On October 15 the Social Security Administration announced there would be no Cost-of-Living-Adjustment (COLA) for 2016. Many perceived this as signifying a hardship for seniors. Lawmakers afterward included a provision in the budget deal to prevent some seniors from facing huge Medicare premium increases which were among the perverse effects that otherwise would have arisen from the zero COLA. This piece explains the basics of Social Security COLAs, as well as how zero COLA years can lead to confused politics and strange policy.
How COLAs work: The annual Social Security COLA is calculated by comparing the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W) in the third quarter of the most recent year to its level in the third quarter of the previous year. CPI-W is just one of multiple measures of general price inflation maintained by the Bureau of Labor Statistics (BLS). Most economists believe it is less accurate than other measures including the Chained Consumer Price Index for All Urban Consumers (C-CPI-U). It just so happened that Social Security COLAs were first established in law before the other more refined measures were developed.
Other quirks of law surround Social Security COLAs. One is that when there is no COLA, neither is there an increase in the amount of wages subject to the Social Security payroll tax – even though normally those automatic annual tax base increases are computed differently than the COLA or CPI.
Another provision of law (the so-called “hold harmless” provision) prevents most beneficiaries’ monthly Social Security benefit checks, net of Medicare Part B premiums, from declining. This means that whenever there is a zero COLA year, roughly 70% of seniors do not face a Medicare premium increase even though their benefit costs have likely gone up. Under law the resulting revenue loss to Medicare is supposed to be made up by higher premiums from high-income seniors and on behalf of low-income seniors (whose premiums are paid for them under Medicaid by the states). This in turn can mean huge premium increases for a minority of seniors on opposite ends of the income spectrum, as would otherwise have happened this year. These various provisions do not add up to a coherent policy, but rather embody a patchwork of responses to the perceived policy and political challenges that accompany zero COLA years.
Zero COLAs are usually more good than bad for seniors, Part 1: Despite what advocacy groups often say, a year without a COLA usually reflects a situation more good than bad for seniors. This is because there is no provision in law allowing for a negative COLA. Thus, if prices rise a large amount one year but fall the next, beneficiaries get a large COLA after the first year but no reduction for the second year. This means that seniors receive higher benefit payments than they would have if current price levels had instead been reached via persistent, regular price inflation. Seniors continue to receive these higher payments in a lower-price environment, with this (usually very small) bonus never taken away.
As Figure 1 shows, prices typically rise by a small percentage each year. But this year prices (per CPI-W) have gone down slightly, modestly increasing beneficiaries’ purchasing power. A similar situation previously arose, to a much greater extent, in the 2008-11 period. Prices rose swiftly in 2008, producing a large COLA for 2009 even though prices went down during that year. Prices didn’t return to 2008 levels until 2011 because the increase in 2010 was not as great as the decrease in 2009 had been. As a result the purchasing power of Social Security benefits outpaced inflation during that period.
Zero COLAs are usually more good than bad for seniors, Part 2: As mentioned earlier, under law Medicare Part B premiums can’t rise for most beneficiaries whenever there is no COLA. This is a clear advantage to seniors, who receive benefits of higher value without paying higher premiums.
Our political discussion often confuses the concepts of prices, costs and spending: Economists generally agree that the CPI-W overstates price inflation relative to a chained index such as C-CPI-U. There isn’t space in this piece to review the details, but this link from CRFB explains some. Yet whenever the annual COLA is zero or quite small, longstanding arguments re-emerge that CPI-W actually understates price inflation as experienced by seniors. AARP for example has argued that an experimental senior price index (CPI-E) would be better, saying that CPI-W “does not accurately represent the buying habits of seniors,” largely because seniors spend more of their income on health care, where costs tend to rise more rapidly.
Much of this discussion confuses the different concepts of prices, costs and spending. COLAs are intended to reflect price changes rather than other factors that increase total costs. Indeed, much health care cost growth does not arise from price inflation but rather the adoption of new technologies.
In general, whether we spend more on any area depends on many factors other than prices. This year you might spend a lot more on plumbing services than you did last year – but not necessarily because the plumber’s prices went up. Instead this may simply reflect your greater need for plumbing services, or the plumber having new services to offer. The fact that seniors spend more on health care as technology progresses in response to their growing needs is indeed an important policy concern. But how much to help seniors afford rising health care costs is primarily an income support issue or a health care policy issue. It is not primarily an issue of price inflation measurement.
Our perplexing system for assessing Medicare Part B premiums makes it difficult to construct sensible policy. As earlier noted, whenever there is no COLA about 70% of Medicare beneficiaries are excused from financing a proportionate share of program cost increases. Under law the revenue loss is to be made up by assessing higher premiums on those who are not so excused: low-income beneficiaries (whose premiums are paid by Medicaid) and higher-income beneficiaries (who pay larger income-related premiums).
This year those premium hikes would have been enormous had Congress not acted. The trustees’ report contains an estimate that the relevant premium would have had to rise from $104.90 to $159.30 even without accounting for the still-higher premiums facing those on the high-income end. Joe Antos has estimated that monthly premiums would have risen to over $500 in the top bracket.
Faced with this situation lawmakers acted to limit the standard premium to about $120 (higher-income beneficiaries will still pay substantially more). The revenue loss resulting from the premium relief would jeopardize program finances, so lawmakers enacted a loan to Medicare from the general treasury, charging affected beneficiaries an additional $3 a month until the loan is repaid.
If you are confused by all this, you’re not alone. Medicare costs keep rising even in a zero COLA year, but the law’s complexities make it very difficult to discern who pays for them. Most beneficiaries aren’t doing so, due to the hold-harmless provision. Lawmakers also just excused high-income beneficiaries from much of the burden of doing so. High-income seniors will still pick up a bit of the cost as will states through Medicaid, who will then pass those on to their residents in various hard-to-track ways. Some of the rest is being picked up by loans from the general US treasury for which all Americans must pay, though none of us know our own share. None of this is a recipe for transparency.
The policy ideal would be a Medicare system in which costs do not rise faster than the ability of senior premium payers to bear. This would require tough decisions about fundamental reforms, eligibility rules and benefit growth rates that the body politic has thus far been unwilling to make. Failing this ideal, the next best outcome would be a system in which beneficiaries and taxpayers each shoulder an appropriate and transparent proportionate share of rising program costs. But this in turn would mean Social Security checks net of premium payments declining in some years, the optics of which have long made for prohibitive politics. As a result, the opaque and seemingly arbitrary process of Medicare premium setting is likely to continue for some time.
Charles Blahous is a senior research fellow for the Mercatus Center, a research fellow for the Hoover Institution, and a contributor to e21. He recently served as a public trustee for Social Security and Medicare.
Interested in real economic insights? Want to stay ahead of the competition? Each weekday morning, e21 delivers a short email that includes e21 exclusive commentaries and the latest market news and updates from Washington. Sign up for the e21 Morning eBrief.