Cost Shifting

Aggregate Cost-Shifting

In 2008, a total of $43 billion in uncompensated hospital care was provided. This so-called “free riding” was the principal justification for the individual mandate included in the Affordable Care Act enacted in March 2010. A 2011 analysis (Appendix A in Holtz-Eakin et al.) breaks this amount down as follows:

• Preexisting conditions. $8.7 billion of the $43 billion reflects care rendered to individuals who would purchase health insurance, but whose preexisting conditions prevented them from doing so; under the Act, they would be guaranteed coverage and so would no longer be uninsured;
• Medicaid Recipients. Of the remaining $34.3 billion, roughly $15 billion must be deducted for cost-shifters who are now newly eligible for Medicaid based on the Act’s expansion of insurance to all individuals and households whose income is at or below 133 percent of the poverty line;
• Illegal Immigrants and Other Nonresidents. Of the remaining $19.3 billion, roughly $8.1 billion is attributable to uncompensated care provided to illegal aliens or other nonresidents of the United States, who will not be subject to the mandate at all; and
• Payments Owed by the Insured. Of the remaining $10.6 billion, another $3.3 billion is attributable to care rendered to insured individuals who nonetheless did not pay their out-of-pocket share, such as co-payments or the like. The ACA would have no effect on these cost-shifters.

Hospital Cost-Shifting

In 2007, Medicare paid on average only 91 percent of the actual cost of hospital care for Medicare patients.  If hospitals made large profits, they might be able to absorb such discounts from patients with public coverage. But the average hospital operating margin averaged only 3.8% in 2006 (margin=(net revenue minus operating expense)/(net revenue); Milliman: Chart 5); nationally, between 1980-2003, average total hospital margins (total revenues/total costs minus 1) ranged from 4-6%  (Exhibit 4). Thus, it’s clear that many facilities would have difficulty absorbing such a deep discount in the form of lower profits.

Instead, the average hospital is able to cover its losses from the surpluses earned on paying patients. As an example, in 2007, hospital payments for the care of privately insured people were equal to about 132 percent of their actual costs of care (Sheils 4-29, Figure 9; see also Dobson).  This is the functional equivalent of a hidden sales tax of 32%!  Of course, realistically, at least some of this wide gap between private payments and the costs of their care reflects hospital market power rather than public sector cost-shifting, that is, it would have occurred even if Medicare and Medicaid were out of the picture. So how much of this total represents the cost-shift?

The actuarial firm Milliman has calculated that if Medicare and Medicaid cost-shifting had been eliminated, privately insured hospital costs in 2006 could have been reduced by 18% (Chart 3a); by the same logic, their costs for physicians in 2007 could have been reduced by 12% (Chart 3b).  These are still sizable amounts, but they also imply a zero-sum situation in which there is a dollar-for-dollar transfer of public plan cost savings to higher costs paid by privately insured patients. Is that really what happens?

According to the Lewin Group, “payments under private insurance are inflated by the cost of covering uncompensated care and payment shortfalls under public health coverage programs.”   But the firm also concedes: “Not all of the shortfalls in payments are shifted to private insurers. The literature indicates that only about 40 percent of uncompensated care and payment shortfalls are passed-on as higher prices for the privately insured. The remainder (60 percent) appears to be absorbed through reductions in costs and net income.” (Sheils, 4-29).  Moreover, Lewin Group estimates that Medicare allowable costs were 7 percent to 8 percent less than hospital’s reported costs in 2007;  Medicare non-allowable costs include advertizing, entertainment, penalties, gifts, donations, employee education, etc. (Sheils 6-25-09: Note 5), so at least part of the discount appears justified. The evidence related to hospital cost-shifting includes:

• Two early studies (using pre-1983 data) showed that at least half of government plan hospital payment shortfalls are cost-shifted (51% in Dranove: 56, 50% in Sloan & Becker);
• However, two other studies using data from the same period showed that hospital cost-shifting is much more limited, though not entirely absent (Zuckerman: 165 and Hadley and Feder: 68).
• Lewin’s own analysis in 1995 which found that 40 percent of the increase in hospital payment shortfalls in public programs were passed-on to private-payers (Sheils and Claxton)
• Even though they differed on the estimated size of the cost-shift, there was a consensus from these various studies that a hospital’s ability to cost-shift would be much more constrained in markets where there was a lot of hospital and health plan competition that would shrink the size of surplus revenues earned from private patients.
• A study using California data from 1993-2001 found that a 1% relative decreasein the average Medicare price was associated with a 0.17 percentincrease in the corresponding price paid by privately insuredpatients; similarly, a 1% relative reduction in the averageMedicaid price was associated with a 0.04 percent increase (Zwanziger: 197). Since the study was conducted in the one of the most competitive markets in the nation, these estimates should be viewed as lower bound measures of the likely size of the cost for the nation as a whole.

Physician Cost-Shifting

The problem of physician cost-shifting has been much less studied.

• One study of physician cost-shifting (using 1989-1990 data) shows no evidence of cost-shifting in response to Medicare fee reductions that for some procedures exceeded 30 percent (Rice: 215).
• This appears to contradict a recent claim (Sheils 4-29-09) that “one study of physician pricing by Thomas Rice et al., showed that for each one percent reduction in physician payments under public programs, private sector prices increased by 0.2 percent” (see Rice, Stearns 1994 in References; this report is not available on-line).
• Rice et al. did find evidence that on average, physicians respond to a 10% reduction in Medicare payments with an 11% increase in their volume of services to privately insured patients, but that is quite different from cost-shifting, which is raising prices without a commensurate increase in the quantity or quality of services provided.
• As noted earlier, the DOJ recognizes that where a health plan accounts for more than 30% of a physician’s practice revenue, the health insurer can have monopsony power to the detriment of patients (AMA: 2). In contrast to hospitals, many of whom might have 30% of revenues attributable to Medicare patients, few physicians rely on Medicare for 30% of their revenue. Thus, it is conceivable that failure to observe cost-shifting in the physician market is that the aggregate revenue losses are not sufficiently high to motivate a change in pricing behavior. But this might change dramatically were a new public plan to piggyback on Medicare’s discount–especially if the plan attracted more than 100 million members.
• Notwithstanding the paucity of evidence regarding physician cost-shifting, the Lewin model that has been used to estimate the cost of various House and Senate health reform bills for committees working on these proposals, assumes that 40% of physician underpayments are cost-shifted (Sheils 6-25-09: 12)

Pharmaceutical Cost-Shifting

Federal law and regulations require that drug manufacturers provide discounts or rebates on pharmaceuticals sold to a number of different federal programs ranging from VA health to Medicaid to TRICARE. Consequently, drug prices under six public programs are 36-59% below the average wholesale price of those drugs (CBO: 8-9). But for every tax dollar saved by these deep discounts, X gets shifted back to private paying patients (CHSR study).

Private Insurer Cost-Shifting

Taking into account cost-shifting from all sources, what is the impact on private insurers?

• Milliman calculates that for a typical family of four with private insurance, the current cost-shift by hospitals and physicians (100% of which is presumed to be shifted to private patients) amounted to about $1,800 in higher premium and out-of-pocket costs (Chart 4).
• Using a 40% cost-shift assumption, the Lewin Group estimates that the mid-July House health reform plan would result in an additional cost shift from the public plan to private plans amounting to 9.7% of private plan premiums, or about $650 annually per private policyholder (Sheils, Figure 2).

References

• AMA. Competition in Health Insurance: A Comprehensive Study of U.S. Markets: 2007 Update. American Medical Association, 2007. [Full Text (pdf)]
• American Hospital Association, Uncompensated Hospital Care Cost Fact Sheet, November 2008.
• American Hospital Association, Underpayment by Medicare and Medicaid Fact Sheet, November 2008.
• Antos, Joseph, Gail Wilensky, and Hanns Kuttner. The Obama Plan: More Regulation, Unsustainable Spending. Health Affairs, November/December 2008; 27(6): w462-w471. [Abstract]
• CBO. Prices for Brand-Name Drugs Under Selected Federal Programs. Washington, DC: Congressional Budget Office, June 2005. [Full Text (pdf)]
• Dobson, Al, Joan DaVanzo and Namrata Sen. The Cost-Shift Payment ‘Hydraulic’: Foundation, History, and Implications. Health Affairs, January/February 2006; 25(1). [Abstract (html)]
• Dranove, David. Pricing by Non-Profit Institutions: The Case of Hospital Cost-Shifting. Journal of Health Economics, March 1988; 7(1):47-57 [Abstract (html)].
  This paper shows that a hospital whose objective function includes output as well as profits may raise price to private paying patients in response to cuts in the price it receives for Medicaid or Medicare patients. Evidence is presented to show that hospitals in Illinois ‘cost-shifted’ in this manner in response to substantial reductions in Medicaid payments in1981-1982. As private sector pricing becomes more competitive, however, the ability and willingness of hospitals to cost-shift will wane.
• FamiliesUSA.  Paying a Premium: The Added Cost of Care for the Uninsured. June 2005. This report includes an analysis by Kenneth Thorpe showing the impact of uninsured uncompensated care on premium costs for family health insurance provided by private employers in 2005 included an extra $922 in premiums due to the cost of care for the uninsured; premiums for individual coverage cost an extra $341. Total care provided to the uninsured not paid for out of pocket was $43.1B.
• FamiliesUSA.  Health Tax: Americans Pay a Premium. 2009. This report includes an analysis by Milliman, Inc. using Medical Expenditure Survey data for 2006 projected forward to 2008. The analysis showed that the uninsured received an estimated $116B in medical care in 2008, of which they paid 37% out of pocket, another 26 percent was paid by third-party sources, such as government programs and charities. The balance, $42.7B was unpaid and estimated to be cost-shifted to privately insured patients; this latter amount increased premiums for family coverage by $1,017 and individual coverage by $368.
• Frakt, A.B. How Much Do Hospitals Cost Shift? A Review of the Evidence. Milbank Quarterly. Vol. 89, No. 1 (April 2011). Abstract.
• Gruber, Jonathan & David Rodriquez, How Much Uncompensated Care Do Doctors Provide?, 26 J. Health Econ. 1151, 1159-61 (Dec. 2007).
• Hacker, Jacob S. The Case for Public Plan Choice in National Health Reform. Institute for America’s Future (December 17, 2008) [Full Text (pdf)]
• Hadley, Jack and Feder, Judy. Hospital Cost-Shifting and Care for the Uninsured. Health Affairs, Fall 1985; 4(3):67-80. [Full Text (pdf)]
  This article examines hospitals’ cost-shifting behavior and its implications for care to the uninsured. Using data from a national sample of 128 private hospitals for 1980 and 1982. Insurers who pay on the basis of hospitals’ full charges seem to pay more for the same care than others. We find no evidence that the markups they pay bear any systematic relation to individual hospitals’ free care, the levels of other insurers’ payments relative to costs, or other revenue needs.
• Hadley, Jack, John Holahan, T. Coughlin, et al., Covering the Uninsured in 2008: A Detailed Examination of Current Costs and Sources of Payment, and Incremental Costs of Expanding Coverage (Washington: Kaiser Commission on Medicaid and the Uninsured, August 2008).
• Holtz-Eakin, Douglas et al. Brief for Amici Curiae Economists in Support of Appellees/Cross-Appellents and Affirmance. May 11, 2011. [Full text (pdf)]
• Kessler, Daniel P. Cost Shifting in California Hospitals: What Is the Effect on Private Payers?, California Foundation for Commerce and Education (2007) [Full text (pdf)].
  The paper reports two key findings. First, cost shifting from Medicare and MediCal is substantial. If, in 2005, the revenues for every California hospital’s Medicare and MediCal patients would have been sufficient to cover these patients’ costs, then private-payer patients’ revenue-to-cost ratio would have declined by 10.8 percentage points, from 1.309 to 1.201. Put another way, a one percent increase in revenue per dollar of MediCal cost leads to a 0.07 percent decrease in revenue per dollar of privately-insured cost; a one percent increase in revenue per dollar of Medicare cost leads to a 0.28 percent decrease in revenue per dollar of privately-insured cost. Second, cost shifting from the uninsured is minimal. If, in 2005, the revenues for every California hospital’s indigent patients would have been sufficient to cover these patients’ costs, then private-payer patients’ revenue-to-cost ratio would have declined by 1.4 percentage points, from 1.309 to 1.295.
• Ledue, C. “Study: 31 Percent of Patient Bad Debt Misclassified, Should Be Charity,” Healthcare Finance News, March 26, 2009.
• Lee, Jason S., Robert A. Berenson, Rick Mayes, and Anne K. Gauthier. Medicare Payment Policy: Does Cost Shifting Matter? Health Affairs Web Exclusive, October 8, 2003 [Abstract (html)]
• Medicare Payment Advisory Commission (MedPAC). A Data Book: Healthcare Spending and the Medicare Program (June 2009). [Full Text (pdf)]
• Miller, T. Covering the Uninsured: Springing a Leak in the ‘Cost Shifting Hydraulic.’ Health Affairs Blog, September 4, 2008.
• Milliman. Hospital & Physician Cost Shift: Payment Level Comparison of Medicare, Medicaid, and Commercial Payers. Presented by Will Fox and John Pickering. December 2008. [Full Text (pdf)].
• Proebsting, Doug. Why hospital cost-shifting is no longer a viable strategy. Milliman Healthcare Reform Briefing Paper, June 2010. Shows that average current ratio of private to Medicare reimbursement (for facility and professional services combined) is 140%. Under the Affordable Care Act, this will increase to 155% by 2015 and to 166% by 2020.
• Rice, T., S. Stearns, S. DesHarnais, D. Pathman, M. Tai-Seale, and M. Brasure. Physician Response to Medicare Payment Reductions: Impacts on public and Private Sectors. Final Report: Robert Wood Johnson Grant No. 20038, September 1994.
• Rice, Thomas. Physician Payment Policies: Impacts and Implications. Annual Review of Public Health 1997; 18: 549-565. [Abstract (html)] This paper reviews recent changes in physician payment policies, examines evidence on their impacts, and discusses their implications for researchers and policymakers. It includes extensive evidence that physicians respond to payment reductions with partially offsetting increases in service volume.
• Rice, T., S. Stearns, S. DesHarnais, D. Pathman, M. Tai-Seale, and M. Brasure. Do physicians cost shift? Health Affairs, Fall 1996; 15(3): 215-225. [Abstract (html)]
  This study analyzes whether physicians charge their privately insured patients more-a practice known as cost shifting-in response to Medicare payment reductions. As part of congressional legislation in 1989 and 1990, Medicare reduced its payment rates for selected procedures by as much as 30 percent. Here we examine whether reductions in Medicare rates increase how much physicians charge privately insured patients. Our data provide no evidence that physicians respond to Medicare payment reductions by shifting costs to their privately insured patients.
• Roy, A. (2011). Myths of the ‘Free Rider’ Health Care Problem, The Apothecary Blog, February 2, 2011.
• Sheils, John. The Impact of the House Health Reform Legislation on Coverage and Provider Incomes. Testimony before the Energy and Commerce Committee, U.S. House of Representatives. The Lewin Group, June 25, 2009; Updated: July 9, 2009. [Full Text (pdf)]
• Sheils, John.  Testimony Before the House Committee on Ways and Means. Falls Church, VA: The Lewin Group, April 29, 2009. [Full Text (html)]
• Sheils, J. and G. Claxton. Potential Cost-Shifting Under Proposed Funding Reductions for Medicare and Medicaid: The Budget Reconciliation Act of 1995. Report to the National Coalition on Health Care, The Lewin Group, December 6, 1995. Cited in Sheils, 4-29-09 (Note 12).
• Sloan, Frank and Becker, Edward. Cross-Subsidies and Payment for Hospital Care. Journal of Health Politics, Policy and Law, Winter 1984; 8(4): 660-685. [Abstract (html)] This study uses hospital data from the 1979 American Hospital Association Reimbursement Survey in a multivariate framework to assess the impact of discounts and third-party reimbursement on hospital costs and profitability. On the basis of the evidence in this study, we find ( I ) that the differential payment for Medicare, Medicaid, and/or Blue Cross is not justified on the basis of differential costs; (2) that the cost-containment efforts of the dominant payers have reduced total payments to hospitals somewhat, but a substantial amount of cost-shifting remains: and (3) that the savings is in profits, rather than in costs.
• Zuckerman, Stephen. Commercial Insurers and All-Payer Regulation. Journal of Health Economics, September 1987; 6(2): 165-187. [Abstract] This study explores commercial insurers’ claims about cost-shifting in the broader context of hospitals’ responses to revenue needs. It relies on hospital survey data from 1980 and 1982, a period that predates Medicare’s Prospective Payment System (PPS) for hospitals. While we Iind that limited amounts of cost-shifting occur, evidence does not support the notion, that it is a perfect safety valve to control financial status. Whatever cost-shifting occurs happens only at hospitals with well below average shares of commercial patients (i.e., where commercial plans lack market power).
• Jack Zwanziger and Anil Bamezai. Evidence Of Cost Shifting In California Hospitals. Health Affairs, January/February 2006; 25(1): 197-203. [Abstract (html)] We used 1993–2001 data from private hospitals in California to investigate whether decreases in Medicare and Medicaid prices were associated with increases in prices paid for privately insured patients. We found that a 1 percent relative decrease in the average Medicare price is associated with a 0.17 percent increase in the corresponding price paid by privately insured patients; similarly, a 1 percent relative reduction in the average Medicaid price is associated with a 0.04 percent increase. These relationships imply that cost shifting from Medicare and Medicaid to private payers accounted for 12.3 percent of the total increase in private payers’ prices from 1997 to 2001.