April 9, 2023
by Stephen Stofka
This week’s letter is about health care spending and its effect on inflation. Economists construct a composite price index number out of many components of the economy. While that construction may have a rigorous methodology we struggle to make causal inferences from the data because price movements in an economy are complex.
In 1965 President Johnson signed the law creating the Medicare and Medicaid programs. At that time, health care spending was 6% of total consumer spending. The radical reformers of that age wildly underestimated Medicare’s costs, particularly for inpatient hospital costs. Since the government now paid for the first 90 days of a hospital stay, doctors were encouraged to take a cautious approach and keep a patient in the hospital if there was a chance of infection or accident at home. The deep pockets of the federal government incentivized medical and pharmaceutical companies to develop new drugs and equipment. Hospitals expanded their surgery and rehabilitation units. Doctors increasingly turned to specialization and their numbers tripled from near 90,000 in 1965 to 284,000 in 1990, according to the National Center for Health Workforce Analysis. In 25 years, healthcare spending (https://fred.stlouisfed.org/graph/?g=12f9k) more than doubled as a percent of consumer spending, coming close to 15% of the total. It has risen to 17% in the past decade and now is 16%. Here’s a chart showing the growing contribution of healthcare spending to total consumer spending (blue line) and healthcare inflation’s impact on overall inflation (red line).

Healthcare spending has a large effect on inflation through two channels: first, during recessions healthcare spending does not decline as much as overall consumer spending; second, prices for health care services have grown faster than the prices of many other goods because the demand for healthcare services remains strong and constant. Since 1990 the prices for all goods and services have increased 81%, far less than the 131% of healthcare prices.
During recessions, total consumer spending falls and that puts downward pressure on inflation. But the healthcare component resists that downward pressure. The Federal Reserve, whose job it is to keep prices stable, might delay lowering interest rates because healthcare spending is keeping the price index elevated above the level of all other goods and services. This in turn could prolong the after effects of a recession: less lending and slower gains in employment. This is what happened after the 1990 and 2001 recessions. During the 1990 recession, inflation (the annual change in price) actually rose a bit before falling, spurred on by an 8.5% increase in healthcare prices. By the first quarter of 1991, healthcare was contributing 40% to overall inflation, rising up from 13% in 1989. The same pattern repeated in the 2000-2002 period.
Even though both recessions lasted less than a year, job recovery was slow. The lingering effect of a recession surely cost President George H.W. Bush a chance at a second term. In 1992, both Bill Clinton and Independent candidate Ross Perot reminded voters that the economy was sluggish and it was time for a change of direction. In the 2000s, Bush’s son, George, learned from his father’s misfortune. He urged the passage of tax cut packages and the Medicare Drug program, which helped secure his victory in the 2004 election despite disapproval of the conduct of the war in Iraq.
The ACA, or Obamacare, capped the growth of inpatient Medicare payments at 2% and this helped keep healthcare inflation (https://fred.stlouisfed.org/graph/?g=12f9b) at or below 2%. Medicaid expansion doubled the contribution level of healthcare prices to overall inflation, but because healthcare inflation was restrained, that helped to contain overall inflation.
The pandemic showed the enduring influence healthcare has on the general price level. When consumer spending had a sharp decline, healthcare prices remained strong. During the 3rd quarter of 2020, healthcare inflation was 2.9% and was responsible for nearly all of the general inflation rate of 1.1%. But here, the paths diverged. As the economy reopened and the general rate of inflation rose during 2021 and 2022, healthcare inflation decreased. That divergence describes the nature of the current overall inflation. It is procyclical, driven by short-to-medium term events, not a fundamental change in the economy.
In a 2017 Federal Reserve Economic Letter, Tim Mahedy and Adam Shapiro (2017) assigned spending categories into two buckets, procyclical and acyclical. Procyclical components that make up 42% of spending are those whose demand and prices vary with the business cycle and changes in employment. These include housing, recreation, food services and some nondurable goods. Acyclical components account for 58% of spending and include healthcare, financial services, many durable goods and transportation. The authors don’t mention energy specifically but I presume that it is an acyclical component of both housing and transportation services.
The pandemic caused shifts within and between these two buckets. During the pandemic demand soared for housing services, but declined for recreation and food services – an example of a shift within the procyclical bucket. We used a lot less energy in our cars but a lot more electricity and gas at home – a shift within the acyclical bucket. We bought a lot of durable goods – a shift between buckets.
I think it is the between shifts that had the most disruption. Supply chains for acyclical goods and services function on a less flexible timeline that does not anticipate sudden changes. Global shipping rates soared, ports were clogged with traffic, parts inventories were depleted, leading to manufacturing delays and an opportunity for companies to raise prices to make up for decreased profits due to shrinking volumes. With long delays from overseas suppliers, big retailers like Wal-Mart and Target increased their orders. As pandemic restrictions lifted, people shifted their spending again from acyclical durable goods to procyclical recreation and food services.
Each of us constructs an instinctive index based on our individual buying habits and circumstances. An American who lives for a while over in Europe has to learn to convert Centigrade temperatures to Fahrenheit. Like the CPI price index, there is methodology for making that conversion. However, it is much easier to remember that 0°C is cold, 10°C is cool, 20°C is comfortable, 30° is hot, and 40° is hell. Much of the time we navigate our daily lives without precision, relying on professionals when we do need exactness. Sometimes the professionals can tell us why something is the way it is but sometimes even they can only guess. Complexity is the result of an interlocking causality that is harder to solve than a Rubik’s cube.
///////////////////
Photo by John Barkiple on Unsplash
Mahedy, T., & Shapiro, A. (2017, November 27). What’s down with inflation? San Francisco Fed. Retrieved April 6, 2023, from https://www.frbsf.org/economic-research/publications/economic-letter/2017/november/contribution-to-low-pce-inflation-from-healthcare/