Abstract
There has been considerable media attention recently to the differences when the compensation and wealth of the top 1% of the population are compared with the bottom 90%. The macro analysis was discussed in the best-selling book Capital in the Twenty-First Century, by the French economist Thomas Piketty.
A new analysis by the Pew Research Center (reported in December) found that the typical U.S. “affluent” family in America now has median wealth of $639,000. That is nearly seven times the wealth of a middle-income family ($96,500), and the middle-income family has roughly 10 times the wealth of a lower-income family ($9,300). For a family of three, in this analysis, an income of about $38,000 was the threshold of “middle income.”
Thirty years ago (in 2013 dollars), the gaps were smaller—$318,100, $94,300, and $11,400.
There was a time in the United States when the gap was closing. From 1948 to 1973, the income of the top 1% fell by nearly a third. The income of the bottom 90% in that era increased at an annual rate of nearly 2.9%, a rate that means incomes doubled every 24 years. Significantly, worker productivity and incomes increased at roughly the same rate, which meant workers shared in the productivity gains.
From 1973 to 1995, the income of families in the bottom 90% fell 0.2% annually. It would have been even worse but this was the period when women in large numbers added second incomes. The share of income enjoyed by the top 1% increased from 8% to 14%. (By one analysis, the increase in the value of employer-provided health care raised real incomes by an added 0.2%.)
From 1995 to 2013, the incomes of the bottom 90% increased by 0.2% but at that rate it would take three centuries to double. Their share continued to fall and is now approaching 50%. Median family income in current dollars showed virtually no change. The incomes of the top 1% now account for 19% of the total, which has more than doubled in 40 years.
Significantly baby boomers are retiring in large numbers and their incomes are reduced. Another factor is that the participation of women has plateaued and has actually drifted lower, which affects family incomes. And the final factor is the increase in part-time workers, which is examined in the data discussed later. Those factors of course affect cash incomes. Those trends come together with the strong stock market recovery to increase the gap between the top 1% and the bottom 90%.
That information has been reported previously by the media. Generally, the analyses are based on macro data—that is to say, the analyses are based on aggregate data for the economy and ignore industry, occupation and job level.
Trends in Total Compensation Increases
The table that follows summarizes the changes in total compensation over two 7-year periods, 1999 to 2007 and 2007 to 2014. The first period starts prior to the 2000 recession and ends before the 2008 downturn. The second period captures the Great Recession years.
Through both periods, benefits have increased more rapidly than cash compensation. Both slowed down dramatically since 2008. That is across the board and says a lot about what has happened to workers.
The macro data and analyses miss an important point—the benefits that are important components of total compensation are not universal and provided or paid for by all employers. Coincidentally, Wharton’s Dr. Peter Cappelli published an article in the November issue of Harvard Business Review, “Google Adds Benefits, Walmart Cuts Them; Oddly the Logic is the Same,” that discusses a significant trend: companies increase or decrease benefits to support their business strategy.
His point is not surprising. Companies are competing for talent, and where a company like Google wants to hire and retain the best people in tight labor markets, they make certain their benefit package is attractive. Conversely, a company like Walmart is not as concerned with the capabilities of their new hires but does feel the pressure to reduce costs.
As I discuss below, significant differences in benefits exist across the levels in organizations, across industries and across companies of different sizes.
He argues this trend makes total compensation more unequal than wage data alone suggest. It is highly likely the inequalities will grow over time. This, of course, is happening to the workers who can least afford it. That is consistent with my reason for writing this editorial.
Another very obvious difference is the value of stock-ownership plans, which of course can add millions to an executive’s total compensation. Stock fluctuations make it almost impossible to develop truly comparative analyses. Despite the problems, ignoring stock-related income badly distorts comparisons of the rich and poor. Significantly, government data do not include the income from stock ownership.
The Patterns of Differences
The data discussed below were provided in a special data dump by the U.S. Bureau of Labor Statistics (BLS). Their data are extensive and planned as a defensible sample of U.S. employers (by size, industry and location) but do not include all the details to develop value estimates.
One of the biggest disparities is the difference in benefits provided by smaller employers (defined as under 50 employees) and larger employers (defined as more than 500 employees). There are roughly 4.9 million firms with 50 or fewer employees, and only 18,000 with 500 or more. (The average business in the United States has just over 20 employees.)
There are also significant differences between “goods-producing” and “service-providing” industries. The best benefits generally are in industries dominated by professionals—finance, technology and professional/scientific. The leisure and hospitality, construction and retail industries have the lowest benefits.
Medical Care Insurance
Overall 69% of employers provide medical care insurance covering 50% of the workforce.
Employers with fewer than 50 employees are not required to provide medical care insurance under the Affordable Care Act (ACA). Only 53% offer medical care benefits and 71% of the employees participate.
In contrast, the BLS data show that 89% of larger firms (over 500 employees) offer medical insurance.
Employers offer medical care benefits to 86% of full-time employees but only to 23% of part-time employees.
When employees are distributed by compensation level, only 34% of the lowest-paid 25% have medical care benefits, while 93% of the highest quartile have medical benefits.
Retirement Benefits
Only 65% of workers have access to a retirement plan. The number covered by a defined benefit pension plan is down to 19% while 60% are eligible to participate in a defined contribution plan.
Again larger employers offer benefits at significantly higher levels—46% maintain a defined benefit plan and 80% have a defined contribution plan. Only 7% of smaller employers offer a defined benefit pension and 47% have a defined contribution plan.
Retirement benefits are provided to 74% of full-time employees but only to 37% of part-time people. Defined benefit plans are maintained for only 8% of part-time employees and 22% of full-time employees.
The industry differences are similar. Defined benefit plans are maintained by 26% of goods-producing companies while 17% of service-providing companies maintain the plans. The percentages for defined contribution plans are 71% and 58%. Utilities are an outlier—79% continue to have a defined benefit plan and 92% have a defined contribution plan.
Paid Leave Benefits
Paid leave is rarely a research subject. Moreover, it is implicit in many discussions that leave policies are reasonably consistent across employers. Recently, there have been reports that a surprisingly high percentage of employees do not use the leave time allowed. However, as the BLS data show, employers now spend more on paid leave than on retirement plans and almost as much as they spend on health insurance.
Perhaps, more important, paid leave is hidden in base salaries. Additional leave days increase compensation per hour worked but does not affect a weekly or monthly rate of pay.
Leave policies are by no means universal. Only 76% of employees, for example, are eligible for paid holidays, 61% can use paid sick days, but only 38% have paid personal leave available.
However, the patterns for retirement and medical care insurance prevail here as well. Larger companies have more generous leave policies. Large companies have paid holidays for 92% of their employees while only 67% of small companies do so. Paid sick leave is available to 81% of employees in large companies but only 50% of those workers in small companies. A similar pattern exists for all paid leave benefits.
The differences are even more apparent when full-time and part-time policies are compared. Paid holidays are provided for 90% of full-time employees but only 37% of part-time employees. A different policy, funeral leave, is provided to 71% of full-time employees but only28% of part-time employees.
Summary
The BLS data show it is a mistake to generalize or to make assumptions. It is also a mistake to rely loosely on averages.
With the cost of benefits now in excess of 40% of base pay—based on misleading averages—market pay analyses that ignore benefits can lead to invalid conclusions. It would be advantageous to assemble better survey data on the prevalence and value of benefits.
A related issue is the pressure to increase the minimum wage. The workers who will be affected by any increase are in many companies the same individuals who are seeing their benefits reduced. The loss of benefits should be factored into future research on comparative income levels.
Finally, when baby boomers retire today, they often can count on the income from a vested defined benefit. Increasingly tomorrow’s retirees will have to rely on the funds in a defined contribution plan. The data suggest many will not have adequate funds to sustain their lifestyle. There will be a growing number who decide they cannot afford to retire. For those in lower job levels their only source of income could be Social Security and typically part-time employment income.
At the macro level, these trends may not be sufficiently material to affect the conclusions from analyses similar to Piketty’s. However, for the micro analyses central to assessing a company’s total compensation levels, the trends are directly relevant.
All of that is going to add to the trends that prompted the conern with the growing income gap. In the U.S. Senator Elizabeth Warren has been more or less alone in focusing public attention on the issues that contribute to the gap. But if the gap continues to increase, I’m likely she will build broader supoport for action.