Feature Article: Can the US Successfully Meet the Threat Posed by Worsening Inequality?
Inequality has always been a difficult subject for people to discuss. Yet for eons humanity has struggled with it, and the emotions and behaviors it triggers.
In 1975, Arthur Okun captured this struggle in what has become a timeles book, Equality and Efficiency: The Big Tradeoff. As he summarized, “Institutions in a capitalist democracy prod us to get ahead of our neighbors economically after telling us to stay in line socially. This double standard professes and pursues an egalitarian political and social system while simultaneously generating gaping disparities in economic well-being.”
A lot has changed since 1975. Okun wrote during a period when there was increasing concern with growing economic stagnation. The elections of Margaret Thatcher (1979) and Ronald Reagan (1980) marked the beginning of the neo-liberal response to this with its greater policy focus on markets and economic efficiency. Forty years later, there is abundant evidence that we have once again overshot, and now face a pressing need to renew our focus on equality.
All around us, we see today the negative effects and dangerous threats that emerge when inequality passes a critical threshold.
Economically, excessive inequality is one of the forces that have weakened aggregated demand. For example, in 2018 the top 10% of households accounted for 35% of total income, but just 23% of consumption spending. The next highest 10% received 18% of income, and accounted for 16% of spending. Unsurprisingly, analysts have pointed to the sharp drop in consumption spending by the top 10% after COVID arrived as an important cause of the current economic contraction.
Socially, the negative effects of rising inequality have also been recognized for a very long time. In his 1899 book, The Theory of the Leisure Class, the economist Thorsten Veblen warned of the cancerous impact of what he called “conspicuous consumption” by economic elites, and the feelings of envy and resentment it stirs up among the masses. In recent years, researchers have developed a deeper understanding of the psychology and neuroscience behind this phenomenon.
For example, in “Interdependent Utilities: How Social Ranking Affects Choice Behavior”, Bault et al find a critical difference between the way people approach decisions when the results will be private versus when they will be observed by others. In the private condition, people are risk averse (i.e., possible losses have a higher weight than possible gains), but in the public condition this reverses, and people become risk seeking, to avoid the envy that comes from watching others’ relative outperformance. The difference is that in the public case, the payoff is involves both money and relative social status.
In a recent paper, Claudius Gros shows how “the repercussions of envy are amplified when societies become increasingly competitive” (“Self-Induced Class Stratification In Competitive Societies Of Agents: Nash Stability In The Presence Of Envy”. See also, “Investigating the Health Consequences for White Americans Who Believe White
Americans Are Wealthy”, by Cooley et al).
Many have argued that this phenomenon has driven the accumulation of debt by the middle class, as they have tried to keep up with the increasingly affluent Joneses. This process has been supercharged in recent years, as the Joneses, egged on by aspirational (envy-based) marketing, have engaged in something resembling competitive conspicuous consumption, which social and other media have made painfully visible to everyone else in the world.
Moreover, as Jack Goldstone and Peter Turchin have shown (in their development of Structural Demographic Theory), increasing inequality within elites is another source of social conflict that has worsened in recent years (see, “The Wealthy and Privileged Can Revolt Too”, by Noah Smith; “Our Suicidal Elites” by Joel Kotkin, “The Discrete Terror of the American Bourgeoisie” by Ed Luce, and for a French view, “Jupiter Falls to Earth”, by Christophe Guilluy).
Political shocks like Brexit and the election of Donald Trump have their roots in the economic struggles, social resentments, and class anger that have built up as inequality has gradually worsened since the 1970s.
In sum, there is no doubt that inequality has already had increasingly negative effects. That is why we included it as one of two critical medium term economic uncertainties in our May 2019 feature article on multipath reasoning about the future (“Multipath Analysis: A Systematic Process for Reducing the Dimensionality of Global Macro Forecasting Challenges”).
However, before discussing what can be done to reverse the worsening effects we see today, we first have to dig into the causes of inequality, some of which may surprise you.
Typically, conversations (and arguments) about the causes of inequality divide into two categories: Those rooted in individual decisions, and those rooted in the nature of “the system” however it is defined. But before delving into both of those, let’s start the cause that is almost never discussed: the impact of randomness (or, if you will, luck).
Let’s assume a world of perfect initial equality, where 50 people each start out with the same endowments of talent, resources, and opportunity. For ease in modeling, we’ll proxy this by (1) everyone starting with an equal endowment of $1,000, and (2) everyone facing an equal chance of their capital earning an annual return of between (20%) and 20%, with all values between those extremes equally likely. Each individual’s annual return is determined by a unique random number generator.
We ran this simulation over 20 years, fifteen times. What do you think we found?
At the end of 20 years, the median multiple (over our fifteen simulations) of highest of the 50 individuals’ resources compared to the lowest was 11x. Across the fifteen scenarios, the highest multiple value was 21x and the lowest was 6x.
What about shares of the total ending resources for all 50 individuals? The top 20% (i.e., the top 10/50) share had a median value of 38%. Across the fifteen scenarios, the highest was 43% and the lowest was 34%.
What about the top 10%? Their median share of total resources was 22%. The highest was 26% and the lowest 20%.
Looking at the top 20%, the median multiple between the top 10%’s ending share and the next 10%’s ending share was 1.4x, with a high of 1.6x and a low of 1.3x. Even within the top 20%, randomness generates inequality.
Finally, after 20 years, what percent of the 50 individuals ended up with less than their initial $1,000 endowment? The median was 60%, with a high of 66% and a low of 50%. This was caused by strings of negative returns from which they never recovered.
In this surprising result, we see a familiar investment lesson: Losses have a larger negative impact on long-term results than gains. If you start with $1,000, make 20% the first year ($1,200), then lose 20% the second year ($960), you have to earn 25% in year three just to get back to $1,200. (Unfortunately, the power of envy causes too many investors to spend far less time on managing risks than finding ways to show higher short-term returns than their peers).
This simple model highlights a critical point: Randomness (luck) has a much larger impact on the emergence of inequality than most of us realize. In fact, because of the cognitive dissonance it creates, we subconsciously refuse to acknowledge, much less admit it.
Now let’s move on to the more familiar roles of personal decisions and the nature of “the system.”
In a January 2020 report, Pew Research found that 60% of Republicans believed that different life choices contributed a great deal to inequality. Only 27% of Democrats agreed (overall, 42% agreed). In contrast, 62% of Republicans agreed that major changes in the system were needed to reduce inequality, as did 88% of Democrats (81% overall). See: “Most Americans Say There is Too Much Economic Inequality in the U.S.”
Perhaps the best-known example of the impact of personal decisions on inequality is Isabel Sawhill and Ron Haskin’s “Success Sequence” – “at least finish high school, get a full-time job and wait until age 21 to get married and have children…Only 2.4 percent of Americans who follow the success sequence live below the poverty line, while over 70 percent enjoy at least middle-class incomes, defined as 300 percent of the poverty line or more. For Americans who—for a host of reasons—don’t follow the sequence, the picture is reversed” (see their paper, “Work and Marriage: The Way to End Poverty and Welfare”. Also, “The Millennial Success Sequence: Marriage, Kids, And The ‘Success Sequence’ Among Young Adults”, by Wang and Wilcox).
Sawhill and Haskin’s research was based on US evidence. Other researchers have arrived at similar conclusions using evidence from other countries. For example, using Canadian data, Christopher Sarlo found that the Success Sequence factors had a strong impact on life outcomes, along with two other decisions: avoiding both addiction and a criminal record (“The Causes of Poverty”).
In terms of our simple model, good decisions can affect both starting resource endowments as well as the distributions of possible annual returns on those endowments that different individuals face.
However, there is also plenty of evidence that system factors (broadly construed) have evolved since Sawhill and Haskins published their paper in 2003, and now have a stronger impact on both individuals and aggregate inequality. In our model, these system factors affect both starting resource endowments and the distributions of possible annual returns facing different individuals.
Consider the Success Sequence. The work of many researchers has made clear that the quality of education varies widely across American school districts, as well as across schools and classrooms within them. And unlike many other countries, the United States severely limits’ parents’ ability to choose better schools for their children.
The quality of university and technical education programs also varies widely, and over the last 50 years the price of the former has sharply risen in real terms.
Moreover, access to these programs is not a level playing field. For example, in “Legacy and Athlete Preferences at Harvard”, Arcidiacono et al “examine the preferences Harvard gives for recruited athletes, legacies, those on the dean’s interest list [i.e., applicants related to large donors], and children of faculty and staff (ALDC). Among white admits, over 43% are in these categories. Among admits who are African American, Asian American, and Hispanic, the share is less than 16% each.” Their model of admissions “shows that roughly three quarters of white ALDC admits would have been rejected if they had been treated as typical white applicants.”
Rising inequality is an emergent result of multiple interacting causes. For example, a poor education system creates an increasingly severe shortage of workers with the skills required to use leading edge technologies as the transition to an economy driven by brains rather than brawn accelerates. Companies that can attract these workers strengthen their competitive advantage over other companies, which enables them to pay higher wages, limiting their competitors’ ability to do so. Research has shown that this is yet another driver of increasing inequality.
This is but one aspect of the challenges associated with getting a good full-time job in today’s economy. A combination of increased use of automation technologies and moving production and supply chains to other countries has eliminated many middle income jobs in the United States, leaving a mix of high income opportunities (that require high quality education and training), and much less secure and poorly paid service and other gig economy jobs. Many of those jobs do not provide health care benefits, whose real cost has also dramatically increased over the past 50 years.
Also, in recent years three fundamental aspects of the US economic system have changed in a way that worsens inequality. First, the rate of labor productivity growth has significantly declined. Second, in many industries, the traditional link between increasing productivity and rising real wages has been broken. Third, labor’s overall share of GDP has shrunk in recent years, for a variety of reasons (see, “Declining Worker Power and American Economic Performance” by Stansbury and Summers, “Labour Share in G20 Economies” by the International Labour Organization, and “A New Look at the Declining Labor Share of Income in the United States” by the McKinsey Global Institute).
These structural changes have been so strong in their impact that they have weakened the link between education and middle class incomes (see, “Education and the Dynamics of Middle Class Status” by Hardy and Marcotte, and “Is College Still Worth It? It’s Complicated” by Emmons et al from the Federal Reserve Bank of St. Louis).
Looking at Sarlo’s two additional Success Sequence factors (avoiding addiction and a criminal record), America’s systems for addressing mental health and addiction issues vary widely in quality, and there is abundant evidence that the criminal justice system treats some groups more harshly than others.
In sum, while making good individual decisions can certainly help to reduce rates of poverty and the extent of inequality, system factors have increasingly reduced their potential positive impact on inequality.
So what should be done to reduce inequality, and how likely is that to happen?
In the immediate term, the sharp income reductions caused by COVID that have disproportionately affected people with lower incomes can be offset with government aid, including direct cash payments, support for food banks, and expansions of eligibility for government provided or subsidized health care (i.e., Medicate and health insurance exchanges). Housing is an equally critical challenge with fewer obvious solutions. While current bans on evictions can be extended, this has knock-on negative effects on landlords’ ability to service their debts. An expedited bankruptcy process is the likely solution.
In the short-term, the challenges to reducing inequality seem even more politically daunting.
Improvements to the education and workforce training systems (or, more broadly, the human capital system) are obviously critical to increasing productivity, but they have repeatedly been met with fierce resistance (e.g., by K-12 teachers unions, colleges and universities, and the large number of competing non-profits focused on worker retraining). While there are proposals to break this logjam (e.g., moving to a national system of stackable certified competencies, similar to the European National Qualifications Framework, and linking participation in a federal jobs program to completing worker retraining), the weight of historical evidence suggests that no matter how critical it is, progress in this area will very likely remain glacial, absent much more aggressive and sustained demands for change by the business community.
Improvements to the US healthcare and social care systems, including processes and institutions focused on mental health and addiction treatment, will also very likely face strong resistance, in addition to squeezed post-COVID budgets. Again, the weight of historical evidence suggests progress will be slow here too, despite polling data showing that reducing health care costs is a critical issue to voters across the political spectrum.
On the other hand, programs that have already been shown to be effective at boosting incomes, like the Earned Income Tax Credit, could be expanded more easily, potentially to include federal wage subsidies. These are likely to be more effective and meet less resistance than attempts to increase the minimum wage, which will put further stress on many businesses that are already barely holding on in the post-COVID economy (e.g., “People versus Machines: The Impact of Minimum Wages on Automatable Jobs” by Lordan and Neumark).
While broad changes to the US education and healthcare systems seem very unlikely at this point, narrower solutions may garner more support and still have a significant impact, particularly integrated approaches to support lower income children early in their lives. As Currie and Rossin-Slater find in their paper “Early-life Origins of Lifecycle Well-being: Research and Policy Implications”, “there is a robust and economically meaningful relationship between early-life conditions and well-being throughout the lifecycle, as measured by adult health, educational attainment, labor market attachment, and other indicators of socio-economic status.”
“However, there is some variation in the degree to which current policies in the U.S. are effective in improving early-life conditions. Among existing programs, some of the most effective are the Special Supplemental Program for Women, Infants, and Children (WIC), home visiting with nurse practitioners, and high-quality, center-based early childhood care and education. In contrast, the evidence on other policies such as prenatal care and family leave is more mixed and limited.”
Changes to the tax system are also critical to reducing inequality. They could include not only higher marginal rates on top incomes, but also and end to provisions like interest deduction (which promotes use of debt in capital structures), taxation of private equity funds carried interest as capital gains rather than income, and imposition of a wealth tax. Two other ideas may have an even greater impact – a carbon tax, with a portion of the revenue rebated to low income taxpayers or used to provide equity to start-up businesses, and a progressive consumption tax that discourages destructive conspicuous consumption.
Structural changes that increase the bargaining power of labor relative to capital could also contribute to the reduction of inequality. The reshoring of supply chains to make them more resilient (and in response to worsening US-China relations) will help. So too would changes that strengthen collective bargaining processes in the private sector. Placing restrictions on companies repurchasing their own equity, or making such buybacks contingent on wage increases could also help. Most difficult of all would be a switch in the US immigration system to one like Canada’s, that gives preference to talented applicants who can make the largest contribution to increasing national productivity growth, in contrast to low skilled immigrants who put downward pressure on low income workers’ wages (an argument which in years past the Democratic Party strongly supported).
After reading this list, it is clear that we do not lack for testable policy ideas for limiting the worsening inequality that has emerged from the interaction of bad decisions, bad luck, bad systems, and now the COVID-19 pandemic.
Moreover, we have at hand a growing number of analyses of how we arrived at today’s predicament, and how we might get out of it (e.g., “Capitalism and the Future of Democracy” by Isabel Sawhill, and “The Economics of Belonging”, by Martin Sandbu).
More darkly, in the growing influence of populist authoritarians on both the left and the right, we are also witnessing the accumulation of painful – and scary -- evidence of the likely consequences if worsening inequality is not reversed.
Yet in our current political climate – even after COVID’s shocking arrival – it will very likely be difficult, if not impossible, to marshal enough political support to enact and effectively implement the reforms needed to do this.
If there is a reason to be hopeful, it rather perversely lies with Xi Jinping. It may well be that only the growing danger and demands of a new Cold War will finally force us to reverse the downward path we are on today.