhere. The best available evidence suggests that high earners are indeed responsive to changes in tax rates, but that the magnitude of the effect is likely quite small.131

A carbon tax could generate about 0.7% of GDP in revenues. The United States arguably should have a carbon tax anyway, in order to shift resources away from activities that contribute to climate change.132

A modest tax on financial transactions, such as purchases of stock shares, would bring in about 0.5% of GDP. Opponents warn that this might cause trading to flee to other financial centers that don’t have such a tax, but the United Kingdom has long had one in place without any apparent damage.

Increasing the payroll tax by 1 percentage point (half a percentage point on employees and half a point on employers) would add about 0.3% to revenues.133 This would leave the payroll tax rate well below that in many European countries, and almost certainly below the level at which it would be a significant deterrent to employment.

We could increase the cap on earnings that are subject to the Social Security payroll tax. A person’s earnings above $127,200 (as of 2017) aren’t subject to the tax. Because a growing share of total earnings in the US economy has gone to those at the top in recent decades, a growing share has been exempt from the tax. In the early 1980s, about 90% of earnings was subject to the Social Security payroll tax; this has dropped to below 85%.134 Raising the cap to get back to 90% would increase tax revenues by about 0.2% of GDP.

Finally, we could do away with the real estate tax credit, which allows homeowners to deduct their state and local property tax payments from the income on which they pay federal income tax. The evidence suggests this credit does nothing to increase homeownership, and it mainly goes affluent taxpayers. This would increase revenues by about 0.2% of GDP.135

This is just one of many possible ways to increase tax revenues.136 The point is that the technical details of getting an additional 10% of GDP are not difficult.

  1. Lane Kenworthy, “Social Democratic Capitalism,” The Good Society. 
  2. Lane Kenworthy, “Health Care,” The Good Society. 
  3. Dylan Matthews, “Donald Trump Promised ‘Insurance for Everybody’. Here’s How He Can Do It,” Vox, 2016; Jacob S. Hacker, “Stronger Policy, Stronger Politics,” The American Prospect, 2016; Sarah Kliff and Ezra Klein, “The Lessons of Obamacare,” Vox, 2017; Paul Starr, “The Next Progressive Health Agenda,” The American Prospect, 2017. 
  4. Aaron E. Carroll, “The Real Reason the U.S. Has Employer-Sponsored Health Insurance,” New York Times: The Upshot, 2017. 
  5. Ezra Klein, “Why an MRI Costs $1,080 in America and $280 in France,” Washington Post: Wonkblog, 2012. 
  6. Kenworthy, “Health Care.” 
  7. Doug Elmendorf, “Revisions to CBO’s Projections of Federal Health Care Spending,” Congressional Budget Office, 2014. 
  8. Robert Pollin, James Heintz, Peter Arno, and Jeannette Wicks-Lim, “Economic Analysis of the Healthy California Single-Payer Health Care Proposal (SB-562),” Political Economy Research Institute (PERI), University of Massachusetts-Amherst, 2017. 
  9. Sean P. Keehan, John A. Poisal, Gigi A. Cuckler, Andrea M. Sisko, Sheila D. Smith, Andrew J. Madison, Devin A. Stone, Christian J. Wolfe, and Joseph M. Lizonitz, “National Health Expenditure Projections, 2015-25: Economy, Prices, and Aging Expected to Shape Spending and Enrollment,” HealthAffairs, 2017, exhibit 4. 
  10. Pollin et al, “Economic Analysis of the Healthy California Single-Payer Health Care Proposal (SB-562).” 
  11. US Department of Labor, “National Compensation Survey: Employee Benefits in the United States, March 2016,” Bulletin 2785, 2016, table 32. 
  12. Jane Waldfogel, What Children Need, Harvard University Press, 2006, ch. 2; Jeanne Brooks-Gunn, Wen-Jui Han, and Jane Waldfogel, “First-Year Maternal Employment and Child Development in the First Seven Years,” Monographs of the Society for Research in Child Development, 2010; Maria del Carmen Huerta et al, “Early Maternal Employment and Child Development in Five OECD Countries,” OECD Social, Employment, and Migration Working Paper 118, 2011; AEI-Brookings Working Group on Paid Family Leave, “Paid Family and Medical Leave: An Issue Whose Time Has Come,” 2017. 
  13. Tommy Ferrarini and Ann-Zofie Duvander, “Earner-Carer Model at the Cross-Roads: Reforms and Outcomes of Sweden’s Family policy in Comparative Perspective,” International Journal of Health Services, 2010; Swedish Ministry of Health and Social Affairs, “Social Insurance in Sweden,” 2016; Social Security Administration, “Social Security Programs Throughout the World: Europe, 2016,” 2016. 
  14. Maya Rossin-Slater, Christopher J. Ruhm, and Jane Waldfogel, “The Effects of California’s Paid Family Leave Program on Mothers’ Leave-Taking and Subsequent Labor Market Outcomes,” Journal of Public Policy Analysis and Management, 2013. 
  15. For a sample proposal, see Christopher J. Ruhm, “A National Paid Parental Leave Policy for the United States,” in The 51%: Driving Growth through Women’s Economic Participation, edited by Diane Whitmore Schanzenbach and Ryan Nunn, Brookings Institution, 2017. 
  16. Lane Kenworthy, “Work-Family-Leisure Balance,” The Good Society. 
  17. According to one estimate, a program with six months of leave and a replacement rate of 66% would cost about 0.5% of GDP. AEI-Brookings Working Group on Paid Family Leave, “Paid Family and Medical Leave: An Issue Whose Time Has Come,” 2017, p. 25. 
  18. Lane Kenworthy, “Equality of Opportunity,” The Good Society. 
  19. Greg J. Duncan, Kathleen M. Ziol-Guest, and Ariel Kalil, “Early-Childhood Poverty and Adult Attainment, Behavior, and Health,” Child Development, 2010. 
  20. Jamie Sturgeon, “Here’s How Much the New Canada Child Benefit Will Give You Each Month,” Global News, 2016. 
  21. Nisha Chikhale, “A Child Tax Credit Primer,” Washington Center on Equitable Growth, 2016, figure 3. 
  22. Theodore R. Marmor, Jerry L. Mashaw, and John Pakutka, Social Insurance, CQ Press, 2014, ch. 5. 
  23. H. Luke Shaefer, Greg Duncan, Kathryn Edin, Irwin Garfinkel, David Harris, Timothy Smeeding, Jane Waldfogel, Christopher Wimer, Hiro Yoshikawa, “A Universal Child Allowance: A Plan to Reduce Poverty and Income Instability among Children in the United States,” 2016. 
  24. House Ways and Means Committee, Green Book, various editions; Michael J. Graetz and Jerry L. Mashaw, True Security: Rethinking American Social Insurance, Yale University Press, 1999; Marmor et al, Social Insurance, ch. 8. 
  25. Marmor et al, Social Insurance. 
  26. Lori G. Kletzer and Robert E. Litan, “A Prescription to Relieve Worker Anxiety,” Policy Brief 01-2. Peterson Institute for International Economics, 2001; Robert J. LaLonde, “The Case for Wage Insurance,” Council on Foreign Relations, 2007. 
  27. Jody Heymann, Hye Jin Rho, John Schmitt, and Alison Earle, “Contagion Nation: A Comparison of Paid Sick Day Policies in 22 Countries,” Center for Economic Development Research, 2009; Lane Kenworthy, “Social Programs,” The Good Society. 
  28. Bureau of Labor Statistics, “Selected Paid Leave Benefits,” 2012, using data from the March 2012 National Compensation Survey. 
  29. Swedish Ministry of Health and Social Affairs, “Social Insurance in Sweden,” 2016; Social Security Administration, “Social Security Programs Throughout the World: Europe, 2016,” 2016. 
  30. 706 Swedish kroner per day. This cap is indexed to inflation. 
  31. Marmor et al, Social Insurance, p. 153. The descriptive information and data in this section draw heavily from chapter 9 of this book. 
  32. David H. Autor and Mark Duggan, “Supporting Work: A Proposal for Modernizing the U.S. Disability Insurance System,” Center for American Progress and The Hamilton Project, 2010; Richard V. Burkhauser and Mary C. Daly, “Social Security Disability Insurance: Time for Fundamental Change,” Journal of Public Policy Analysis and Management, 2012; Virginia P. Reno and Lisa D. Ekman, “Social Security Disability Insurance: Essential Protection When Work Incapacity Strikes,” Journal of Public Policy Analysis and Management, 2012. 
  33. Kenworthy, “Social Programs,” The Good Society. 
  34. Michael Harrington, The Other America: Poverty in the United States, Simon and Schuster, 1962; Charles Murray, Losing Ground: American Social Policy, 1950-1980, Basic Books, 1984; David T. Ellwood, Poor Support, Basic Books, 1988; Theodore R. Marmor, Jerry L. Mashaw, and Philip L. Harvey, America’s Misunderstood Welfare State, Basic Books, 1990; Christopher Jencks, Rethinking Social Policy, Harvard University Press, 1992; Theda Skocpol, Protecting Soldiers and Mothers, Harvard University Press, 1992; Rebecca M. Blank, It Takes a Nation: A New Agenda for Fighting Poverty, Russell Sage Foundation and Princeton University Press, 1997; Martin Gilens, Why Americans Hate Welfare, University of Chicago Press, 1999; Ron Haskins and Isabel V. Sawhill, Creating an Opportunity Society, Brookings Institution, 2009; Lisa Thiebaud Nicoli, Half a Loaf: Generosity in Cash Assistance to Single Mothers Across U.S. States, 1911-1996, PhD dissertation, Department of Sociology, University of Arizona, 2012; Sandra K. Danziger, Sheldon Danziger, Kristin S. Seefeldt, and H. Luke Shaefer, “From Welfare to a Work-Based Safety Net: An Incomplete Transition,” Journal of Policy Analysis and Management, 2016. 
  35. Sharon Hays, Flat Broke with Children, Oxford University Press, 2003; Jason DeParle, American Dream, Penguin, 2004, ch. 14; Sandra Morgen, Joan Acker, and Jill Weigt, Stretched Thin: Poor Families, Welfare Work, and Welfare Reform, Cornell University Press, 2010. 
  36. Kathryn Edin and Laura Lein, Making Ends Meet, Russell Sage Foundation, 1997; Katherine S. Newman, No Shame in My Game: The Working Poor in the Inner City, Vintage Books and Russell Sage Foundation, 1999; Katherine S. Newman, Chutes and Ladders: Navigating the Low-Wage Labor Market, Russell Sage Foundation and Harvard University Press, 2006; DeParle, American Dream; Kathryn Edin and Maria J. Kefalas, Promises I Can Keep: Why Poor Women Put Motherhood Before Marriage, University of California Press, 2005; Paula England and Kathryn Edin, eds., Unmarried Couples with Children, Russell Sage Foundation, 2009; Timothy J. Nelson and Kathryn Edin, Doing the Best I Can: Fathering in the Inner City, University of California Press, 2013. 
  37. D. Ben-Galim and A. Sachraida Dal, eds., Now It’s Personal: Learning from Welfare-to-Work Approaches Around the World, Institute for Public Policy Research, 2009. 
  38. Jason DeParle, “Welfare Limits Left Poor Adrift as Recession Hit,” New York Times, 2012; Kathryn J. Edin and Luke K. Shaefer, 2.00 Dollars a Day: Living on Almost Nothing in America, Houghton Mifflin, 2015. 
  39. Center on Budget and Policy Priorities, “Policy Basics: Introduction to the Supplemental Nutrition Assistance Program (SNAP).” 
  40. Lane Kenworthy, “Inclusion: The Elderly,” The Good Society. 
  41. Thomas Geoghegan, “Get Radical: Raise Social Security,” New York Times, 2011; Michael Lind, Steven Hill, Robert Hiltonsmith, and Joshua Freedman, “Expanded Social Security,” New America Foundation, 2013. 
  42. Teresa Ghilarducci, When I’m Sixty-Four, Princeton University Press, 2008. 
  43. Graetz and Mashaw, True Security; Alicia H. Munnell, “Bigger and Better: Redesigning Our Retirement System in the Wake of the Financial Collapse,” in Shared Responsibility, Shared Risk, edited by Jacob S. Hacker and Ann O’Leary, Oxford University Press, 2012; U.S. Senate Committee on Health, Education, Labor, and Pensions, “The Retirement Crisis and a Plan to Solve It,” 2012. 
  44. Chris Farrell, “No 401(k)? Your State May Come to the Rescue,” NextAvenue, 2016; Mary Williams Walsh, “California Aims Retirement Plan at Those Whose Jobs Offer None,” New York Times, 2016. 
  45. The United States has one of the highest elderly employment rates among affluent nations. 
  46. Edward Glaeser, Triumph of the City, Penguin, 2011; Ryan Avent, The Gated City, Amazon Digital Services, 2011; Matthew Yglesias, The Rent Is Too Damn High, Simon and Schuster, 2013. 
  47. Lane Kenworthy, “A Decent and Rising Income Floor,” The Good Society. 
  48. The following description and data draw from Robert Collinson, Ingrid Gould Ellen, and Jens Ludwig, “Low-Income Housing Policy,” Working Paper 21071, National Bureau of Economic Research, 2015; Congressional Budget Office, “Federal Housing Assistance for Low-Income Households,” 2015; Matthew Desmond, Evicted, Crown Books, 2016; Center on Budget and Policy Priorities, “Policy Basics: Federal Rental Assistance.” 
  49. Collinson et al, “Low-Income Housing Policy.” 
  50. Robert Haveman, “Do Housing Vouchers Work?, Pathways, 2013; Will Fischer, “Research Shows Housing Vouchers Reduce Hardship and Provide Platform for Long-Term Gains Among Children,” Center on Budget and Policy Priorities, 2014; Collinson et al, “Low-Income Housing Policy.” 
  51. The Congressional Budget Office estimates that providing a voucher to the 8 million households with incomes below 50% of the area median and that don’t currently receive one would cost about $40 billion. 
  52. Adam Carasso, Gillian Reynolds, and C. Eugene Steurle, “How Much Does the Federal Government Spend to Promote Economic Mobility and for Whom?,” Economic Mobility Project, 2008; Eric Toder, Marjery Austin Turner, Katherine Lim, and Liza Getsinger, “Reforming the Mortgage Interest Deduction,” Urban Institute and Tax Policy Center, 2010; Steven C. Bourassa, Donald R. Haurin, Patric H. Hedershott, and Martin Hoesli, “Mortgage Interest Deductions and Homeownership: An International Survey,” Swiss Finance Institute Research Paper 12-06, 2015; Nisha Chikhale, “U.S. Homeownership Tax Policies Are Expensive and Inequitable,” Washington Center for Equitable Growth, 2017. 
  53. Lane Kenworthy, “Early Education,” The Good Society. 
  54. Lynda Laughlin, “Who’s Minding the Kids? Child Care Arrangements: Spring 2011,” US Census Bureau, 2013, table 6, using data from the Survey of Income and Program Participation (SIPP). See also Ajay Chaudry et al, “Child Care Choices of Low-Income Working Families,” Urban Institute, 2011; ChildCare Aware of America, “Parents and the High Cost of Child Care,” 2012. 
  55. Eric A. Hanushek, John F. Kain, Jacob M. Markman, and Steven G. Rivkin, “Does Peer Ability Affect Student Achievement?,” Working Paper 8502, National Bureau of Economic Research, 2001; James J. Heckman, “Schools, Skills, and Synapses,” Working Paper 14064, National Bureau of Economic Research, 2008; Robert Bauchmüller, Mette Gørtz and Astrid Würtz Rasmussen, “Long-Run Benefits from Universal High-Quality Preschooling,” AKF Working Paper, 2011; Barnett, “Getting the Facts Right on Pre-K”; Elizabeth U. Cascio, “Does Universal Preschool Hit the Target? Program Access and Preschool Impacts,” Working Paper 23215, National Bureau of Economic Research, 2017. 
  56. Timothy Bartik proposes an alternative that he estimates would cost half as much, or about 0.5% of GDP. It includes universal pre-K available to all four-year-olds, targeted full-time full-year childcare and pre-K available for all disadvantaged children from birth to age five, and targeted parenting services for first-time disadvantaged mothers and their children from the prenatal period until age two. See Bartik, Investing in Kids: Early Childhood Programs and Local Economic Development, W.E. Upjohn Institute for Employment Research, 2011, ch. 7. Other proposals for smaller-scale programs include Obama Administration 2014 Budget Proposal; Cynthia G. Brown, Donna Cooper, Juliana Herman, Melissa Lazarín, Michael Linden, Sasha Post, and Neera Tanden, “Investing in Our Children: A Plan to Expand Access to Preschool and Child Care,” Center for American Progress, 2013. 
  57. National Center for Education Statistics, “Expenditures.” 
  58. New York City and Boston each spend about $10,000 per child on their well-regarded public prekindergarten programs. Data source: David Kirp, “How New York Made Pre-K a Success,” New York Times, 2016. 
  59. OECD, Doing Better for Families, figure 1.11. 
  60. Chaudry et al, Cradle to Kindergarten. 
  61. Barbara R. Bergmann, Saving Our Children from Poverty: What the United States Can Learn from France, Russell Sage Foundation, 1996; Janet C. Gornick and Marcia K. Meyers, Families That Work, Russell Sage Foundation, 2003; OECD, Doing Better for Families; Claire Lundberg, “Maybe Working Moms Can Have It All — in France,” Slate, 2012. 
  62. Amy E. Lowenstein, “Early Care and Education as Educational Panacea: What Do We Really Know About Its Effectiveness?” Educational Policy, 2010; Charles Murray, “Response to Heckman: Weighing the Evidence,” Boston Review, 2012; Grover J. “Russ” Whitehurst, “Can We Be Hard-Headed About Preschool? A Look at Universal and Targeted Pre-K,” Brookings Institution, 2013; Whitehurst, “Does Pre-K Work? It Depends on How Picky You Are,” Brookings Institution, 2014; Will Wilkinson, “Does Subsidized Preschool Pay Off?” The Economist: Democracy in America, 2013; David J. Armor and Sonia Sousa, “The Dubious Promise of Universal Preschool,” National Affairs, 2014; Timothy J. Bartik, “Grading the Pre-K Evidence,” Investing in Kids, 2014. 
  63. Mary Eberstadt, “The Post-Welfare State Family,” The Weekly Standard, 2013. 
  64. Claudia Goldin and Lawrence F. Katz, The Race between Education and Technology, Harvard University Press, 2008, figure 6.1; Census Bureau; National Center for Health Statistics. 
  65. OECD, “SF1.3: Living Arrangements of Children,” OECD Family Database. 
  66. Reihan Salam, “The House Budget Committee on the Inequality Landscape,” National Review Online: The Agenda, 2011. 
  67. Controlling for education and other relevant factors, federal government employees have higher compensation (wages and benefits) than their private-sector counterparts but state and local government employees don’t. Jeffrey Keefe, “Debunking the Myth of the Overcompensated Public Employee: The Evidence,” Economic Policy Institute, 2010; Philipp Bewerunge and Harvey S. Rosen, “Wages, Pensions, and Public-Private Sector Compensation Differentials,” Working Paper 227, Griswold Center for Economic Policy Studies, 2012; Congressional Budget Office, “Comparing the Compensation of Federal and Private-Sector Employees,” 2012. 
  68. Harry Holzer, “Creating New Pathways into Middle Class Jobs,” Progressive Policy Institute, 2015; James Rosenbaum, Caitlin Ahern, Kelly Becker, and Janet Rosenbaum, “The New Forgotten Half and Research Directions to Support Them,” William T. Grant Foundation, 2015; Nancy Schwartz and Robert Hoffman, “Pathways to Upward Mobility,” National Affairs, 2015. 
  69. Schwartz and Hoffman, “Pathways to Upward Mobility.” 
  70. Holzer, “Creating New Pathways into Middle Class Jobs.” 
  71. Kenworthy, “College Education,” The Good Society. 
  72. Christopher Jencks, “The Graduation Gap,” The American Prospect, 2009. 
  73. Matthew M. Chingos, “Who Would Benefit Most from Free College?,” Brookings Institution, 2016; Kenworthy, “College Education.” 
  74. David Leonhardt, “Why Does College Cost So Much?,” New York Times, 2011. The Tennessee Promise program, which made community college tuition-free for students, seems to have yielded a significant enrollment response. Celeste Carruthers and William Fox, “Aid for All: College Coaching, Financial Aid, and Post-Secondary Persistence in Tennessee,” Economics of Education Review, 2016. 
  75. Richard Murphy, Judith Scott-Clayton, and Gillian Wyness, “Lessons from the End of Free College in England,” Brookings Institution, 2017. 
  76. College Board, “Trends in College Pricing,” various years. 
  77. Adam Davidson, “Is College Tuition Really Too High?,” New York Times, 2016. 
  78. Matt Phillips, “College in Sweden Is Free but Students Still Have a Ton of Debt. How Can That Be?,” Quartz, 2013. 
  79. Kevin Carey, “A Quiet Revolution in Helping Lift the Burden of Student Debt,” New York Times, 2015; Susan Dynarski, “America Can Fix Its Student Loan Crisis. Just Ask Australia,” New York Times, 2016; Dynarski, “What Does Cutting Rates on Student Loans Do?,” Brookings Institution, 2016. 
  80. William G. Bowen and Derek Bok, The Shape of the River, Princeton University Press, 1998; Barbara F. Reskin, The Realities of Affirmative Action in Employment, American Sociological Association, 1998. 
  81. Jeremy Ashkenas, Haeyoun Park, and Adam Pearce, “Even with Affirmative Action, Blacks and Hispanics Are More Underrepresented at Top Colleges Than 35 Years Ago,” New York Times, 2017. 
  82. Richard Kahlenberg, The Remedy: Class, Race, and Affirmative Action, Basic Books, 1996; Richard Kahlenberg and Halley Porter, “A Better Affirmative Action,” Century Foundation, 2013; Sigal Alon, Race, Class, and Affirmative Action, Russell Sage Foundation, 2015. 
  83. Lane Kenworthy, “Employment,” The Good Society. 
  84. Helen Ginsburg, Full Employment and Public Policy: The United States and Sweden, D.C. Heath, 1983; Gosta Rehn, “Swedish Active Labor Market Policy: Retrospect and Prospect,” Industrial Relations, 1985; Anders Björklund and Richard B. Freeman, “Generating Equality and Eliminating Poverty, the Swedish Way,” in The Welfare State in Transition: Reforming the Swedish Model, edited by Richard B. Freeman, Robert Topel, and Birgitta Swedenborg, University of Chicago Press, 1987; Jon Kvist and Niels Ploug, “Active Labour Market Policies: When Do They Work — and Where Do They Fail?,” Danish National Institute of Social Research, 2003; OECD, “Benefits and Employment, Friend or Foe? Interactions between Passive and Active Social Programmes,” in OECD Employment Outlook, 2003; Per Kongshøj Madsen, “How Can It Possibly Fly? The Paradox of a Dynamic Labour Market in a Scandinavian Welfare State,” in National Identity and the Varieties of Capitalism: The Danish Experience, edited by John L. Campbell, John A. Hall, and Ove K. Pedersen, McGill-Queen’s University Press, 2006. 
  85. Kenworthy, “Social Programs,” The Good Society. 
  86. David Card, Jochen Kluve, and Andrea Weber, “What Works? A Meta Analysis of Recent Active Labor Market Program Evaluations,” Working Paper 21431, National Bureau of Economic Research, 2015. 
  87. Robert Pollin, “Back to Full Employment,” Boston Review, 2011; Dean Baker and Jared Bernstein, Getting Back to Full Employment, Center for Economic and Policy Research, 2013; Jared Bernstein, The Reconnection Agenda: Reuniting Growth and Prosperity, 2016. 
  88. Peter Gottschalk, “The Impact of Changes in Public Employment on Low-Wage Labor Markets,” in Generating Jobs: How to Increase Demand for Less-Skilled Workers, edited by Richard B. Freeman and Peter Gottschalk, Russell Sage Foundation, 1998. 
  89. Lane Kenworthy, “How Do We Know?,” The Good Society. 
  90. David Card and Alan B. Krueger, Myth and Measurement: The New Economics of the Minimum Wage, Princeton University Press, 1995; Arindrajit Dube, T. William Lester, and Michael Reich, “Minimum Wage Effects Across State Borders: Estimates Using Contiguous Counties,” Review of Economics and Statistics, 2010; John Schmitt, “Why Does the Minimum Wage Have No Discernible Effect on Employment?,” Center for Economic and Policy Research, 2013; Doruk Cengiz, Arindrajit Dube, Attila Lindner, and Ben Zipperer, “The Effect of Minimum Wages on the Total Number of Jobs: Evidence from the United States Using a Bunching Estimator.” 
  91. David R. Howell, Kea Fiedler, and Stephanie Luce, “What’s the Right Minimum Wage?,” Washington Center for Equitable Growth, 2016; Nick Hanuer, “Confronting the Parasite Economy,” The American Prospect, 2016. 
  92. Census Bureau, “Historical Income Tables: People,” table P-43. 
  93. Arindrajit Dube, “Designing Thoughtful Minimum Wage Policy at the State and Local Levels,” Hamilton Project, 2013; Alan B. Krueger, “The Minimum Wage: How Much Is Too Much?,” New York Times, 2015. 
  94. For further discussion of this debate, see Lane Kenworthy, Social Democratic America, Oxford University Press, 2014, pp. 134-141. 
  95. Lane Kenworthy, “Do Employment-Conditional Earnings Subsidies Work?,” ImPRovE Working Paper 15-10, Herman Deleeck Centre for Social Policy, University of Antwerp, 2015. 
  96. This may affect its attractiveness to recipients, the degree to which it incentivizes employment, and the ways recipients spend the benefit money. Sarah Halpern-Meekin, Kathryn Edin, Laura Tach, and Jennifer Sykes, It’s Not Like I’m Poor: How Working Families Make Ends Meet in a Post-Welfare World, University of California Press, 2014. 
  97. Kenworthy, “Social Programs.” 
  98. Center on Budget and Policy Priorities, “Policy Basics: State Earned Income Tax Credits,” 2014. 
  99. Kenworthy, “A Decent and Rising Income Floor”; Kenworthy, “Shared Prosperity,” The Good Society. 
  100. The current federal EITC averages $2,300 per household. 
  101. Kenworthy, “Do Employment-Conditional Earnings Subsidies Work?” 
  102. Even if it does reduce wages, it’s worth noting that this kind of hazard exists with all insurance. Public pension programs encourage people to save less during their working years than they otherwise would. Unemployment insurance encourages people to remain out of work longer than they otherwise might. Affordable healthcare encourages people to use more health services than they truly need. In each case, we judge the likely cost to be smaller than the gain in economic security, psychological well-being, and social justice. Arguably, the same is true here. 
  103. Donna Cooper, “Meeting the Infrastructure Imperative,” Center for American Progress, 2012; American Society of Civil Engineers, “Report Card for America’s Infrastructure,” 2013; Federal Highway Administration, “Deficient Bridges by State and Highway System,” 2014; Monica Anderson and Andrew Perrin, “13% of Americans Don’t Use the Internet. Who Are They?,” Pew Research Center, 2015; Rosabeth Moss Kanter, Move: Putting America’s Infrastructure Back in the Lead, W.W. Norton, 2015; Federal Communications Commission, “Internet Access Services Report,” various years. 
  104. Evan Soltas, “The Myth of the Falling Bridge,” Bloomberg View, 2013. 
  105. World Economic Forum, The Global Competitiveness Report 2013-2014, 2015. 
  106. Cooper, “Meeting the Infrastructure Imperative.” 
  107. Congressional Budget Office, “Estimated Impact of the American Recovery and Reinvestment Act on Employment and Economic Output,” 2011; Mark Zandi, “An Analysis of the Obama Jobs Plan,” Moody’s Analytics: Dismal Scientist, 2011; Sylvain Leduc and Daniel Wilson, “Highway Grants: Roads to Prosperity,” Federal Reserve Bank of San Francisco Economic Letters, 2012. 
  108. Economic Innovation Group, “The New Map of Economic Growth and Recovery,” 2016. 
  109. Matthew Whitaker, Resolution Foundation. 
  110. Jed Kolko, “Trump Was Stronger Where the Economy Is Weaker,” FiveThirtyEight, 2017. 
  111. Harold Meyerson, “Place Matters,” The American Prospect, 2017. 
  112. Wikipedia, “List of Minimum Annual Leave by Country”; Rebecca Ray, Milla Sanes, and John Schmitt, “No-Vacation Nation Revisited,” Center for Economic and Policy Research, 2013. 
  113. Robert W. Van Giezen, “Paid Leave in Private Industry Over the Past 20 Years,” Bureau of Labor Statistics, 2013; Society for Human Resource Management, “2017 Holiday Schedules.” 
  114. Lane Kenworthy, Social Democratic America, Oxford University Press, 2014. 
  115. Lane Kenworthy, “Taxes,” The Good Society. 
  116. Lane Kenworthy, “Income Inequality,” The Good Society. 
  117. The amount paid by households in the bottom fifth is calculated as $1.8 trillion (the total tax revenue needed) multiplied by .02 (this group will account for 2% of the revenues) divided by 25.2 million (the number of households in this group) = $1,428. The calculation is analogous for the other four groups. With the top fifth, we can go further and break it down into subgroups. Those between the 80th and 90th percentiles would pay $21,100 more per year, those between the 90th and 95th percentiles $30,600, those between the 95th and 99th percentiles (average income $320,000) $54,600, and those in the top 1% (average income $1.8 million) $340,000. 
  118. Lane Kenworthy, “Are Progressive Income Taxes Fair?,” Consider the Evidence, 2011. As Adam Smith put it in The Wealth of Nations (book 5, ch. 2, part 2), “The subjects of every state ought to contribute towards the support of the government, as nearly as possible, in proportion to their respective abilities; that is, in proportion to the revenue which they respectively enjoy under the protection of the state.” 
  119. This assumes high-income households don’t alter their behavior in response to the increase in the effective tax rate. 
  120. Kenworthy, Jobs with Equality, Oxford University Press, 2008, figure 8.12. 
  121. This estimate is based on information in Alan B. Krueger, “A Future Consumption Tax to Fix Today’s Economy,” New York Times: Economix, 2009; Eric Toder and Joseph Rosenberg, “Effects of Imposing a Value-Added Tax to Replace Payroll Taxes or Corporate Taxes,” Tax Policy Center, 2010; Robert J. Barro, “How to Really Save the Economy,” New York Times, 2011; Andrea Louise Campbell, “The 10 Percent Solution,” Democracy, 2011. 
  122. See, for instance, Robert Kuttner, “Progressive Revenue as the Alternative to Caps, Commissions, and Cuts,” prepared for the Scholars Strategy Network, 2010. 
  123. In fact, a consumption tax can be made progressive. See Robert H. Frank, “Progressive Consumption Tax,” Democracy, 2008. 
  124. Bruce Bartlett, “Tax Reform That Works: Building a Solid Fiscal Foundation with a VAT,” New America Foundation, 2012. 
  125. Kenworthy Progress for the Poor, Oxford University Press, 2011, ch. 8. 
  126. Gabriel Zucman, The Hidden Wealth of Nations: The Scourge of Tax Havens, University of Chicago Press, 2015; Reuven S. Avi-Yonah, “International Tax Evasion and Avoidance: What Can Be Done?,” The American Prospect, 2016; Jared Bernstein, “We’re Going to Need More Tax Revenue. Here’s How to Raise It,” The American Prospect, 2016. 
  127. Lane Kenworthy, “Were the Bush Tax Cuts Worse for Progressivity or for Revenues?,” Consider the Evidence, 2011. 
  128. James Surowiecki, “Soak the Very, Very Rich,” The New Yorker, 2010. 
  129. Institute on Taxation and Economic Policy, “Who Pays Taxes in America,” various years. 
  130. In 2016, average pretax income of households in the top 1% was $1.827 million, according to the Institute for Taxation and Economic Policy. There were about 1.26 million households in this group. 6% of their total income is about 7% of the country’s $1.8 trillion GDP. 
  131. Cristobal Young, The Myth of Millionaire Tax Flight, Stanford University Press, 2017. 
  132. Lane Kenworthy, “Climate Stability,” The Good Society. 
  133. The total payroll (Social Security and Medicare) tax rate is about 15%, and since the mid-1980s it has consistently collected 6.6% to 7.0% of GDP. 
  134. John S. Irons, Testimony before the National Commission on Fiscal Responsibility and Reform, June 30, 2010. 
  135. Chikhale, “U.S. Homeownership Tax Policies Are Expensive and Inequitable.” 
  136. For more, see Bernstein, “We’re Going to Need More Tax Revenue. Here’s How to Raise It.”