— Former senator Jim DeMint (R-S.C.), writing in The Wall Street Journal, Oct. 17, 2013
Jim DeMint, president of the Heritage Foundation, wrote an opinion article in which he declared that the organization would continue to fight the Affordable Care Act, a.k.a. Obamacare, despite the failure of the government-shutdown strategy that the organization had advocated earlier this year.
In making the case that the cost of the health-care law was sure to grow, DeMint cited some figures about Medicare that struck us as a bit fishy. So we decided to investigate.
There’s no question that any new social program often has unintended costs. We often point to the fascinating story of the creation of Medicaid, detailed in a 1993 article by former White House aide Joseph A. Califano Jr., which unexpectedly has led to one-third of the Medicaid budget today going to nursing homes. That’s an excellent example of a good intentions and congressional deal-making gone awry. More than 30,000 people have read this article since we first posted it on Scribd last year.
But DeMint gets into trouble when he starts tossing around 50-year old figures and compares them to today’s numbers. There is, for instance, something called inflation which makes these numbers not easily comparable. In the 1970s, inflation hit double digits, which is why most budget experts compare budget figures over time by using a metric such as a percentage of the overall economy or payroll costs or something else that allows for apples-to-apples comparisons.
Let’s take the first figure — that the Medicare Part A budget in 1990 is six to seven times higher than originally estimated. In February 1994, after Texas billionaire Ross Perot used this figure in a debate with then-Vice President Al Gore, Robert J. Myers wrote an article for the newsletter of the Society of Actuaries explaining the problems with this figure.
Who is Robert J. Myers? He was the long-time Social Security actuary who developed the original estimate for the House Ways and Means Committee.
In the article, Myers is forthright about the fact that he underestimated the potential costs, but he says that inflation in the 1970s makes a dollar comparison fairly useless. Instead, he says the best approach is to compare costs as a percentage of taxable payroll. Thus, his original estimate was that the cost would be 1.61 percent of taxable payroll in 1990, whereas it turned out to be 2.41 percent. Thus the ratio of actual adjusted outgo, as a percentage of taxable payroll, is 150 percent higher, though he makes some other adjustments to account for changes in the law and ultimately arrives at a figure of 164 percent. (Note: as readers have pointed out, this translates into a difference of 64 percent.)
That’s not great, he admits, but it’s very different than 644 percent. “A deviation such as this is nothing to be proud about; it is, however, much better than that based on the dollar values alone,” wrote Myers, who was president of the Society of Actuaries in 1971-72. “Nonetheless, the only thing for me to do now is to commit bah-kari!”
“We never stated that we were using inflation-adjusted numbers,” said Ken McIntyre, spokesman for Heritage. “But adjusting for inflation, as you note, also shows a large discrepancy in anticipated costs.”
As for DeMint’s other figure, he uses the slippery word “likewise” to suggest that his next figure is a similar comparison. But what he is actually doing is plucking a 1965 figure — $500 million a year — and suggesting that was also a prediction for 1990. The quote — that Medicare Part B would require “federal appropriations of about $500 million a year from general tax revenues” — is not from a budget document but appears to come from an obscure and relatively minor New York Times article that appeared on March 11, 1965. But it was not a 50-year estimate, like the previous statistic.
Califano also mentions the $500 million number in his article. He says that this was the price of getting the Medicare bill out of the Senate Finance Committee — to “agree to pay hospitals on a cost-plus basis, and doctors’ fees that were ‘reasonable,’ ‘customary’ and ‘prevailing’ in their communities, thereby giving physicians the power to raise their own fees.”
Califano says that Johnson looked at the figure and dismissed it as peanuts. “His single-minded focus on access for the poor and elderly led him to grossly underestimate the price of giving hospitals and doctors the keys to the federal Treasury,” he wrote.
McIntyre said the 4,400 percent figure was inflation-adjusted. “The point of the DeMint piece — that Medicare spending has vastly exceeded projections — holds, regardless of what data set you want to use,” he said.
The Pinocchio Test
For the Pinocchio test, DeMint’s assertions pose an interesting question. On the one hand, it is broadly correct that original estimates of the costs of Medicare were too low. Certainly, as Califano noted, there were unintended consequences to legislative deals made to get the bill through Congress that ended up burdening the U.S. Treasury.
At the same time, DeMint’s figures are ridiculously overstated. He preferred to use trumped-up stats rather than refer to the concrete examples cited by Califano. As a general proposition, comparing estimates over nearly a half century is a bit of a fool’s errand, as the world of 1965 is so different from today. It reminds us a bit of Mitt Romney’s efforts to compare today’s Navy to the Navy of 1916 by counting ships — gunboats versus aircraft carriers.
In researching this column, we found that DeMint’s numbers are not particularly original and have been recycled over the years by various skeptics of major changes in the health-care system. He could have made his points without relying on such tendentious numbers. It’s time to retire them, and so DeMint earns Three Pinocchios.
Check out our candidate Pinocchio Tracker