Seniors won't receive a raise on their Social Security payments this year. It s the first time that has happened since 1975, when cost-of-living adjustments to keep pace with inflation were made automatic.

The freeze has drawn fresh attention to the longstanding argument that the current COLA system understates the true cost of growing old in America.

New research has broken the matter down to dollars and cents.

In 1972, Congress voted overwhelmingly to link payments for Social Security and a handful of other federal programs to the Consumer Price Index for Urban Wage Earners and Clerical Workers. It's similar to the nation's headline inflation measure, the Consumer Price Index for All Urban Consumers (CPI-U), but it represents a smaller portion of the population (32% versus 87%), and it places a slightly lower weight on costs like medical care, housing and recreation and a slightly higher weight on those like clothing, food and transportation.

Pensioner advocacy groups argue that neither measure adequately captures how rising costs have affected the elderly. For example, the old typically have much higher health-care expenses than the young, and health-care costs have risen far faster than the overall inflation rate in recent decades. Premiums for Medicare Part B, which covers the cost of outpatient care and doctor fees for the old, have increased about 14-fold since 1976, versus about a four-fold increase in broad cost-of-living measures.

In 1987, Congress directed the Bureau of Labor Statistics, which publishes the CPI-U and CPI-W, to begin calculating a consumer price index for the elderly. Thus was born the experimental CPI-E. In 2008, when the measure turned 25 (it was calculated using a base date of 1982), the BLS published the results of its experiment. From December 1982 to December 2007, the CPI-W rose 110%, the CPI-U rose 115.2% and the CPI-E, the index of elderly costs, rose 126.5%.

Had Social Security payments been linked to the CPI-E all along, payments would today be much larger than they are.

A recent study by researchers at Stanford University looked at the amount of Social Security benefits seniors retain after paying out-of-pocket medical costs. For the average man born in 1918 who began collecting Social Security benefits in 1983 (at age 65), payments net of medical expenses rose from an initial $542 a month to $867 a month by the end of 2007. If his net payments had kept up with inflation for non-medical goods, he would have had $1,081 in net payments that year. Even though the gross income was rising, health-care inflation effectively cut his benefits. Were payments indexed to the CPI-E, the researchers found, the same man would have collected payments net of medical bills of $957 in 2007, easing but not quite eliminating the cost-of-living squeeze.

The Senior Citizens League, an advocacy group, has renewed calls for full implementation of the CPI-E for use in calculating COLAs. Late last month, Rep. Earl Pomeroy (D., N.D.) introduced the Seniors Protection Act of 2010, which would give seniors, veterans and other collectors of federal benefits a one-time payment of $250 to make up for the lack of a COLA this year.

There are alternative views, however, that go beyond the matter of cost. (A 2003 study by the Federal Reserve Bank of New York concluded that CPI-E implementation would bankrupt Social Security five years sooner than projected.)

The BLS, which doesn't take sides on policy matters, emphasizes that the CPI-E is an experimental measure that might not accurately reflect costs facing recipients of Social Security payments and other federal benefits. For one thing, it's based on age 62 and up. Many benefits collectors are young and disabled; their spending patterns might not match those of the old. Also, the CPI-E captures the habits of workers age 62 to 67 who have delayed Social Security payments in exchange for bigger benefits. Again, they might spend differently than the retired. To construct an index that more faithfully represents federal benefits collectors, the BLS would need a new directive, funding and plenty of time to study the group's consumption habits, says John Greenlees, a Bureau economist.

There's also a chance that CPI-E implementation could backfire. The index weights health-care costs much more heavily than its sibling indexes, but also has offsetting features that hold down the cost of living. For example, the CPI-E assumes seniors enjoy discounts on many of their purchases. BLS research shows that gap between health-care inflation and overall inflation is narrowing, and so is the difference between CPI-E and other measures. Over the decade ended 1993, the CPI-E rose at 4% a year versus 3.7% for the CPI-U and 3.5% for the CPI-W. Over the next decade, the CPI-E increased at 2.8% a year, versus 2.6% for each of the other measures.

If health-care inflation reverts to the mean and matches overall inflation in coming years, or even lags it, a switch to the CPI-E could constrain senior payments. The influential American Association of Retired Persons says more research is needed on the CPI-E, and that a switch to the CPI-U for Social Security would be the best thing for now.

As for the one-time payment of $250, not everyone thinks it's a good idea. The Center on Budget and Policy Priorities, a Washington, D.C. think tank, points out that the COLA process was adopted to eliminate ad hoc adjustments and congressional bidding wars for senior votes. In a paper published late last year, the CBPP called the case for a 2010 benefit increase for seniors "weak." Last year's increase of 5.8%, the largest in 25 years, was based largely on a jump in energy prices, and the timing of COLA calculations (which are done at the end of the third quarter for the subsequent year). Energy prices plunged over the following year. If Social Security payments were strictly based on inflation measures, the group notes, seniors would get a benefits cut this year under any inflation measure, including the CPI-E.