Post-Redistribution Liberalism

Changing the tax structure, even radically, won’t really change much. We need to look to pre-tax inequities to transform society.

By Isabel Sawhill

President Obama once called inequality “the defining issue of our time.” How right he was. The question now is not whether we have a big problem; the question is what to do about it.

Progressives naturally gravitate to the idea of redistributing income. They often take the market distribution of income as a given. Why not simply raise taxes on the wealthy and use the proceeds to beef up spending on programs that help the middle class and the poor?

First off, let me make it very clear that I am in favor of such redistribution. But we liberals may need to start better recognizing the difficulty of achieving such redistribution in practice, and in light of those difficulties, consider other ways of reducing inequality that focus more on market incomes and less on tax and benefit programs.

Given the extraordinary rise in income inequality that has occurred over the past four decades, we need all of the weapons we can find to fight the war on inequality. More specifically, it’s time to take a closer look at what might be achieved if we could persuade or nudge American businesses, through tax policy, to adopt what I will call a “stakeholder model of business.” That means not making every decision based on maximizing short-term profits and shareholder value, but instead focusing at least as much attention on broader goals and values, such as the kinds of compensation schemes and training investments that will enhance worker productivity and pay. This can be done without sacrificing long-term profitability. In fact, it may actually improve economic competitiveness and long-term growth. Relying on redistribution alone to accomplish the same goal is a flawed strategy.

The Limits on a Redistributionist Agenda

The difficulty of enacting a redistributionist agenda is primarily political, not economic. The United States is a wealthy country and can afford to tax the wealthy and corporations more without slowing economic growth in the ways (wrongly) emphasized by supply-side conservatives. Nonetheless, the kinds of tax increases needed are more substantial than our politics may allow at this time. It’s easy to fantasize about a post-Trump liberal politics that would open the door to such tax increases. But we need to remember that even when Democrats had a clear majority in Congress, they have not enacted such measures. Obama proposed to limit deductions for those with incomes above $250,000 in his budget every year. That’s the kind of base broadening that tax experts have long favored, and it would be very progressive. But Congress never took any action on it, even when Democrats held a majority. When the Obama Administration tried to do away with the tax benefits associated with 529 college savings plans, benefits that disproportionately help those with six-figure incomes, his own party rebelled. My Brookings Institution colleague Richard Reeves writes about this incident in his book, Dream Hoarders. He goes on to explain how the upper middle class must play a critical part of any redistributionist solution. The entire burden of funding government can’t be put on the top 1 percent. The “Dream Hoarders,” with their six-figure incomes, are also part of the problem.

Lest we assume that a Bernie Sanders-type figure is going to arrive on a white horse with columns of liberal troops behind him that are capable of radically changing the battlefield, we also need to understand the deeper roots of this war.

Once a certain distribution of income is in place, changing it requires creating clear winners and losers. The losers will feel their pain much more acutely than the winners will appreciate their gains. Behavioral scientists call this “loss aversion.” Experiments have shown that the psychological effect of a loss is about twice the benefit of an equivalent gain—an irrational, albeit widespread phenomenon first unveiled by Daniel Kahneman and Amos Tversky. In addition, if the pain is concentrated on a small group (say, the top 1 percent) and the gains are spread among a much larger number of people (say, the poor and the middle class), the imbalance between loser pain and winner gain will be even greater. The small group of losers will mobilize to protect their status while the much larger group of winners will be a lot harder to organize. Of course, the power of that small group’s money in politics makes matters worse, although political scientists who study its effects are less convinced than the rest of us that it is the determining factor in most electoral outcomes, especially at the national level (if it were, Jeb Bush would have won the Republican primary.

Studies have shown that  Americans believe that merit is more important than luck in explaining a high income.

Another reason why it is difficult to redistribute in any truly decisive way is the fact that, in the United States, public attitudes are more conservative than they are in many other advanced countries. Americans believe that they live in a is a land of opportunity where anyone can make it if they work hard enough. The truth is that most of us who do well in life benefited from the luck of the draw. We were born into a rich rather than a poor country, to the right kind of family, with good health, and good genes. These advantages have nothing to do with merit. Of course, we need to play whatever cards we were dealt with skill and energy, but the cards we begin with matter a lot. Yet, as a number of studies have shown, the American public believes that merit is more important than luck in explaining a high income. And even when it is good fortune that leads to a high income, people in the United States think one should be rewarded anyway.

In a nationally representative survey in which Americans were asked what they thought the top marginal tax rate on incomes should be, the bulk of the responses were between 20 and 40 percent with a median of 30 percent and a mean of 33 percent. These averages are well below the existing rate of 39.6 percent. One possible reason for this finding is a lack of information about how much money people at the top actually have. (Another possibility is that the respondents didn’t understand the concept of marginality. A high marginal tax rate doesn’t lead to a high average rate in a progressive system.)

And here’s the even more dispiriting part. When one group of individuals was supplied with data on the high degree of inequality in today’s economy and another group was not, the responses of the more informed group changed only a little. They favored a top marginal rate that was just 1 percentage point higher than before. This research was conducted by Kenneth Scheve from Stanford and David Stasavage of New York University. They have studied the history of taxation and believe that, unless there is a crisis (such as a war), or unless people believe that the government has treated the rich unfairly, they are not likely to favor higher top rates. And unfairness, they note, has nothing to do with inequality. In the public mind, a lack of fairness is related to being able to take advantage of complex rules by hiring an accountant and then getting away with manipulating the rules in self-serving ways. Focus on the complexity and its uneven impacts and one might have a politically salient argument. But that’s a very different rationale than arguing that the rich should pay a higher rate because they can.

How Much Could Redistribution Achieve?

Setting aside some of the political difficulties for the moment, suppose we raised marginal tax rates on the highest income households from 39.6 percent to 50 percent. This “what if” experiment has already been carried out by my colleagues, William Gale, Melissa Kearney, and Peter Orszag using a tax-simulation model created by Brookings and the Urban Institute. They find that the increase would raise taxes by an average of $6,464 for those in the 95-99th percentiles (those with average incomes of $321,000 in 2013). Households in the top 1 percent (with average incomes of $1.571 million in 2013) would pay an additional $110,968 and those in the top 0.1 percent an additional $568,617. By one measure, this change by itself has a tiny effect on the overall distribution of income, reducing the United States’s after-tax Gini coefficient from .574 to .571. (The Gini is a summary index of inequality that is equal to 1 when all of society’s income goes to just one person and to 0 when there is complete equality.)

Now imagine that all of the revenue collected from this change was distributed evenly to the bottom 20 percent. The total revenue raised is $95.6 billion and allows each household at the bottom to have an extra $2,650 in post-tax income. The Gini now falls to .560, not terribly different from the .574 where we started. This leads the authors to conclude that “such a sizable increase in the top personal income tax rate leads to a strikingly limited reduction in overall income inequality” and that this “speaks to the limitations of this particular approach to addressing the broader challenge. It also reflects the fact that the high level of U.S. income inequality is characterized by a wide divergence in income between higher-income households and those at the middle and below.” However, the ratio of top to bottom incomes (the 90/10 ratio) falls from 16.7 to 12.5. Because the action here is at the tails of the distribution, this latter measure (of the ratio between incomes) may be a better guide to understanding the effects of the policy than the Gini Index alone. We could debate whether an increase in marginal tax rates to 50 percent is enough (recall that the United States had far higher rates in the 1950s). We could also debate whether an income bounce of $2,650 to those at the bottom is large or small. To my way of thinking, this would be well worth doing, although I agree with the authors that it is not as powerful a remedy to inequality as many might believe.

Progressives have been inspired by the Trump election to think more boldly than in the past. Some are talking about a universal children’s allowance. Some are talking about a guaranteed jobs program. Others about a universal basic income. But the problem remains that we haven’t thought enough about how to raise the revenues that would be needed to pay for any of these schemes. All of them are frightfully expensive. They would require very large tax increases at the top, but also increases that would affect the upper middle class.

One solution is to leave taxes on earnings untouched and focus on unearned and inherited wealth. The 3.8 percent surtax on net investment income that was enacted as part of the Affordable Care Act is a nice example. Another worthy proposal is to tax capital gains at death rather than let very large accumulations of wealth be transferred to the next generation tax-free.

Another solution would be to scrap the income tax for everyone except those with six-figure incomes and replace it with a value added tax (VAT), a type of national sales tax. Progressives typically react to such a suggestion by noting that a VAT would be regressive. But it would also raise a ton of revenue that could be redistributed in a progressive fashion. The net result might accomplish more in the fight against inequality than first meets the eye. Europeans finance their social welfare systems with a VAT and it seems to work quite well. A single-payer health care system financed by a VAT, for example, has a lot to recommend it. Moving to an entirely new form of taxation also helps to get around the loss-aversion problem, as it would be much less clear whose ox was being gored. Because individuals would be freed from the burden of having to file their taxes every year, and worrying about whether they were being screwed by other people taking advantage of deductions not available to them, most middle-class families might be thrilled. Notably, a number of Republicans endorsed a version of the idea in the 2016 presidential primary. As Larry Summers once quipped, progressives don’t like a VAT because it’s regressive; conservatives don’t like it because it is a revenue raiser. We will get a VAT when conservatives realize it is regressive and progressives realize it is a revenue raiser. In short, there is a lot to like here from both a political and a substantive perspective.

While any number of these changes would be a good idea, wholesale restructuring of the tax system is also difficult to imagine even under unified Democratic control of the government. But there may be a way to organize the private sector that would make that kind of heavy redistributive lift significantly less arduous.

Stakeholder Capitalism to the Rescue?

Stakeholder capitalism is not a new concept, but it seems to have gone out of favor in recent decades. It means paying attention not just to shareholders but also to workers, customers, and the community. It has proven to be a successful strategy for many companies. They have showcased what can be accomplished when the private sector takes greater responsibility for helping workers—whether in the form of profit sharing, training, or providing benefits such as paid leave and flexible hours. The fact is that without such an approach, it will be difficult to achieve broadly based economic growth. It would simply require too much redistribution after the fact. We need instead to test the limits of equalizing the distribution of market incomes before taxes and benefits enter the picture. Let me emphasize that this is not an either-or proposition. We need to do both.

It’s important to note that this focus on what businesses can do is important because the private sector is responsible for nearly 85 percent of all jobs in the economy and for most of the income that ends up in people’s pockets.

In the past, workers relied on unions to bargain for higher pay and better benefits and working conditions. But we all know unions’ glory days are gone, particularly in the United States. I don’t want to argue that they can’t still make a difference; just that we shouldn’t count on them to be the major driver of higher pay and improved working conditions in the future.

A better approach, I think, is to build on what some progressive employers are already doing, to spread the word about their successes, and to catalyze others to follow suit. There is plenty of evidence that stakeholder capitalism need not be at odds with shareholder capitalism. Businesses can still do well while doing good. Those that focus on the longer term, invest in their workers and their communities, and share their profits may actually improve overall productivity, reduce turnover, and engage their employees, all of which may end up being a win-win for both shareholders and employees.

The first thing that has to change is the mindset of corporate America with respect to one central, and harmful, idea that has come to dominate corporate thinking. As Steven Pearlstein, a business columnist with The Washington Post, put it: “In the recent history of management ideas, few have had a more profound—or pernicious—effect than the one that says corporations should be run in a manner that ‘maximizes shareholder value.’ ” Indeed, you could argue that much of what Americans perceive to be wrong with the economy these days—the slow growth and rising inequality; the recurring scandals; the wild swings from boom to bust; the inadequate investment in R&D, worker training and public goods—has its roots in this ideology.

The funny thing is that this supposed imperative to “maximize” a company’s share price has no foundation in history or in law. Nor is there any empirical evidence that it makes the economy or the society better off. What began in the 1970s and ’80s as a useful corrective to self-satisfied managerial mediocrity has become a corrupting, self-interested dogma peddled by finance professors, money managers, and over-compensated corporate executives.”

These are strong words, and there are, as always, two sides to this story. Corporations are not evil and managers must deal with competing goals even when they subscribe to a stakeholder philosophy. But I think Pearlstein is right about both the law and the evidence. He further suggests that what may be good for any one company is not necessarily good for society, and that those social benefits create a strong case for using corporate tax reform to recognize the social benefits. There is now broad agreement that corporate tax rates need to be reduced. However, they also need to be reformed in a way that will contribute not just to more growth but to more inclusive, longer term growth.

Stakeholder Capitalism: It’s Legal and It Works

Is a corporation that fails to maximize shareholder value breaching the legal terms of its incorporation? There have been numerous court cases on this issue, including the pivotal Delaware decision of 1919, Dodge v. Ford, stating, “A business corporation is organized and carried on primarily for the profit of the stockholders. The powers of the directors are to be employed for that end.”

This view has become the norm in recent decades. However, as Lynn Stout, a prominent legal scholar on this subject, writes:

contrary to popular belief, the managers of public companies have no enforceable legal duty to maximize shareholder value. Certainly they can choose to maximize profits, but they can also choose to pursue any other objective that is not unlawful, including taking care of employees and suppliers, pleasing customers, benefiting the community and the broader society, and preserving and protecting the corporate entity itself. Shareholder primacy is a managerial choice—not a legal requirement.

Still, norms are a powerful thing. As Matthew T. Bodie put it, “Although the vibrancy of shareholder primacy has at times been called into question as a matter of law, both boardrooms and courts have taken the normative call for shareholder wealth maximization increasingly to heart.”

At the same time, many firms have found that investments in employees and communities are not mutually exclusive with making a profit. Indeed, some of the most profitable and productive companies in the country have generous benefit packages, share profits with employees, and invest in worker training and environmental protection. What is often unclear, however, is whether their profits allow for these investments or whether the causal arrow goes in the other direction. As I will argue below, it seems to be some of both, but one cannot exclude the possibility that these investments actually enhance the bottom line.

In 30 states and the District of Columbia, companies can now legally incorporate as “benefit corporations,” giving directors encouragement to consider the interests of all stakeholders—including workers, consumers, and communities—in addition to their shareholders.

One well-known example is Ben & Jerry’s. From the beginning, this popular Vermont-based ice cream company built stakeholder capitalism into its business model. It pays its lowest-wage workers more than double the minimum wage and is a certified “B Corporation.” Its mission includes, among other goals, “increasing value for our stakeholders and expanding opportunities for development and career growth for our employees.” As David Gelles reported, “Ben & Jerry’s is proceeding with its activism-infused capitalism, one pint of Chunky Monkey at a time.”

I do not necessarily think that every company should go out and become a Ben and Jerry’s, but I do believe that the private sector could play a bigger role in ensuring that prosperity is broadly shared without undermining their profitability. Workers are part of a team of people who ensure a company’s success. Shouldn’t they share in that very success?

A Focus on Workers and Their Wages

From the time of Henry Ford, employers have recognized that whether we are talking about wages, benefits, or working conditions, minimizing employment costs is not always the best strategy. In 1914, Ford introduced a $5 per day wage at his automobile factory, doubling the daily wage while simultaneously reducing the average workday from nine hours to eight. His company was rewarded with improved worker productivity, reduced turnover, and higher profit margins. Estimated productivity gains ranged from 30 to 50 percent as a direct result of these increased wages, according to Daniel Raff and Larry Summers. Prior to this wage increase, Ford had some of the worst worker retention and highest absenteeism rates in the industry.

Of course, one can take this principle to an extreme. In 2015, Dan Price, the CEO of a Seattle-based credit card processing company, Gravity, announced that he was going to pay all his employees a minimum of $70,000 a year. The figure was chosen on the basis that it was what most people needed to live comfortably in a city with a notoriously high cost of living. Price was an instant celebrity, flooded with new customers and lauded by grateful employees. However, not everyone was happy. Some employees groused that it wasn’t fair to pay everyone the same regardless of their skills and contributions and some key employees left. Price’s brother, a co-owner, sued on the grounds that he had not been informed and, as a major shareholder, would be hurt in the process. Other businesses in the area worried that he was setting a standard that would bankrupt most of them. At this writing, the company is still in business and appears to be doing fine, but no one should assume that there are no downsides to raising worker pay to such a high level.

Firms have found that investments in employees are not mutually exclusive with profiting. They may even help.

I think the moral of these stories is that employers have a choice to become either what some have called “high-road” or “low-road” employers and others call a stakeholder or a shareholder company. A high-road employer needn’t necessarily go as far as Henry Ford or Dan Price. There are limits and trade-offs, but the point at which better pay and benefits for workers becomes inconsistent with higher profits isn’t probably as clear-cut as your Economics 101 textbook suggests. Further, employers, like the rest of us, are sheep. They pay what the competition pays. When unionized workers successfully raised wages in the 1950s, nonunion firms followed suit. That behavior has economic boundaries, to be sure, but also a socially defined component. Within some ill-defined range, the cost of labor doesn’t determine the ability of a firm to be successful. When unions were stronger than they are now, they bargained for higher wages, and unionized workers earned what economists like to call “rents”—meaning more than they were supposedly worth according to the market. It helped, of course, that bargaining was industry-wide and there was little or no competition from other countries. But the benefits were also clear: middle-class communities with stable families and a broadly based prosperity that created plenty of demand for the goods the workers were producing.

Joseph Blasi and Douglas Kruse at Rutgers and Richard Freeman at Harvard have done extensive research and writing on this topic. Their book, The Citizen’s Share, is essential reading for anyone interested in the topic. They use the phrase “shared capitalism” instead of “stakeholder capitalism” to describe various ways of giving workers a stake in the success of their companies. These include profit sharing, employee ownership, gain sharing, and broad-based stock options. Profit sharing is the most common form of shared capitalism. These shares can be paid out as cash bonuses on a yearly or more frequent basis and depend on company profitability. Worker compensation is thus partly tied to how well the company does and can fluctuate quite a bit as a result. On average, firms provide a cash bonus that is 5 percent of an employee’s annual compensation. In 2016, Ford hourly workers received an average pre-tax profit sharing check of $6,900. Southwest Airlines’s record profits in 2015 allowed it to pay out profits of about 15.6 percent of the average employee’s compensation to more than 49,000 of their employees.

Shared capitalism is much more commonplace than you might think: 47 percent of all for-profit employees participate in some form of it, with the greatest number (37 percent) participating in profit sharing. However, many companies use combinations of shared capitalism—for example, profit sharing to motivate workers in the short term, and stock ownership to motivate them in the longer term.

There are now hundreds of studies on the effects of various forms of shared capitalism, including meta-analyses that attempt to extract some general lessons from all of the individual studies. Typically, the studies look at the output or value added by a firm and try to determine how much difference some form of shared capitalism makes, after holding constant or adjusting for as many other firm differences as possible or by matching similar firms with and without employee ownership or profit sharing. As always, the findings are mixed but suggest the following:

  • The effects on output or productivity are typically positive.
  • Profit sharing has larger effects than employee ownership.
  • Financial gains to workers do not come at the expense of their regular pay; they usually add to it.
  • Firm profits and share prices are either not affected or given a boost.
  • Any positive effects are usually small when averaged across firms. Negative effects are very uncommon.

Most of the studies cited above conclude that, while profit sharing is associated with increased productivity, it is difficult to determine whether profit sharing is the product of increased performance or the cause for it. Most of the studies try hard to adjust for selection or for confounding variables but identifying the effects of profit sharing is difficult. At least one study avoided this problem by using a randomized controlled design, involving 21 fast food franchises. The study found that those randomly assigned to the profit sharing group experienced greater productivity and profitability and reduced employee turnover.

If the evidence is mostly positive, why is profit sharing or some other form of shared capitalism not more widespread? What’s not to like about it? The answer may be simple inertia or a lack of knowledge about its benefits. It may be related to the kind of short-termism and focus on quarterly profits that is dictated by financial markets. And it may simply be ideological, reflecting the strong belief among economists and many business leaders that competition will ensure that both labor and capital markets are efficient. They see widening gaps in incomes and earnings, with the top ranks pulling away from everyone else, as the result of market forces rather than imperfect competition and social norms. With increases in the scale of global markets and the size of corporations, for example, a winner-take-all bidding up of top salaries might have occurred. On the other side of the argument, increasing levels of industry concentration and other noncompetitive forces, along with evolving social norms and close ties between top managers and their boards, may be creating unearned economic rents for executives. A greater role for workers and a broader sharing of such rents need not harm the performance of the economy. In fact, it should improve that performance.

Shared Capitalism and Corporate Tax Reform

Shared capitalism, especially in the form of employee ownership, has had bipartisan support in the past. Conservatives like employee ownership because it gives workers a stake in capitalism, and liberals support it because it reduces inequality. As John Case puts it in The Atlantic, “The left favors spreading the wealth. The right wants to create more capitalists. With employee ownership, they can both get their way.”

The idea of encouraging broad-based capitalism has deep roots in our nation’s history. As Blasi, Freeman, and Kruse note, the founders from Washington to Madison and Jefferson all believed that democracy would not survive and flourish unless property ownership was widespread. In their day, the major form of property was land. The Homestead Act of 1862 and other laws were intended to encourage ownership. Tax preferences for home ownership have a similar aim. But in an industrial or post-industrial era, the major form of property is financial capital. Yet what we are seeing is a growing concentration of such capital in the hands of a very small number of very wealthy people. In 1974, with the strong backing of Senator Russell Long, Congress enacted the Employee Retirement Insurance and Security Act (ERISA). It included tax preferences for the creation of Employee Stock Ownership Plans. Again, the purpose was to spread the wealth. Currently, such plans receive numerous tax benefits. Broad-based profit sharing, which as we have seen, is even more beneficial, has not been given a similar push.

There are any number of ways to address this and, in the process, to create a more worker-friendly form of capitalism. One idea would be to change a provision in the Internal Revenue Code, Section 162(m). The section was initially created to rein in excessive executive pay and was last changed in 1993. It limits corporate income tax deductions for executive pay in publicly held companies to $1 million per year. Covered employees include only the CEO, the CFO, and the three next highest-paid employees. However, the rule comes with certain exclusions, including performance-based compensation. This means that compensation paid to the CEO and other top executives can be deducted from a firm’s corporate income taxes, even if that compensation exceeds $1 million, as long as the board certifies that the extra compensation is performance-based.

Although Section 162(m) was originally intended to limit excessive executive compensation, multiple studies have found that the cap has not had its intended effect. Executive compensation has increased dramatically. Steven Balsam of the Economic Policy Institute estimates that tax deductions for executive compensation cost the U.S. Treasury an estimated $7.5 billion per year.

Section 162(m) could be amended so that corporations receive these deductions only if they offer profit sharing or a share-ownership plan to all (or almost all) of their employees. The proposal is not to disallow performance-based pay, but to require that it be broadened. If a corporation does not wish to broaden performance-based pay to most of its employees, then salaries exceeding $1 million would no longer be deductible.

This option is only one of many that could be considered if and when Congress decides to reform corporate income taxes. And because noncorporate businesses are growing in importance, there may be a role for changes to noncorporate taxes as well if we want to encourage more profit sharing, more training, and other worker-oriented benefits.


Our failure to achieve anything close to broadly based economic growth in the United States is very troubling. As we know, one way to address the issue is by redistributing income from the rich to the poor and middle class. However desirable this may be on equity grounds, this approach cannot, however, be the only, or even the primary way of creating a fairer society. What is needed is a much bigger role for the private sector in sharing the rewards of capitalism with workers. The evidence that this is option compatible with making profits is quite extensive, despite what many might think. If we want more of it, tax reform may be the vehicle to ensure that it becomes the norm and not the exception.

Isabel Sawhill is a Senior Fellow in Economic Studies at the Brookings Institution. She is the author, most recently, of The Forgotten Americans: An Economic Agenda for a Divided Nation.

Also by this author

Saving Democracy

Click to

View Comments

blog comments powered by Disqus