Tracking Health Care Costs:

Originally published by the Center for Studying Health System Change (HSC). HSC was a nonpartisan policy research organization funded principally by the Robert Wood Johnson Foundation.

Long-Predicted Upturn Appears

Issue Brief No. 23, November 1999 -- Paul B. Ginsburg

After three years of watching and waiting, the long-anticipated rise in health care costs finally materialized. Premiums for employment-based insurance climbed roughly 5 percent in 1999, a noticeable jump from the 3 percent increases recorded in 1998. Meanwhile, the underlying costs of private insurance -- measured with a one-year lag -- also rose by about 2 percentage points. Many observers had expected premiums to shoot up more sharply than costs, which would have signaled the turn of the insurance underwriting cycle. That shift had not yet taken place. This Issue Brief follows the trajectory of health care cost growth, premium changes for employer-sponsored coverage, and what these trends mean for working Americans.

Putting Cost Trends in Context

Premium increases for employer-based insurance were climbing, but the roughly 5 percent gains of 1999 were nowhere close to the double-digit spikes that defined the 1980s and early 1990s. Tracking these premiums matters because they influence whether employers continue offering coverage, what types of plans they make available, how much of the cost employees are asked to shoulder, and whether workers can afford the plans their employers provide.

To read premium trends correctly and anticipate where they might head, you have to look at what drives them -- the underlying costs of delivering medical care. In 1998, the growth rate for costs behind private insurance jumped by roughly 2 percentage points. When you zoom out and include Medicare and other public programs in the broader national health expenditure picture, the overall spending growth was more modest than what private insurers experienced. Medicare played a big role here. The program saw an unusually low rate of spending growth in 1998, thanks to budget cuts, success cracking down on fraud and abuse, and other factors. As a result, total national health spending rose only slightly in 1998. Still, that overall rate of increase outpaced growth in gross domestic product (GDP) -- a sign that the unusual stretch where health spending grew more slowly than the economy might be nearing its end.

Health Expenditure Patterns

Health care spending can be measured several different ways: by tracking what insurers and patients pay for services, by tallying provider revenues, or by examining the costs providers incur in delivering care (with labor costs being the most significant component). This analysis draws most heavily on data about provider revenues and labor costs at health care establishments.

Per capita provider revenues grew from 3.9 percent in 1997 to 4.5 percent in 1998, based on the expanded Milliman & Robertson Health Cost Index. Because general inflation actually fell between those years (from 1.9 percent to 1.0 percent), the real increase in spending was even more pronounced. Even so, the growth rate remained far below what the country experienced during the 1980s and early 1990s.

Among the different categories of provider revenue, prescription drugs stood out sharply from the pack. Drug spending had been climbing at double-digit rates since 1995, and the 1998 increase was markedly steeper than in 1997. Most of this acceleration reflected growth in volume rather than price -- drug expenditures rose 14.3 percent while the drug component of the Consumer Price Index increased only 3.7 percent. Meanwhile, spending on inpatient hospital services and physician care continued to grow at comparatively low rates.

Payrolls in health services establishments, measured per capita, grew 4.2 percent in 1998 but only 2.9 percent through the first eight months of 1999. This represented a notable drop from the 5.7 percent rise in 1997, and the downward movement ran counter to the upward trend in provider revenues.

The slower payroll growth in 1998 compared to 1997 was entirely accounted for by a deceleration in hours worked. Average hourly wage increases in health care -- 3.5 percent in 1998 -- were slightly higher than in 1997 but still below the increases seen in other industries. For many years, wages in health care had risen much faster than in other sectors. That gap closed in 1994, and by the late 1990s, health care workers were seeing notably lower wage growth than workers in other fields.

The sharp decline in payroll per capita during early 1999 suggested that further acceleration in health care cost growth was unlikely in the near term. Providers were under heavy pressure from Medicare reimbursement cuts and managed care organizations to keep costs down.

Insurance Premium Developments

As had become a pattern, anecdotal reports through the year suggested much steeper premium increases than the systematic data eventually confirmed. But recently released surveys did show a genuine uptick. The Kaiser Family Foundation/Hospital Research and Educational Trust (KFF/HRET) Survey of Employer-Based Health Plans reported premiums rising 4.8 percent in 1999, up from 3.3 percent in 1998. Data from the Hay Benefits Report told a similar story, with estimates of 5.4 percent growth in 1999 versus 3.9 percent in 1998. This 1.5-percentage-point acceleration was broadly consistent with the increase in underlying cost trends (lagged one year) estimated by Milliman & Robertson -- 5.2 percent for 1998 versus 3.3 percent for 1997.

Health policy analysts had widely expected premium growth to outstrip underlying cost growth as a result of the insurance underwriting cycle. During the mid-1990s, insurers had been cutting premiums aggressively to break into new markets and grab market share. By 1998 and 1999, that dynamic had reversed, with insurers pulling out of selected markets. However, strong employer pushback against premium hikes appeared to have delayed or dampened the next phase of the underwriting cycle, in which premiums typically rise faster than costs. Eventually this phase would arrive, restoring plan profitability and drawing fresh capital into the industry -- at which point premium growth would slow relative to costs again.

One telling result from the KFF/HRET survey was a sizable gap between premium increases for fully insured and self-insured plans. Fully insured plans saw 5.8 percent increases in 1999, versus 3.7 percent for self-insured arrangements. This implied that fully insured plans were pricing more aggressively than actuarial cost projections would justify -- projections reflected in self-insured rates -- suggesting the underwriting cycle shift might be starting.

The pattern during the 1990s was clear. From 1992 through 1994, premium growth ran well ahead of cost increases, and health plans posted strong profits. That reversed in 1995. From 1996 through 1998, premiums lagged behind costs and plan margins shrank considerably. If 1999 cost trends held steady, the gap between cost growth and premium growth would narrow substantially, potentially setting the stage for premiums to exceed costs in 2000.

Premium increases also varied by plan type. In both the KFF/HRET and Hay surveys, indemnity plans saw the steepest hikes -- nearly 3 percentage points higher than HMOs and point-of-service plans in the Hay data. This divergence was more aligned with expectations than in prior years, when indemnity and managed care premium growth had been surprisingly close.

What This Meant for Consumers

When premium increases began falling in the early 1990s, employers did not immediately share those savings with workers. Many actually increased the share of premiums that employees had to pay, especially for family coverage. As premiums dropped further, that trend reversed and workers got meaningful relief from rising health costs. By the late 1990s, both the KFF/HRET and Hay surveys showed employees paying a larger share of family premiums but a smaller share of single coverage premiums -- perhaps because employers were responding to the growing number of households with two wage earners and dual coverage options.

Throughout the decade, consumers saw a significant drop in out-of-pocket medical spending. The average person's out-of-pocket costs for medical services were 6 percent lower in 1997 than in 1990, according to the Consumer Expenditure Survey, even as overall per capita spending on medical services climbed 36 percent over the same period. Consumer spending on drugs rose 27 percent during those years -- far less than the 97 percent jump in per capita drug expenditures.

Much of this decline in out-of-pocket costs relative to total spending reflected the shift to managed care. An HMO enrollee might pay $10 per doctor visit and $5 per prescription, compared with a $250 deductible and 20 percent coinsurance under a traditional indemnity plan. Of course, some of this financial benefit came with trade-offs: restrictions on provider choice and other managed care constraints.

Looking Ahead

Over the longer term, costs would remain the dominant force shaping premium trends, though premiums were likely to climb faster in 2000 as the underwriting cycle shifted. Several factors pointed toward continued cost acceleration. In response to the managed care backlash, plans were loosening their restrictions on care. New or proposed managed care regulations at both the state and federal level, along with the growing threat of lawsuits, stood to push costs higher. Expensive new medical technologies, especially pharmaceuticals, kept entering clinical practice. And ongoing provider consolidation had the potential to give hospitals and physician groups more bargaining power, leading to higher reimbursement rates.

Despite these pressures, a return to the massive double-digit increases of the late 1980s and early 1990s seemed unlikely. Health care markets had become fundamentally more competitive. Employers continued to push plans hard on premiums, and plans in turn pressed providers for discounts. Those competitive dynamics would keep moderating the pace of cost growth.

Still, even modest growth in health costs carried broad consequences. Significant premium increases would leave workers with less money for other needs. Rising Medicare and Medicaid spending would threaten the favorable fiscal outlook that governments then enjoyed. And higher costs could push more people out of coverage altogether. An HSC study found that 20 percent of uninsured Americans had access to employer-sponsored coverage but turned it down -- mostly because of cost. Any meaningful uptick in premiums risked making coverage even less affordable for these workers.

Data Sources

The 1999 KFF/HRET survey included 1,939 employers drawn as a random sample of firms. The 1999 Hay Benefits Report covered 1,009 employers from a convenience sample of Hay's client base.

The Milliman & Robertson Health Cost Index was designed to reflect cost increases faced by private insurers, limited to commonly insured services -- inpatient and outpatient hospital care, physician services, and prescription drugs. Because provider revenue data generally cover all patients, M&R analysts subtracted Medicare payment data from the totals. The HCI tracked the experience of a private policy with a $250 deductible.

Payroll cost data from the Bureau of Labor Statistics covered all nonsupervisory workers in health services establishments (SIC 8000). Nonsalaried health professionals were not included. Though payrolls represent just one component of provider costs, they are an important one and are available with the shortest time lag.

The Department of Labor's Consumer Expenditure Survey sampled approximately 5,000 households about their spending by category over the prior quarter. Health-related categories included health insurance premiums, medical services, drugs, and medical supplies.

Sources and Further Reading

CMS — National Health Expenditure Data — Official U.S. health spending data.

Kaiser Family Foundation — Health Costs — Health care cost analysis.

Health Affairs — Peer-reviewed health policy research.

Robert Wood Johnson Foundation — Health policy research.

Commonwealth Fund — Research on health care costs.