Between 1945 and 1980, real GDP per capita doubled. Equally striking, the rise in living standards was actually faster for those at the bottom than those at the top. Educational attainment soared, keeping pace with technological progress. Children grew up with a (reasonable) expectation they would outearn their parents . The American Dream was alive and well.

Then, something — well, many things — changed. GDP growth continued after 1980, with the usual pauses for recessions. But the rising tide no longer carried all boats.

The portion of national income going to labor fell, and the fruits of productivity growth increasingly went to a moneyed, highly educated elite. In 1980 to 2014, the share of national income earned by those in the top 1 percent rose steeply, to about 20 percent. Meanwhile, the share going to the bottom half declined from 20 percent to 12 percent.

These statistics will not surprise most readers. Nor do the often-cited causes, among them global economic integration, technological change that sharply increased the returns from higher education, the greater market power of large companies, changes in social norms and unionization rates and, of course, the broader policy environment.

But what can be done? Populist remedies that depend on turning back the clock on international trade or factory automation are self-defeating. Social norms and the market power of large companies are difficult to change. But governments do control two crucial things: how tax dollars are spent, and how tax dollars are collected. Here I focus on the role of tax policy, offering a brief outline of the issues, and highlighting the major changes in taxes whisked through Congress in December, 2017 that will only continue these malign trends.

Some Nuts and Bolts

The tax system plays a crucial role in affecting income inequality, both by changing the incentives facing individuals in the economy (to work, save and invest) and by collecting more in taxes from some people than others. The federal personal and corporate income taxes, and the estate tax, are progressive. The federal payroll tax, on the other hand, is regressive since the marginal rate (on the non-Medicare portion) falls to zero above a threshold ($128,400 in 2018). However, payroll tax revenues are dedicated to Social Security benefits, which are themselves progressive, since the benefits relative to pre-retirement income are higher for low-income households.

While the chart above shows that the net effects of taxes and transfers are to reduce inequality, some taxes are clearly more progressive than others. According to the Congressional Budget Office, the federal income tax is one of the most progressive because the earned income tax credit creates an average tax rate for the lowest fifth of households of minus 7.2 percent, and average tax rates climb steadily as incomes increase, reaching 15.5 percent for the top quintile. (These figures are all before the new tax law.)

The only taxes that are more progressive than the income tax are the estate tax, which is paid by only the wealthiest one-fifth of 1 percent of estates (before the tax law change), and the corporate tax, which is mostly borne by those at the top of the distribution, since capital income is highly concentrated.

While there is debate over how much of the corporate tax burden is passed on to workers in the form of lower wages, the near consensus among nonpartisan analysts – including the Joint Committee on Tax, the Tax Policy Center and the Congressional Budget Office — is that the bulk of the burden falls on the owners of capital or shareholders. According to U.S. Treasury economists, about three-quarters of the corporate tax base consists of excess profits — profits exceeding the level needed to induce companies to remain in business. Taxes on excess profits are borne by shareholders, not workers. In addition, exhaustive studies of the cross-country evidence have failed to reveal any effects of corporate taxation on wages.

Other important impacts of tax policy on behavior are more difficult to estimate. For example, the tax system sends important signals about social norms, and tax rates may serve as cues in determining how aggressively economic actors bargain. Observers have noted a suspiciously tight, negative relationship between tax rates faced by those at the top of the distribution and the share of national income accruing to those at the top.

Earlier this century, tax increases at the high end of the income distribution preceded falling income shares for the top 1 percent of the population, whereas more recently, decreases in the top rate preceded increases in the share of income received by the top 1 percent.

Of course, every good economist knows that correlation need not imply causality, and there are multiple explanations for this pattern. For example, one possibility is that tax evasion and avoidance are more attractive when tax rates are high, so high incomes are simply more apparent when tax rates are lower.

However, it is also possible that low tax rates feed the dynamic of inequality by leading those with high incomes, such as managers and executives, to be more aggressive in bargaining to increase their share of total compensation. In contrast, when tax rates are higher, taxes may serve as a brake. Lower taxes allow high-income earners to keep more at the margin, giving them stronger incentives to demand more. More subtly, lower taxes on higher earners may serve as a signal that society believes the rich deserve their incomes, improving the bargaining position of those at the top.

According to this view, the brake provided by higher tax rates was relaxed just as other forces in the global economy (such as technological change, globalization and market power) shifted demand away from less-skilled labor. As a result, pay packages at the top surged.

While more research is needed to properly understand these influences, the tax system affects income distribution in subtle as well as unsubtle ways. Our tax code is a statement of values. When we give preference to some activities over others, it affects economic incentives and also conveys society’s sense of what is valuable.

Despite the significant widening of income inequality over the past four decades, we haven’t countered these troubling trends by increasing the progressivity of the tax system. Quite the contrary: many (though not all) of the tax law changes of the past 35 years have made the tax system less progressive. In particular, declines in top tax rates on earned income, capital gains and dividends, alongside the exemption of a vast majority of large estates from taxation, have reduced progressivity for those at the top of society. And while the overall effect of the tax system is to reduce inequality, it reduces inequality less than it did in the past.

The Tax Cuts and Jobs Act: Making a Bad Situation Worse

The 2016 election consolidated control of the federal government in a single party. Republicans have long asserted the virtues of less government and have been fairly consistent in honoring their values by cutting taxes when in power. One could argue that cutting taxes rather than spending puts the cart before the horse. But, of course, tax cuts are popular while spending cuts are not. The Machiavellian explanation is that starting with tax cuts is part of a broad strategy to “starve the beast,” forcing those on the left to reduce their spending down the road when they return to power, in order to prevent ballooning budget deficits.

The Tax Cuts and Jobs Act (TCJA), signed in December, is centered on business tax cuts. The legislation lowers government revenues by $1.5 trillion over 10 years. This is certainly consistent with a starve-the-beast strategy as well as the belief that business should be taxed less in order to spur economic growth. Most of the personal income tax provisions in the legislation are relatively minor. On average, there are small cuts in household taxes in the early years of the legislation and small increases in the later years when most of the provisions expire. The corporate tax cuts, however, are permanent, and indeed understated in the budget cost.

That’s because a special low rate (8 or 15.5 percent) on past offshore corporate earnings is expected to raise about $340 billion in revenue in the short run, but actually represents a large tax cut compared with the previous law in the long run. (Before the adoption of the TCJA, this foreign income would have been taxed at 35 percent upon repatriation, less foreign taxes paid.)

Analyses of the distributional impact of the legislation indicate that, once the dust settles, the great bulk of the net tax cuts will go toward lightening the tax burden on high earners. In 2018, the percentage gain in after-tax income for the top 5 percent of the distribution will be five times the gain of those in the bottom 40 percent. In 2018, the bottom 80 percent of the income distribution will enjoy an average tax cut of a bit less than $800, whereas the top 1 percent will get an average cut of over $50,000.

By the time the law is fully phased in and the personal income tax cuts expire, the bottom 80 percent of the population will on average pay $15 more in tax annually, while the top 1 percent will retain net benefits exceeding $20,000. The index used to adjust tax brackets to inflation has also been changed, causing a growing stealth tax increase on personal income relative to prior law.

A Missed Opportunity

As noted above, the rate cuts for corporations are permanent. And while there has long been bipartisan acknowledgment that the corporate tax was in need of reform, this change amounts to a missed opportunity since it doesn’t pair a cut in rates with base-broadening provisions that would make the tax both more equitable and less costly. Indeed, in several key ways, the corporate tax base is even narrower than before.

For example, the international provisions of the tax law (ignoring the special low rate on prior offshore earnings) actually lose a bit of revenue compared with prior law. This was quite difficult to manage, since estimates suggest that the U.S. government was already losing over $100 billion in revenue each year because of creative accounting that shifted profits to low-tax countries. At the end of the day, businesses will get large tax cuts, and the cost of those cuts will be borne by middle-income households and future taxpayers facing higher federal debt burdens.

Waiting for the Growth Fairy

Of course, businesses are people, too; when businesses get tax cuts, someone benefits. Those people tend to be well-off. Still, proponents of the bill argue that the legislation will usher in an era of investment-led growth that will spur large increases in wages. In fact, Trump administration officials assert that wages will rise by as much as $4,000 to $9,000 annually.

While it would be pleasing to live in a world of free lunches, both theory and evidence contradict these claims. Few economists support these rosy trickle-down scenarios. On the contrary: 37 of 38 top economists from across the ideological spectrum regularly polled in the University of Chicago business school survey did not agree that the legislation would lead to substantially higher growth. Claims of high wage effects have also been debunked, often by the very same economists who wrote the studies that were cited by Trump administration officials.

For growth to materialize, companies would need to increase investment in response to those tax cuts. And for those investments to benefit workers, they would need to increase worker productivity. Last but not least, the higher productivity would need to lead to higher wages.

Start with the first big assumption — that lower tax rates will lead to a big surge in investment. In the sort of economic models featured in introductory economics textbooks, lower taxes do mean higher expected returns on investment and therefore higher rates of investment. But real world features of our tax system make for a big difference in expected magnitudes.

Consider first that debt-financed investments — a substantial portion of investment — already receive a net subsidy through the tax system, and these investments are not treated more favorably under the new tax law. Second, while the law reduces the nominal corporate tax rate from 35 percent to 21 percent, the effective tax rate reduction for companies already adept at exploiting loopholes in the law to reduce their taxes will be much smaller.

Third, some of the corporate tax provisions, such as the minimum tax on foreign earnings, perversely encourage offshore investments at the expense of domestic investments since the bite of the tax is smaller when companies hold more assets offshore, and the tax rate is lower for income earned abroad than income earned at home. Also, the move to a “territorial” tax system (which exempts foreign-earned income from taxation) makes explicit — and permanent — the tax preference for earnings outside the United States.

In any event, we are hardly in an environment in which companies with good investment opportunities are constrained by their access to capital. On the contrary: we are awash in capital. Corporate after-tax profits are far higher than they were in decades past, and corporate savings have been rising substantially as a share of GDP. Instead, we face a shortage of investment opportunities relative to the supply of capital, as emphasized by the secular stagnation and savings glut concerns of Larry Summers and Ben Bernanke.

What is limiting investment opportunities? In part, the economic weakness of the middle class is responsible. For companies to justify expansion, they must have a reasonable expectation of increased sales. And when wage growth is weak and households are loaded up with student debt, mortgage debt and credit card debt, consumer demand is attenuated.

Now consider another link on the alleged chain from lower corporate tax rates to higher wages. Even if lower taxes materially raise investment, it is far from obvious this will increase the demand for labor. Some types of investment displace semiskilled labor. Indeed, American manufacturers produce 80 percent more value than 30 years ago with just 70 percent as many hours of work. And there is good reason to believe that labor displacement will continue. For example, the predicted shift to autonomous vehicles will probably devastate the market for truck drivers. By no coincidence, wage growth has lagged behind productivity growth in recent years, leading to a persistent decline in labor’s share of income.

Of late, some companies have announced small wage increases or bonuses, linking these decisions to the recent tax law changes. These wage increases usually represent very tiny shares of the tax cut magnitudes, and, often, these announcements reflect publicity goals rather than economics. For example, it should take substantial time for any additional investments to raise wages, even if the tax cuts did lead to new investments. In one colorful example, Walmart announced a $2-an-hour increase in its base wage, linking the news to the passage of the tax cut. Yet the announcement came the same day it announced 63 store closings, and the company was already under pressure to raise wages to remain competitive in a tightening labor market.

Some proponents of corporate tax cuts rely instead on a demand-side case for such tax cuts to benefit workers. Under this logic, even the rich will spend some share of added post-tax income, and this will raise aggregate demand and economic output accordingly. Yet while such a mechanism is widely accepted to exist in a recession, we are now at the “full employment” level of output. Indeed, from this starting point, the Federal Reserve will work to offset any net stimulus by raising interest rates, since fueling demand in an economy at full employment is more likely to generate inflation than sustainable growth. Remember, too, that these tax cuts will increase federal deficits by about $1.5 trillion, and thus act as a drag on growth in the long run by “crowding out” productive investment or sucking in capital from abroad. The latter is no disaster. But since extra borrowing from abroad must eventually be repaid, it will tend to reduce Americans’ standards of living in the future.

Before the TCJA, the United States’ debt was expected to increase faster than GDP over the next decade as a result of increasing spending on Social Security and Medicare to cover the aging of retired baby boomers. By further adding to this already large debt burden, Washington will have substantially less leeway to stimulate the economy with lower taxes or increased spending when the next recession arrives. And recessions always do arrive.

Better Tax Policy

The new tax law is a large step in the wrong direction: it increases deficits, widens income inequality and makes the tax system far more complex. In an era in which government spending must increase in response to demographic factors, any tax reform should be revenue neutral. And after decades of rising income inequality, any tax cuts should surely go to those in the bottom half of the income distribution. Let tax cuts for the middle class bubble up to help businesses, rather than relying on tax cuts for those at the top to trickle down to workers.