Nonwage benefits have come to play an increasingly crucial role for workers in recent years. Roughly 40 years ago, these benefits constituted a mere 5% of the average worker’s compensation package; as of 2021, they make up roughly 30%. In addition, provisions such as sick leave have become increasingly relied upon in the wake of the COVID-19 pandemic, as workers return to offices and navigate rampant new strains of the virus. These benefits represent a progressively large component of total compensation – and continue to become ever more essential to employers and employees alike.
Despite this growth in importance, however, benefits have received considerably less attention than pure wages in both academic and popular discussions. For evidence of this disparity, look no further than the ways in which discussions around worker compensation are framed: Topics such as income inequality, the gender pay gap and CEO compensation all focus on salary rather than benefits.
A new paper by UNC Kenan-Flagler Business School Professor Paige Ouimet and University of Maryland Smith School of Business Professor Geoffrey Tate aims to fill this gap by looking specifically at three of the most important nonwage benefits: health insurance, retirement benefits and paid leave. The authors find greater variation across firms in average benefits as compared to average wages; by contrast, within a given firm, benefit levels vary far less than wages across employees. These divergent patterns in benefits and wages demonstrate that looking at wages alone fails to capture the full picture of how total compensation is distributed among workers.
In addition, the paper also finds a positive correlation between wages and benefits, meaning that the amount of benefits that workers receive increases with their income. High earners also reap generous benefit packages, while low-wage workers receive far lower benefits, if any.
The paper also finds that the amount of benefits offered by a firm is heavily affected by the specific industry and location in which that firm operates, as well as by the overall composition of its workforce. The authors conclude with a discussion of implications for worker turnover and firm hiring practices. Together, these findings demonstrate how the provision of nonwage benefits alters labor market dynamics and creates new ones, holding tremendous significance for our collective understanding of how workers and firms operate in the modern environment.
In three separate analyses of health insurance, retirement benefits and paid leave, Ouimet and Tate examine total variation in the benefits and wages received by workers. They then break down how much of the variation can be attributed to differences across firms as compared with differences within firms. The data used covers all full-time employees from 2004 to 2014, save for those employed in the public, military and agriculture sectors, as well as unpaid and overseas workers.
Differences among firms account for 86.3% of the total variation in health insurance, 86% of the total variation in retirement benefits and 72.8% of the total variation in paid leave. By contrast, differences among firms account for only 59% of the total variation in wages. Put more simply, the firm at which workers are employed has much greater bearing on the level of benefits they receive than it does on their pure wage level.
While differing wages within a firm account for 41% of the total variation in wages, differing benefits within a firm account only for 13.7% of the total variation in health insurance, 14% of the total variation in retirement benefits and 27.2% of the variation in leave. These results, which continue to appear even after controlling for a variety of potential confounding factors, demonstrate the tendency for firms to standardize the benefits they offer to their entire workforce. Ouimet and Tate hypothesize that this is largely a result of regulations that limit an employer’s ability to offer different benefit levels across their workforce. These policies and their implications are discussed in greater detail below.
For the second stage of their analysis, Ouimet and Tate turn their attention to other possible factors that may impact the level of benefits a firm offers. They hypothesize that the presence of high-wage workers – who are more fiercely pursued by companies and have more bargaining power – will cause firms employing such workers to offer more generous benefits to these employees but also across the firm because of the arguments above. As such, low-wage workers in the same firm will, on average, receive more generous benefit packages. Indeed, the paper finds that the presence of high-wage workers elsewhere in the firm leads to an estimated 11.5% increase in benefits.
As the authors note, nonwage benefits that are standardized within a firm hold roughly the same financial value for all workers, so they make up a larger percentage of low-wage workers’ total compensation and are thus especially important for this subset of employees. One additional implication of this finding involves discussions around inequality, which has become an economic buzzword in recent years. By leaving out nonwage benefits, their growing share of all compensation packages and their outsized importance to low-wage workers, current measures of inequality may actually understate the true gap in compensation. This especially holds true given the positive correlation between wages and benefits observed in the data. Since higher-paid workers also receive higher benefit packages, inequality is likely underestimated when nonwage benefits are not taken into consideration.
When considering the impact of nonwage benefits on both firms and employees, it’s especially important to consider the unintended consequences resulting from current policies around these benefits. The Health Insurance Portability and Accountability Act mandates that firms provide the same health insurance offerings to all “similarly situated” workers that do not vary by a broad, employment-based categorization, such as geographic location or full-time vs. part-time status. Similarly, the Employment Retirement Income Security Act prevents firms from offering disproportionately greater retirement benefits to its highest-paid workers. Paid leave, by contrast, lacks the same mandatory standards as health or retirement benefits.
As a response to these regulations, firms tend to standardize their benefit offerings across their employees, thus reducing the variation in benefits received by workers within a given company. Alternatively, firms may offer uniform sets of benefits to employees across multiple wage levels to minimize their own administrative burden or prevent any ill will among their workforce that different benefit levels might cause.
The paper’s findings indicate that this standardization may produce both positive and negative effects. As firms compete over higher-skilled workers, they must offer benefit plans that are attractive to that tier of workers – and, because of the standardization practices outlined above, must also offer those plans to lower-skilled workers within that tier. Thus, lower-skilled workers may benefit from these policies, which force a firm to offer all employees the benefits it offers the highest paid within its ranks.
Indeed, while greater benefits levels are strongly associated with decreased employee turnover, the effect is most pronounced among low-wage workers. Ouimet and Tate find that increasing the level of benefits by one standard deviation decreases turnover among low-wage workers by 0.6 percentage points, compared with a decrease of 0.2 percentage points among high-wage workers. In addition, firms that provide more generous benefits appear to shed low-wage workers. The firm may accomplish this by automating the tasks usually performed by low-wage workers or by outsourcing that work to another company. While further research is needed to more precisely decompose the mechanisms by which this occurs, the data clearly demonstrate that firms move away from the workers at the lower end of their payroll – concurrently showing how these most vulnerable workers may be harmed by the status quo.
The takeaway is not just that benefits are factoring ever more heavily into the behavior of firms and employees. It also means that as these benefits become larger components of compensation packages, the dynamics they generate – as well as the policies and regulations governing their provision – will similarly increase in importance.