Network News

X My Profile
View More Activity
Posted at 6:00 AM ET, 12/21/2007

Hillary Bakes Some Stats

By Michael Dobbs

Johnston, IA, Dec. 17.

"During the Nineties, health care inflation was kept low. In the last seven years, health care costs have doubled."

--Hillary Clinton, Town Hall meeting, Johnston, IA, December 17, 2007

This is one of those cases where a politician mixes and matches data to best support his or her argument. Take a number from one data set, combine it with a figure from a different data set, add a tablespoon of political rhetoric, stir well, cook at 450 degrees (25 degrees in Iowa), and voila, we have a tasty dish for the credulous voter. Prior to serving, make sure you remove any warning labels.

Hillary's goal, needless to say, was to show that life for the average American was much better under Clinton I than under Bush II. But there are other ways of cooking the stats to produce a very different taste. Pay attention while I demonstrate how it's done.

The Facts

First, some background. Health care costs have been rising for decades, well above the rate of inflation. The argument is about differences in the rate of growth. According to the Centers for Medicare and Medicaid Services, annual growth rates in per capita health expenditures dipped during the mid-90s to a low of 5.2 percent in 1996, but then began to rise every year until peaking in 2002 at 9.1 percent. Since 2002, the rate of growth declined to 6.9 percent in 2005, and will probably decline further in 2006.

The following graph shows how the growth in per capita health care expenditures began to accelerate after 1996-1997, during Clinton's second term.

Another way of measuring the growth in health care costs is to look at health insurance premiums, which are generally more volatile than overall health expenditures. According to the Kaiser Family Foundation, the annual growth rate in premium costs dipped to 0.8 percent in 1996, but then started rising. The growth rate had reached 8.2 percent in 2000 and climbed to a peak of 13.9 percent in 2003. See this table. After 2003, the growth rate began declining to an estimated level this year of 6.1 percent.

In order to come up with her sound bite on health care costs, Hillary Clinton used two different statistical measures, one for the Nineties, another for the last seven years. To make the point that the Clinton administration held down health care "inflation", Clinton campaign spokesman Phil Singer cited U.S. government data showing that overall health spending increased from 13.7 percent of Gross Domestic Product in 1993 to 13.8 percent in 2000. To support the statement about the doubling of health care "costs" over the last seven years, he cited the Kaiser Family Foundation study on health care premiums.

Note also the comparison of a carefully chosen inflation rate (the increase in health spending over the increase in GDP), for Clinton I, and the costs of health care premiums, for Bush II. By this definition, a zero or negative "health care inflation" rate could still result in a considerable rise in "health care costs."

Techniques that are perfectly acceptable for chefs are not so kosher for statisticians. To compare health care costs under Clinton and health care costs under Bush, you should work from the same ingredients. Otherwise, you are comparing apples with oranges.

There are many reasons why premiums began to rise after 1996, according to Gary Claxton, vice president of the Kaiser Family Foundation, who has followed the issue closely. The collapse of the Clinton health care plan played a role, as did the consolidation of various HMOs and hospitals. Another factor was the Balanced Budget Act of 1997, which cut back on Medicare payments to hospitals.

The Pinocchio Test

It is true that the rates of growth in both health care costs and premiums have peaked under Bush. But it is also true that the rise in annual growth rates began under Clinton and is now declining under Bush. It is also a fact that the annual increase in health care costs and premiums is lower today than it was in 2000, when Clinton left office, as this chart illustrates.

In other words, the same statistics can be twisted to support many different arguments.

By mixing and matching data, Hillary Clinton paints a rosier-than-justified picture of the halycon days of Clinton I. This is standard operating procedure for presidential candidates, but it is still misleading.

(About our rating scale.)

By Michael Dobbs  | December 21, 2007; 6:00 AM ET
Categories:  2 Pinocchios, Candidate Record, Candidate Watch, History, Social Issues  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Romney's 'Flip, Flop, Flip'
Next: Marching with King

No comments have been posted to this entry.

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2011 The Washington Post Company