Report | Health

Paid sick days: Measuring the small cost for New York City businesses

Policy Memorandum #194

 Download PDF
 Press release

Share this page:

For more than two years, the New York City Council has been considering legislation to require employers to provide paid sick days to employees. During that time, several proposals to achieve this goal have been considered. Nearly 40 members of the council have signaled their support by signing on as co-sponsors of a paid sick days bill. Thus far, Speaker Christine Quinn has not called a vote on the bill. Mayor Michael Bloomberg is opposed to the proposal (Chen 2012). This policy memorandum is based on an illustrative bill providing five paid sick days to both hourly and salaried employees (details may vary in the final legislation put forward).

Proponents of paid sick days legislation say it would provide job and income security, particularly for low-wage workers, and reduce public health risks arising from the spread of illnesses to consumers and vulnerable populations. Opponents argue it would be costly for New York City employers and could lead to job reductions at a time when the city is struggling to increase employment. One such opponent, The Partnership for New York City (2010), made this case in 2010, but its findings were based on an unscientific sampling of businesses and a misinterpretation of the proposed legislation.1

In this policy memorandum, we examine the cost of providing five paid sick days to New York City employees in a range of industries. This paper provides concrete economic data (based on the U.S. Census Bureau’s Economic Census) to help policymakers better understand the potential costs incurred through implementation of paid sick days legislation. Specifically, it provides information on the relative size of potential costs as compared with average sales2 for New York City firms by industry. The data clearly show that the potential cost is in fact extremely small relative to the total sales of a firm. In addition, available research shows potential savings for employers that provide paid sick days, largely resulting from reduced employee turnover.

Furthermore, by demonstrating that potential costs associated with implementation are very small as a share of sales, this paper provides a clearer picture of how employers might adjust to or absorb such costs. While there are a variety of strategies employers might choose to comply with paid sick days legislation, the bottom line is that costs would be so low that compliance could easily be achieved through very modest adjustments to other areas of compensation or to prices, without reducing employment.

This memo’s key findings include:

  • If all employees used all five paid sick days, the cost to an employer that currently provides no paid sick days ranges by sector from between 0.12 percent to 0.92 percent of sales.
  • Among workers who currently have access to exactly five paid sick days, the number of days taken across industrial sectors ranges from 1.5 to 3.0 days. If employees used the average number of paid sick days for each sector, the total cost for firms currently providing no paid sick leave would range from 0.06 percent to 0.54 percent of sales.
  • For the many New York City employers already providing five or more paid sick days, there would be no additional cost.
  • Because many employers already provide five or more paid sick days, legislating paid sick days would level the playing field within New York City.

For firms that currently offer no paid leave, the cost of providing paid sick days is low

Many firms in New York City already provide paid sick leave benefits that meet or exceed the five days of sick time anticipated by this policy memorandum. For those that do not currently provide such benefits, data from the U.S. Census Bureau’s Economic Census can be used to calculate the cost to employers across a variety of industrial sectors of providing sick leave.3

Under the paid sick days model outlined here, an employee would earn one hour of paid sick time for every 40 hours worked, up to a maximum of 40 hours per year. The maximum cost to an employer that currently provides no sick time would be the cost of paying for an additional 40 hours of wages for each employee.

Table 1 presents the estimated employer costs based on two scenarios of employee sick day usage. The maximum usage scenario (the second column of data) is based on maximum possible use of paid sick days by every employee of the company (i.e., all employees accrue and use the full 40 hours of paid sick leave).4 Across all industries, the average cost for all employers that do not currently provide paid sick days would range from 0.12 percent of sales in wholesale trade to 0.92 percent of sales in administrative services.5

Table 1 Table 1 (continued)

Estimated cost of implementing five paid sick days for employers currently providing no paid sick leave, as share of total sales, by sector

Industry sector* Payroll as share of employer sales** Cost of maximum usage scenario (40 hours of sick leave), as share of total sales Average use (in days) of sick leave (given 40 hours of sick leave) Cost of average usage by sector, as share of total sales
Wholesale trade 6.5% 0.12% 2.36 0.06%
Retail trade 10.3 0.20 2.59 0.10
Real estate and rental and leasing 15.4 0.30 3.01 0.18
Administrative and support and waste management and remediation services*** 47.8 0.92 2.92 0.54
Health care and social assistance 42.3 0.81 2.53 0.41
Arts, entertainment, and recreation 30.0 0.58 2.79 0.32
Accommodation and food services 27.2 0.52 1.50 0.16
Other services (except public administration) 17.9 0.34 2.70 0.19

* Industry sectors lacking data on payroll and sales excluded from analysis here include: utilities, information, finance and insurance, management of companies and enterprises, construction, educational services, and professional, scientific, and technical services.

** "Sales" refers to "employer sales, shipments, receipts, revenue, or business done," as defined in the U.S. Census Bureau’s Economic Census.

*** This broad sector includes general administrative support services such as office administration and security, as well as services specifically related to waste management and remediation.

Source: Authors' analysis of U.S. Census Bureau (2007) and Miller (2011)

While this scenario provides a useful “upper limit” figure, it likely overstates the real cost because several factors would probably reduce overall sick day usage. Three such factors include:

• Eligibility restrictions: The maximum usage scenario likely overstates the actual cost because it assumes that all employees would be eligible to use all 40 hours of paid leave per year. Most paid sick days bills include probationary periods of several months for new employees. At any given time, some of an employer’s workers are likely to be within their probationary period and therefore not yet eligible to use any paid sick time, making maximum usage by all employees impossible.

• Employees already receiving paid sick leave: In most firms some employees (particularly those in management or salaried positions) already receive paid sick leave benefits. Those employers with employees already receiving paid sick leave of five or more days would incur no additional costs for these employees under the sick leave legislation modeled in this paper.

• Availability of other forms of leave time: Most proposals for paid sick time allow employers to count other existing forms of paid time off toward their required paid sick leave allotment if they allow such time to be used when workers are sick or tending to sick family members. Firms providing such paid time off would not need to add any additional days off nor redesignate existing time off as sick time, and thus would incur no additional costs.

Most importantly, actual costs would be lower than those presented in the maximum usage scenario because, as data from the National Health Interview Survey (NHIS) indicate, employees who do have paid sick days tend to use substantially less than all of their available sick time. As shown in the third column of data, among workers who currently have access to exactly five paid sick days, the number of days taken ranges across industrial sectors from 1.5 to 3.0 days.

The average usage by sector scenario (the fourth column of data) reflects the estimated cost of providing paid sick days using this sector-specific data from the NHIS on actual paid sick day usage (data are drawn from Miller 2011). Relying on data for employees with access to exactly five paid sick days, the total cost of providing sick days to employees who currently do not receive them drops even lower under this scenario. This is because it reflects that the average employee in a given sector would not use all of her available sick days. Across all industries for which data are available, the cost falls dramatically (from a low of 0.06 percent of sales in wholesale trade to 0.54 percent in administrative services). Note that the figures under the average usage by sector scenario still overestimate the costs due to the three factors outlined previously (eligibility restrictions, employees already receiving paid sick leave, and availability of other forms of leave time). Therefore, this calculation of cost to employers still represents a conservative estimate in the sense that it overstates costs. Nonetheless, it is the most accurate estimate available because it is based on reliable empirical data on the use of sick leave benefits by employees.

Reduced “presenteeism” and turnover: Two of the unexplored but important savings from implementing paid sick days

The data on employer costs associated with the provision of paid sick days show that the costs are smaller than is widely assumed. Less attention has been paid to the potential for employer savings from the provision of paid sick days. A growing body of research attempts to measure employer operating costs associated with the absence of sick leave policies. While this approach is less direct and these costs are harder to measure, it is clear that employers can enjoy substantial savings by implementing paid sick day policies.


“Presenteeism”—the opposite of absenteeism—is gaining recognition as a drain on productivity, and businesses are exploring ways to mitigate its costs (Hemp 2004). Presenteeism is defined as productivity loss from workers who show up to work with real health problems; it does not include workers who might pretend to be ill to avoid working to their full capacity or workers who simply are unproductive at work. Workers without paid sick time are more likely to go to work sick (Smith and Kim 2010), particularly if they are among the millions of low-income workers who lack paid sick time but need a weekly paycheck to pay their bills. Employers bear the cost of the lost productivity that results—a cost that may well exceed the cost of providing paid sick time.

For individual employers, measuring the costs associated with presenteeism can be challenging. Sick workers may try to hide their illness out of fear of punishment, and a small decrease in productivity is difficult to detect. The costs vary depending on the nature of an employee’s ailment and its impact on his or her performance; however, research shows that these costs are real and can be quite large (Hemp 2004).


Businesses incur sizable costs from job turnover among employees without paid sick time. Poor working conditions and environments that do not support work-family obligations can increase job turnover and ultimately disrupt labor supply (Josephson et al. 2008). Job turnover and rehiring and training are costly for employers, and there is considerable evidence that good working conditions and health benefits—including paid sick days—can mitigate job turnover. Employers that offer more generous health benefits to workers are known to have fewer employees quit and lower turnover costs. Cooper and Monheit (1993) find that having paid sick time reduces voluntary job mobility by 5.6 percentage points for married men and 3.6 percentage points for married women.

The effect of providing paid sick leave on worker turnover and its associated costs turns out to be quite large. Paid sick time may increase worker loyalty to an employer and reduce employers’ hiring and training costs, which are typically high in low-wage industries (Lovell 2005). Lovell finds that employers’ savings from reduced turnover are substantial; nationally, paid sick days can reduce employer costs by about $25 billion annually. Clearly, the status quo—presenteeism in the workplace—is not free of costs; research suggests the relative costs of providing paid sick time are less than the cost of doing nothing (Lovell 2005).

Further research on the savings to employers from reduced presenteeism and reduced turnover would be valuable.

Cost of providing paid sick days is miniscule for New York City employers and industries as a whole

Given the low cost of providing paid sick days, it is not surprising that many New York City employers already provide at least five paid sick days. Debate about paid sick leave legislation’s full impact on the New York City economy must consider the substantial share of employees in the city who already receive adequate sick leave benefits. For employers that already provide their workers at least five paid sick days, there would be no cost increase at all.

Companies will successfully adjust to required paid sick days

Given the very small additional cost likely to confront most employers, adjustment need not be burdensome. Successful businesses learn how to absorb unexpected increases in the cost of energy or health insurance—cost increases that account for a much larger share of sales.

Because it would apply to all firms and all sectors, paid sick day legislation would actually help level the playing field within New York City. Firms would choose among adjusting through prices, wages, or other forms of compensation, knowing that their competitors are weighing the same choices. For firms competing across state lines (though employers would find similar requirements in Connecticut),6 the considerable savings from reduced turnover and from fewer unhealthy workers refilling the salad bar or otherwise interacting with the public should substantially offset the fractional increase in labor costs as a share of total sales. Importantly, many workers who would be covered under the legislation modeled in this paper are employed in service-sector jobs that are far less vulnerable to interstate competition.

More significant, the overall economic impact of implementing paid sick leave is negligible. The estimated costs for employees are very low; to the extent that there are costs to employers, they are offset considerably by benefits stemming from healthier, more productive employees.


New York City has good reason to take pride in its relatively high share of employers that provide paid sick days. By adopting legislation that further extends paid sick days to its workers, New York City can build on its well-founded reputation for having a highly educated, healthy, and productive workforce. Without such regulation, policymakers and citizens can expect a continued “race to the bottom” wherein competitive pressures cause employers to reduce benefits that improve the health of employees and, in this case, the health of the public.

Instead, establishing a minimum standard for the provision of paid sick days would level the playing field among employers within the city and improve the health of families and workplaces. The empirical evidence shows that the costs of such a policy for New York City’s employers would be extremely low, and therefore easily absorbed through very small adjustments without a measurable impact on employment.

The authors would like to thank Kevin Miller, formerly with the Institute for Women’s Policy Research, for technical assistance with NHIS analyses of sector-specific sick day usage.


1. Kevin Miller (2012) provides a more complete critique of the group’s methods and results.

2. “Sales” refers here and throughout the paper to “employer sales, shipments, receipts, revenue, or business done,” as defined in the U.S. Census Bureau’s Economic Census.

3. Complete data are not available for the following sectors: utilities, information, finance and insurance, management of companies and enterprises, construction, educational services, and professional, scientific, and technical services.

4. This scenario assumes employees work 40 hours per week and 52 weeks per year.

5. This paper assumes that employer practices for replacing absent employees are the same for paid and unpaid leave.

6. See Hall and Gould (2011) for an analysis of the Connecticut paid sick days legislation.


Chen, David W. 2012. “Weakened, Council Bill on Sick Days Gains Favor.” New York Times, Sept. 27.

Cooper, Philip, and Alan Monheit. 1993. “Does Employment-related Health Insurance Inhibit Job Mobility?” Inquiry, vol. 30, no. 4 (winter), pp. 400–416.

Hall, Doug, and Elise Gould. 2011. “Paid Sick Days: Measuring the Small Costs for Connecticut Businesses.” Economic Policy Institute, Policy Memo No. 177.

Hemp, Paul. 2004. “Presenteeism: At Work – But Out of It.” Harvard Business Review, Oct. 2004, reprint R0410B.

Josephson, Malin, Per Lindberg, Margaretha Voss, Lars Alfredsson, and Eva Vingård. 2008. “The Same Factors Influence Job Turnover and Long Spells of Sick Leave: A 3-year Follow-up of Swedish Nurses. The European Journal of Public Health Advance Access, vol. 18, pp 380–385, doi:10.1093/eurpub/ckn009.

Lovell, Vicky. 2005. Valuing Good Health: An Estimate of Costs and Savings for the Healthy Families Act. Institute for Women’s Policy Research, publication no. B248.

Miller, Kevin. 2011. Analysis of Centers for Disease Control and Prevention’s 2008 National Health Interview Survey [unpublished]. Institute for Women’s Policy Research.

Miller, Kevin. 2012. Response to Partnership for New York City Report on Paid Sick Days Law. Institute for Women’s Policy Research.

Partnership for New York City. 2010. Impact of Paid Sick Leave on NYC Business: A Survey of New York City Employers. New York City Investment Fund Report.

Smith, Tom, and Jibum Kim. 2010. “Paid Sick Days: Attitudes and Experiences.” Report prepared by the National Opinion Research Center for the Public Welfare Foundation.

U.S. Census Bureau. 2007. 2007 Economic Census, Table EC0700A1, “All Sectors: Geographic Area Series: Economy-Wide Key Statistics: 2007.”

See related work on Unions and Labor Standards | Health

See more work by Elise Gould and Doug Hall