| ||
Between 1999 and 2001, we analyzed the impact of AIDS on six corporations, four of which were subsidiaries of transnational corporations, based in South Africa and Botswana. The companies were large by the standards of developing countries, reporting sales of between $35 million and $3.4 billion at the time of the study. They operated in six industries—mining, metals processing, utilities, agribusiness, retail, and media—and employed between 500 and 35,000 people each. We refer to them as companies A–F because they asked to remain anonymous.
All the organizations first carried out voluntary surveys to estimate the number of employees who were infected with HIV. We divided the workforce into groups based on risk factors such as sex, age, job level, and location. Then, using the survey data, we estimated the likely incidence of HIV in each of the groups. The results of all the surveys, supplemented by epidemiological data from other sources, allowed us to create ten-year forecasts of new HIV infections and AIDS-related deaths for each company, and we used these later in the cost-benefit analysis.
If any of the executives in these companies doubted the seriousness of the AIDS threat, the surveys laid those doubts to rest. The prevalence of HIV ranged from 7.9 percent, approximately one in every twelve employees in company A, to 29 percent, nearly one in three in company C (South Africa's national prevalence rate was 25 percent in 2001). Companies in mining, metals processing, and agribusiness were affected the most, with more than 23 percent of their employees suffering from HIV/AIDS. Unskilled and skilled workers were two to three times more likely to be infected than supervisors and managers. In company C, for example, an estimated 39 percent of unskilled and skilled workers and 14 percent of supervisors and managers were infected. Most U.S. companies would regard even the lower infection rate as a disaster.
We worked with human resources managers, finance executives, and medical personnel to estimate the direct and indirect costs the companies had incurred because of HIV-infected employees. We then worked out the cost to each company of losing one HIV-infected employee at different levels of the organization. Since we could not assess the indirect organizational costs (the expenses on the lower right side of the exhibit "The Costs of AIDS to an Employer"), our estimates were much lower than the costs the companies actually incurred. Finally, we discounted each cost using the 7 percent real lending rate in South Africa in October 2001.
We found that the cost of one HIV infection ranged, on average, from less than half the affected employee's annual salary at company E to more than 3.5 times the employee's yearly pay at company C. After generating the present values of the costs of AIDS for five different levels of the workforce, we combined them with the forecast of infections to arrive at the cost that AIDS would impose on each company over the next ten years. The annual AIDS "tax" on business, as we called it, ranged from 0.4 percent of the annual wage bill at company E to 5.9 percent of the wage bill at company C in 2001. In absolute terms, it was as high as $11.9 million per year at company A.
| The cost of one HIV infection ranged … to more than 3.5 times the employee's yearly pay. |
| —Sydney Rosen, et al |
It's worth noting that the AIDS tax effectively divided the six companies into a high-cost group (companies A, C, and F) and a low-cost group (B, D, and E), but only partly because of differences in HIV prevalence. Unskilled workers at B, D, and E were not eligible for many of the benefit payments that other employees received, and lower-level workers received only minimal health care benefits. The companies had also capped their annual contributions to employee benefits funds, holding costs constant even as claims rose, which left their workers to bear more of the financial burden of HIV/AIDS on their own.
Benefits versus costs
Given the staggering cost of HIV/AIDS, what can corporations do? The only cost-effective way to respond is to fight the epidemic. That can take two forms: prevention programs to reduce the number of employees who will get infected and treatment programs to extend the working lives of employees who already have HIV or AIDS. Prevention and treatment are not mutually exclusive options; they are two parts of a comprehensive anti-AIDS strategy. We looked at the tactics in isolation, however, in order to better understand their cost-benefit implications. When companies invest in HIV prevention and AIDS treatment programs, they incur certain costs and derive certain benefits. The benefits, as we pointed out earlier, are the savings to the company because employees do not fall ill due to the epidemic. So our estimates of the companies' savings served as the benefit side of the cost-benefit comparison. To complete the analysis, we needed the costs to companies of implementing prevention and treatment programs and an assessment of the programs' effectiveness.
| The prevalence of AIDS ranged from one in every twelve employees, to nearly one in every three. |
| —Sydney Rosen, et al |
Prevention programs. Companies have tried many ways to prevent the spread of the virus in the workforce, but not all of them have proved to be effective. For example, education programs that some businesses invested in during the 1990s produced no clear benefits. Recent studies, however, show that employees' risk of HIV infection is reduced when companies provide—in addition to education and counseling—treatment for other sexually transmitted diseases, which facilitate HIV transmission. And encouraging employees to discover their HIV status through voluntary testing allows them to protect themselves if the virus has not infected them—or protect others if it has. State-of-the-art prevention programs therefore have four elements: educating employees, families, and the surrounding community about HIV/AIDS and how it can be avoided; distributing condoms to employees; treating other sexually transmitted diseases; and providing free counseling and testing services for employees and families.
Very little information is available about the returns on prevention programs, largely because their effectiveness in workplaces has rarely been measured. But according to our research, HIV prevention programs last year cost companies in South Africa between $10 and $15 annually per employee and achieved substantial reductions in the infection rate. For example, one South African mining company that wasn't in our sample estimated that its prevention program in a community of 4,000 miners would cut the HIV infection rate among employees by 50 percent. We calculated that if the companies we studied had reduced the workforce infection rate by 50 percent with programs that cost $10 per employee per year, the returns would have been positive for the three high-cost businesses. At the low-cost companies, we believe the investments would have been profitable if all the organizational costs of the epidemic had been taken into account. While the returns were not large, the fact that they were positive suggested that companies should invest more in HIV prevention programs.
Treatment programs. A few companies in developing countries have begun providing employees, either directly or indirectly, with free treatment for HIV/AIDS in the form of highly active antiretroviral therapy (HAART). The therapy consists of a combination of three drugs taken daily, with close monitoring by a medical professional, either on-site or off. There are three basic models in use in South Africa. Some companies offer third-party health insurance plans, much like those in the United States, and eliminate or reduce the employee copayment. Others contract with stand-alone HIV/AIDS management programs to provide services to workers. And a number of the very large mining and agricultural companies manage the health care of their employees in-house through company clinics.
To come up with the cost for a treatment program, we first assumed that patients would require the medicines eight years after being infected, on average, and would stay on them for the rest of their working lives. Responses to the treatment vary, but we assumed, based on medical research conducted in North America and Europe, that the therapy would allow employees to extend their working lives an average of five years. The annual cost of treatment was taken to be $500 per patient per year.
All the companies would have earned positive returns on their investments had they provided HAART at no cost to employees, according to the mathematical model we used. The annual reduction in the AIDS tax would have ranged from 0.8 percent in the case of company E to 40.4 percent at company F. In fact, providing free antiretroviral therapy at every level of the workforce made eminent financial sense for all six companies.
The Cost of AIDS to an Employer
| Direct Costs | Indirect Costs | |
|---|---|---|
| From one employee with HIV/AIDS (individual costs) |
|
|
| From many employees with HIV/AIDS (organizational costs) |
|
|
| Total cost of AIDS | ||
| Direct Costs | Indirect Costs | |
|---|---|---|
| From one employee with HIV/AIDS (individual costs) |
|
|
| From many employees with HIV/AIDS (organizational costs) |
|
|
| Total cost of AIDS | ||