According to a new study from the Manhattan Institute they do. Proving the old adage that you can fool all of the people some of the time, but you can fool the Washington Post all the time, the Post ran a column yesterday by Frank Lichtenberg, the author of the Manhattan Institute’s study. Lichtenberg’s main point was that new drugs are big life savers and therefore we should be generous with our spending on drugs. Lichtenberg analyzed data on the increase in life expectancy by state between 1991 and 2004. He compared it with the rate at which states adopted newer drugs. This was measured by the average year in which the FDA first approved the active chemical compound in drug prescribed for Medicaid or Medicare patients. Lichtenberg finds a strong correlation between the use of newer drugs and increases in life expectancy. He also finds that productivity (measured as output per worker) grew most rapidly in states that adopted newer drugs more rapidly. And, best of all, there is a negative and insignificant relationship between total per person health care expenditure and the rapid adoption of newer drugs. In other words, it seems that any additional spending on new drugs is more than offset by lower medical costs elsewhere – a real free lunch. Before we start celebrating these great new findings, a closer look is in order. The study also finds a very strong positive relationship between per person nursing home and home health care expenditures and new drug use. It also finds a negative, but insignificant, relationship between education and productivity. (Maybe we really are wasting our money sending our kids to college.) Confronted with these mysterious findings, I did a quick regression which related Lichtenberg’s index of new drugs to the median age of state population. Bingo! There is a very strong positive relationship between the newness of drugs in Lichtenberg’s index and the median age of a state’s population. In other words, if you have an older population, then you would have been more likely to adopt newer drugs. This is presumably because most new drugs were developed to meet the needs of an older population. (Lichtenberg’s measure of newness actually doesn’t correspond well to new drugs as approved by the FDA. Most new drugs do not use new chemical compounds. This lead to the wonderful result that the Lichtenberg’s measure of the newness of drugs in Medicare is actually associated with less spending on drugs. Maybe if we restrict spending on drugs we will force doctors to make better choices, which will lead to longer life expectancies.) The age story also explains Lichtenstein’s finding that education has no effect on productivity. The youngest cohorts are the most educated. People in their twenties are far more likely to have a college degree than people in their forties. However, productivity increases with age, at least until workers reach their forties. This means that if we don’t control for age, it is entirely possible that a state with a more educated workforce will actually be less productive. Lichtenstein’s study probably bears somewhat closer examination, but the lesson for Washington Post should be to talk to someone who knows something about the topic before rushing a column into print. The editors are on record as being big supporters of high drug prices, but they should not be cherry-picking research to publicize that supports their position.
--Dean Baker
