It solves while decreasing costs and increasing positive effects on inequality and alleviating poverty. Independently avoids unemployment incentives.

Economists since Adam Smith have marveled at the potential magic of unconstrained competitive markets. Under certain conditions, markets can efficiently allocate resources in such a way that no intervention could make anyone better off without making someone else worse off. Those preconditions are routinely violated in the real world, but the free enterprise system produces a dizzying array of products and gets them to consumers around the world who are willing and able to pay for them. In the process, it produces an enormous amount of wealth. The problem is that the gains from economic activity are increasingly concentrated in a sliver of the population. Half of the population has seen little increase in real wages for decades. This raises normative issues about the desirability of our economic system, but even people with a high tolerance for inequality might be concerned that a system that increasingly leaves most voters behind is not politically sustainable over the long run. This article develops a straightforward correction to wage stagnation: a refundable wage tax credit, which I call a universal earned income tax credit (UEITC). The credit would match earnings up to a maximum amount. I illustrate two variants: a 100 percent credit on earnings up to $10,000 plus a $2,500 fully refundable child tax credit; and a 50 percent credit on earnings up to $14,000 plus a fully refundable $2,000 per person (adults and children) universal basic income (UBI). The cost of the new tax credits could be offset by a dedicated tax—either a broad-based, value-added tax (VAT) of 11 percent or a surtax of 12 percent of taxable income. The maximum credit would be indexed to economic growth as measured by GDP per capita. For the first time in decades, low- and middle-income workers would share in economic gains even if the factors suppressing market wages do not reverse. Workers at all income levels, rather than only the highest earners, would benefit from economic prosperity, finally fulfilling John F. Kennedy’s promise that “a rising tide lifts all boats.” A substantial wage subsidy is more efficient and direct than the trade and immigration restrictions currently favored by the right and the regulatory responses advanced by the left. Polls show that the public strongly supports assisting working people. The earned income tax credit (EITC), which provides a substantial wage subsidy for lower-income families with children, has garnered broad bipartisan support. But the EITC provides little or no support for workers without children at home or for families with somewhat higher earnings. The proposed UEITC would replace the EITC and provide meaningful assistance for low- and middle-income workers whether or not they have children living with them. Unlike the EITC, the UEITC would not directly penalize two-earner couples. Because the wage credit is universal and refundable (that is, it does not phase out with income and workers can receive it even if they do not owe income taxes), workers could safely request an advance payment from an employer. Self-employed workers and those with multiple employers would ultimately claim the credit less any advance payments on income tax returns. Like other income, but unlike the EITC and most other tax credits, the UEITC would be subject to income taxes. A UEITC would also beeasier for the Internal Revenue Service (IRS) to administer than the current EITC because eligibility would be simpler to determine. All adults with earnings would be eligible for the UEITC. The proposed options are quite progressive. Low- and middle-income workers would receive far more in tax credits than they would pay in additional taxes. I assume the new credits and their dedicated funding sources would phase in and the EITC would phase out between 2020 and 2023. In 2023, when fully phased in, the proposals would cut average tax bills for most people in the four lowest income quintiles. The federal tax cut for the bottom quintile would average about $3,800 to $4,500 (23.2 to 27.8 percent of after-tax income), depending on the option chosen. The tax cut for the middle quintile would average $2,600 to $3,100 (4.2 to 5.1 percent of after-tax income). High-income people would pay more because the VAT burden or surtax increase with income. The tax increase on the top quintile would average between $12,000 and $22,000 (4.7 to 8.0 percent of after-tax income). The tax burden on the top 1 percent of earners would increase by 7.5 to 13.8 percent of after-tax income ($128,000 to $235,000). By convention, the static distributional estimates ignore behavioral responses, but those responses increase the overall benefit for low-wage workers. Because the proposal significantly increases the reward to working, some individuals would choose to enter the work force or work more hours, raising their income. Over the long run, they would accumulate more human capital (job skills) that would increase their productivity and, thus, wages. The increased labor supply could depress wages, but evidence from the EITC suggests that this effect is small. Watson (2020) estimates that the EITC increases workers’ welfare by more than the static estimates despite a substantial labor supply response.1 An alternative to the UEITC, which received a boost from Andrew Yang’s presidential primary campaign, would be a UBI alone. The UEITC provides no benefit to people who are out of the work force. However, work-based subsidies have strong political advantages. The UEITC and enhanced CTC are targeted at workers and families with children, two groups that many Democrats and Republicans are willing to support. The public is wary of assisting able-bodied adults who do not work. Eighty-seven percent of respondents in a recent poll supported work requirements for public assistance programs (Doar, Bowman, and O’Neil, 2016). Section II describes the motivation for the study, explaining the challenges facing lower- and middle-class workers and why a wage subsidy could mitigate those challenges. Section III describes the proposal in detail, compares the UEITC and enhanced CTC or UBI to the current-law EITC and CTC, summarizes the distributional and revenue effects of the proposed reforms using the Tax Policy Center’s microsimulation tax model, and discusses the overall economic effects. Section IV compares the UEITC variants with other reform options and considers several possible complementary policy changes. Section V offers some conclusions. The proposals are designed to achieve four objectives: (1) Mitigate rising economic inequality and wage stagnation; (2) Broadly share the gains from economic growth; (3) Support work and families; and (4) Be available to anyone who works. Below, I outline the positive arguments for objectives 1 through 3 and the political rationale for objective 4. While much of the recent policy debate has been devoted to rising economic inequality in the United States, the more salient and politically destabilizing factor may be the sense among middle-income workers that they can’t get ahead. 2 Median real wages for full-time male workers have barely budged in 40 years (figure 1). That conclusion is only slightly dampened when wages are indexed by the deflator for personal consumption expenditures (PCE), which allows for more substitution among consumption items. Median wages understate compensation because of the exclusion of fringe benefits, especially the value of employer-sponsored health insurance (Schieber and Nyce, 2018). However, it is scant consolation to the typical worker that all pay increases since the 1970s have gone to cover the cost of increasingly expensive health insurance. Recent political discourse has blamed trade and immigration as the main culprits for stagnant wages, but many economists believe that technology is a prime factor (Krueger, 2012; Summers, 2013). 3 The growing availability of machine substitutes for human labor depresses wages. For example, cashiers now compete with self-checkout machines. The cashier’s wages will depend only on how much it costs to purchase and operate the machine equivalent. As the price of the machine falls and quality improves, there will be more and more downward pressure on wages. Muro, Maxim, and Whiton (2019) conclude that all occupations will be affected by automation in the future, although the extent of exposure will vary widely. Skilled workers are not exempt. Frey andOsborne (2013) predict that skilled laborers, such as lawyers, accountants, and economists, will also soon face steep competition from computers.4 Autor and Dorn (2013) develop the implications in a model where technology substitutes for workers engaged in routine tasks in the manufacturing sector, but less so for those with higher skills and for low-skilled workers in the service sector. Computers compete for routine tasks performed by production workers (for example, robots replace assemblers and painters in an automobile plant), but they are less useful as substitutes for low-skilled workers in service jobs. The demand for services has increased dramatically since the 1970s, which has increased wages and employment at the bottom of the skill distribution because many service-sector jobs are not easily mechanized, but computers have driven down demand for workers performing routine tasks. The robots have made managers and other highskilled workers more productive because firms can now produce more with less labor. Autor and Dorn argue that this explains patterns in both employment and earnings since 1980, and they rule out globalization as an alternative explanation. If automation of routine tasks is the prime explanation for wage stagnation, these patterns are likely to persist. Wage stagnation might simply represent a transitional problem as we adjust to new technology. Past waves of technological change initially cost jobs but ultimately vastly improved middle-class workers’ living standards. (Allen 2007) However, the historical context is much different now. The 30 years preceding the current malaise corresponded to very rapid real wage increases, and American workers might not be willing to wait decades more for the economy to fully adapt to new technology. The success of populist presidential candidates who proposed radical policies to help the middle class in the 2016 election—Donald Trump and Bernie Sanders—is evidence of growing impatience with workingclass wage stagnation.5 However, computers and robotics are getting cheaper and better at a remarkable rate. Many activities that reach a large enough scale to overcome the fixed costs of investment in automation will be automated and many workers displaced. Low- and medium-skilled human workers will still be valuable in small-scale start-up enterprises and in some kinds of service jobs, but 21st century workers will need to be highly adaptable and may find themselves in places where only low-wage jobs are available.6This trend might be exacerbated by the COVID-19 pandemic for at least two reasons. First, robots don’t get sick (Bloom and Prettner, 2020).7 Second, the economic collapse accompanying the pandemic disproportionately hurt smaller businesses, which are the least likely to automate.8 The new economy with more economic activity in large businesses will automate faster. Other factors also contribute to wage stagnation. In addition to trade, immigration, and technology, Clausing (2019) identifies declining unionization, the diminishing real value of the federal minimum wage, rent seeking, a winner-take-all economy, and tax policy as important factors in wage stasis. Baker (2016) argues that misguided public policies promote economic concentration, rent seeking, and outsize executive compensation at the expense of rank-and-file workers. Reversing the policies that tilt the economy in favor of the rich would raise incomes for the bottom 99 percent, but those policies have powerful advocates. Raising minimum wages and repealing antilabor laws could similarly help some low- and middle-income workers, but vigorous debate persists in the economics literature about the costs and effectiveness of such policies.9 The evidence, therefore, suggests that wage stagnation is real and unlikely to change in the foreseeable future. Assuming the economy continues to grow, that means middle-class workers’ relative position will continue to erode compared with those at the top of the income distribution. This dynamic will become increasingly politically unstable over time. Middle-class malaise creates a growing risk of populist responses such as trade barriers, immigration bans, and burdensome wage and employment regulations that could harm the economy (Burman, 2013). Such responses might cause the middle-income worker’s position to rise in relative, but not absolute, terms. Inequality would diminish, but everyone could end up worse off. Developing an efficient compensation mechanism that would induce middle-class voters to endorse pro-growth policies could thus benefit all income groups. Value judgments aside, there is a positive economic argument for developing efficient public policies to address middle-class wage stagnation directly because they may be necessary to permit progrowth policies to survive.