Rising Health Care Costs -

¶ … Rising Health Care Costs - Everyone Pays

Usually, when rising health care costs are discussed, it is the employees or the end users that are the ones caught up in the price. However, from an employer's point-of-view these costs can also be very problematic. For this reason, it is important to look at the issue from a labor economics standpoint. Some of the information regarding this issue will be presented in table form throughout the paper in order to show as much information as possible in a small amount of space and allow for a thorough discussion of the issue at hand.

Within the field of labor economics it is important to note that unions have a generally positive influence on both the wages that individuals receive and the fringe benefits such as health insurance. There has been a lot of literature written on the difference in wages between union and non-union employees, but in more recent years some of the interest has shifted slightly to the difference in fringe benefits between union and non-union members. According to much of this new literature, employers that are unionized provided benefits such as pensions and health insurance much more often than those that are non-union. During the decade of the 1980s, health insurance and other fringe benefits got a lot more attention in literature on economic issues mostly because money wages had somewhat declined when looked at as a proportion of the total amount of compensation for employees.

For example, between the years 1966 and 1977, fringe benefits grew at a rate that was much faster than wages in the private economy. By 1977, health insurance and other fringe benefits represented over 23% of the compensation for employees. By 1994 it had peaked at almost 29%. Since that time, it has declined only slightly to just over 27%. Many studies have looked at what type of impact unionization has on health insurance and other fringe benefits but most of the data has been qualitative. In other words, economists have looked at estimating what type of impact unionization has on the probability that health insurance and pensions will be provided by employers.

Using data from population surveys, researchers found that approximately 50% of the gap in pension coverage between union and non-union individuals and approximately one-third to two-thirds of the union to non-union gap in health insurance coverage can be explained by labor market characteristics that are atypical and belong largely to unionized workers. These include longer work weeks, larger establishment size, and higher wages. In a study performed slightly later than that, researchers found that, when they control for the size of the firm, having a workplace that was unionized increased significantly the probability that both health insurance coverage and pensions would be provided.

Between the years of 1983 and 1988, however, the probability of employers providing health insurance coverage declined. This was largely believed to be based on the fact that benefits had changed somewhat over time and the costs to employers had risen significantly. Because of this employers must find ways to deal with the issue of health insurance costs and they often do this by limiting the coverage that they provide or limiting which employees can receive coverage. Often, part-time employees cannot receive health insurance coverage and do not qualify for it through their employers. This has not really changed that much over time but the amount of part-time employees that many companies have now has changed over time in order to battle some of the health care costs. Overall, the requirements for an individual to qualify for health insurance have not changed significantly over time and the benefits have not changed all that much although many of the benefits have gotten better and more comprehensive if the money is there to pay for them.

Unionization has such a large impact on fringe benefits that it is very important to discuss it. Many of the methods that were utilized to determine this impact have been re-examined in recent years but the earlier findings have not generally been challenged too strongly. Some authors have criticized studies that have been done previously because they did not take into account the simultaneous issue of fringe benefits and wages when using either probit or logic equations in order to help estimate the impact that unions had on the probability of health insurance coverage. However, despite these criticisms it has generally still been found that the likelihood of health insurance coverage and pensions are both increased if a union is present. Most studies have looked at the probability of various fringe benefits being provided by employers and only a few of the studies have looked at whether having a union actually raises the expenditure levels that employers have for these same fringe benefits.

In 1984 a study done using data from the Bureau of Labor Statistics Survey of Private Establishments indicated that unionization not only increased the chances of health insurance coverage and pensions but also increased the expenditures on health insurance. In other words, companies that had unions not only gave their employees health insurance at a higher rate but also spent more on the health insurance presumably to give their employees better coverage overall.

There are many different things included in fringe benefits such as paid leave, life insurance, accident insurance, pension plans, and Social Security. However, health insurance is particularly important because of the costs that have risen so dramatically in recent years. It looks as though these health insurance costs will continue to rise and employers will be paying the largest amount of this if they keep health care coverage for their employees. The health care expenditures for workers in a union generally average around $2.26 per hour. This is compared to an average of $1.04 per hour for workers that are not unionized. Because of this it is suggested that unionization has a strong impact on the compensation differential and this has been underestimated quite significantly when the expenditures of employers on fringe benefits overall are looked at and the expenditures on health insurance are largely ignored. This information is summarized in the following table, which was taken from a study conducted in 2002 by Olson and reproduced here in its entirety for ease of understanding an explanation of the issue.

Total Compensation, Wages and Salaries, Fringe Benefit Costs, and Health

Insurance Costs for Union and Non-union Employees in Private Industry,

Wages and Compensation Salaries

All Union Employees

All Non-union Employees

Unadjusted Percentage Differentials

Total Benefit

Health

Costs

Insurance Costs

All Union Employees

All Non-union Employees

Unadjusted Percentage Differentials

Source: Internet:.

The Annual and Hourly Union-Non- union Differential in Health Insurance

Expenditure by Employers, by Sex, 1996-2000

Tobit

Marginal

Annual

Hourly Percentage

Coefficients Effects

Percentage

Differential in Health

Differential in Benefits

Health Benefits

Men

Women

The Hourly Union -- Non-union Differential in Wages and Health Insurance

Expenditures Compared with the Wage Differential in Percentage by Sex,

Wages and Health Insurance the Wage Differential (2)

Differential (1)

Men

Women

Difference between (1) and (2)

Men

Women

Source: Olson, 2002.

What has been done with the health-care coverage of employees has not always been noted but the Bureau of Labor Statistics began looking at how much employers were spending on employee health insurance in 1996. Complex calculations are often utilized to determine this type of issue. Findings of a recent studies support the findings from the 1980s which did suggest that those who are unionized not only provide health-care benefits at higher rate for their employees but also pay higher premiums for better coverage.

Implications regarding this include the fact that both male and female employees that work full-time, year-round jobs generally benefit from working in a company that is unionized. In other words, collective bargaining agreements have a measure of protection against the rapidly rising cost of health care. Also, studying the impact of unionization on benefits such as health insurance generally requires better collection techniques for data. Investigations that are more rigorous would benefit from much more detailed information on expenditures and benefit plans at the individual employer level. This type of information would help to advance the understanding of the total compensation differential between union and nonunion companies, and is generally something that should be looked at more strongly in the future.

Is also important to look at the differences between costs for benefits for those that work full-time and part-time. Generally, the cost for each hour worked when it comes to paying health insurance is greater for those that work part-time than it is for those that work full-time, and this is largely the reason that most employers do not offer health insurance coverage to part-time employees. However, there are some that do. The costs that employers paid for total benefits for employees grew approximately 50% faster than the wages for those employees when looking at the period between 1980 and 1994. By March of 1996, 28% of the total compensation for private workers in the United States was related to non-wage compensation such as health insurance and other fringe benefits. Because of these types of figures, it should come as no surprise that economists and others that analyze this type of issue are greatly interested in what type of role benefits play within the labor market.

Utilizing a simple theory of the labor market indicates that employers are generally concerned regarding the level of total worker compensation as opposed to the division between the wages that they actually pay and other compensation such as benefits. However, looking at this so simply ignores many of the important differences between benefits and wages. Generally it has been asserted that benefits represent what are called quasi-fixed costs, which means that they do not vary with the number of hours that are worked as wages do but rather they vary instead with the number of workers at the company. Because of this, what type of structure employee compensation packages have often influences the demand that an employer has for both part-time and full-time workers as well as the decisions that they make regarding overtime hours and pay.

There are several issues that need to be considered here and they will be explained briefly and then presented in table form. Although the data for this these particular tables comes from 1994, it is still relevant today because the issues that are dealt with here have not changed significantly over the period of time with the exception of the higher price of health care. As can be seen in the table below, which comes from a 1999 study, the differences in the rate of coverage between part-time and full-time jobs vary considerably. This ranges from approximately 19% for savings plans to around 63% for benefits such as health insurance. Some benefits are not required by law and for those benefits that are not required by law full-time jobs are much more likely to have benefit plans than are part-time jobs.

Table 2. Benefit provision and costs for full- and port-time workers March 1994

Characteristic

Vacation

Holiday

Percent of jobs covered

Part time

Full time

Average cost per hour

Part time

Full time

Average annual cost

Part time

Full time

Number of covered jobs

Part time

Full time

Health

Characteristic

Sick leave

Insurance

Percent of jobs covered

Part time

Full time

Average cost per hour

Part time

Full time

Average annual cost

Part time

Full time

Number of covered jobs

Part time

Full time

Legally

Characteristic

Pension

Savings required

Percent of jobs covered

Part time

Full time

Average cost per hour

Part time

Full time

Average annual cost

Part time

Full time

Number of covered jobs

Part time

Full time

Source: Microdata from the March 1994 Employment Cost Index. Buchmueller, 1999.

In order to help investigate how employers generally prorate the costs that they have associated with health insurance benefits, one study took a sample of 253 establishments that had at least one full-time and one part-time employee that both had health insurance. These firms contributed 787 full-time jobs and 424 part-time jobs to the employment cost index sample. The information presented below in table 3 indicates three different cost measures utilizing the establishment as the unit of observation. These include the average cost for the employer of the health-care benefits that were provided to part-time jobs, this same type of cost provided to full-time jobs, and the difference within the establishment between these two. The mean for each one of these variables was reported as well as the 25th, 50th, and 75th percentiles for comparison.

It can be seen by looking at this that the difference in the overall mean yearly costs for health insurance plans between those that were part-time jobs and those that work full-time jobs is approximately 24% which is very close to the differential that was to be expected. The difference between the median costs is also very similar. This suggests once again that those that offer health insurance to workers that are only part-time help to structure benefits, cost sharing, and other provisions of the plan in ways that help partially prorate the cost based on employee hours.

Table 3. Differences in annual health insurance costs between full- and part-time jobs, March 1994

Characteristic

Part time

Full time

Difference

Number of jobs

Number of establishments

Mean

Standard deviation

25th percentile

Median

75th percentile

Source: Microdata from the March 1994 Employment Cost Index. Sample restricted to establishments reporting data for both full- and part-time jobs. Buchmueller, 1999.

The last issue that must be looked at here in table form is the average provision of many of these plans and whether the establishments that were utilized for earlier tables are classified as being segregated or integrated. Of the 264 establishments dealt with here that have at least one full-time employee on the health insurance plan and at least one part-time employee that also has a health insurance plan, 192 of them are seen to be integrated and 72 are segregated. It is estimated by looking at the tabular data below that there are three specific ways in which establishments that are segregated work to lower the cost of health benefits that they provide to part-time workers.

The first one of these is to require that part-time workers pay a larger monthly premium than full-time workers. In approximately 73% of health insurance plans that are offered to full-time employees by segregated establishments the entire premium each month for coverage on a single person is paid by the company. In contrast to this the employers pay the entire monthly premium in less than 50% of the health insurance plans that they offer to employees that work part-time. For plans that require workers to make a contribution, however, the differences that are seen between part-time employees and full-time employees are very small. Based on single coverage, full-time employees require an average contribution of $40.19 while part-time workers require an average contribution of $45.86. For those that have family coverage as opposed to single coverage the difference between the two groups is even smaller than this and is not even seen to be statistically significant.

Another way that part-time worker plans differ from those of full-time workers is that part-time worker plans have more restrictions on issues such as pre-existing conditions. This helps to lower the benefit cost for employees that work only part-time in many establishments. The third and final way that these plans differ so strongly between full-time and part-time is the eligibility requirements based on length of service. Many part-time plans have a probationary period and almost 50% of them have a probationary period that is six months or longer. In contrast to this, fully 98% of plans for full-time employees that have a probationary period now have a period of three months or less.

Table 4. Proportion of full- and port-time jobs covered by various medical plan provisions by type of establishment (integrated or segregated), 1993-94 in percent, except where noted]

Integrated establishments

Characteristic

Part-time

Full-time jobs

Fee arrangement:

Fee-for-service

PPO

HMO

Self-insured

Employee contribution for single coverage: