Thursday, May 8, 2014

The latest data from the Bureau of Economic Analysis (BEA) of the U.S. Department of Commerce estimate that spending for health care services jumped by about $50 billion in the first quarter (Q1) of 2014 compared to fourth quarter (Q4) of 2013 (seasonally adjusted annual rates). This translates to an annual rate of increase of more than 10%, a figure that has been widely cited and discussed.[1]

Some will interpret this as an indication that we are on track for a 10% rise in health spending over the full year. This is NOT a correct interpretation, and it is important to explain why. Moreover, this 10% figure conflicts with the 7% growth that we are reporting in our latest Health Sector Economic Indicators; this, too, warrants explanation. Finally, I have a few thoughts related to the preliminary nature of the BEA estimates.

(1) Why does a 10% annual rate for Q1 not suggest 10% growth for the year? The basis for the 10% annualized rate of growth is the estimated $50 billion increase between Q4 of 2013 and Q1 of 2014. The annualization of this figure imagines that this rate of increase will continue for the remaining three quarters of 2014. In other words, spending in each quarter would be roughly $50 billion higher than in the preceding quarter. This means that spending in Q4 of 2014 would be about $200 billion greater than spending in Q4 of 2013, an increase of roughly 10%.[2]

However, the $50 billion increase in spending in Q1 is NOT a good indicator of growth during the remaining quarters, because it reflects a surge in the number of people with health insurance due to the Patient Protection and Affordable Care Act (ACA) that will not be repeated in subsequent quarters. Discussions with BEA analysts suggest that more than half of the $50 billion growth estimate is attributable to expanded coverage. ACA enrollment is, for the most part, closed for the rest of the year, and therefore no additional surges of this magnitude should be expected. For the sake of illustration, suppose that exactly half the growth in Q1 was due to expanded coverage and that no additional expansion was anticipated after Q1. Then we might expect $25 billion in growth in each of the remaining three quarters. In this case, Q4 of 2014 would be $125 billion greater than Q4 of 2013, an increase of about 6.25%.[3]

In short, because the surge in numbers of insured individuals will not be repeated in subsequent quarters, the 10% annualized rate of growth in spending is NOT a good indicator of how the full year is likely to play out.[4]

(2) What formula produces the 6–7% growth rates in our HSEI health spending estimates for early 2014? There are a number of reasons why the health spending growth rates shown in our latest spending brief differ from the 10% figure under discussion. For one thing, our health spending estimates include all of the spending categories in the official national health expenditure accounts. But the main reason is that our annualized increase is based on comparing spending in the current month to spending in that same month from the prior year. Thus, the 7% increase shown for March 2014 indicates that spending is 7% higher than March 2013. This seems to be a more reasonable indicator of our current growth path than the 10% annualized rate discussed above and may even be a bit high (see below).

(3) Is the acceleration in spending real, or will future data revisions show slower growth? I am quite conflicted on this issue. On the one hand, the pattern in our data (see below) is in line with expectations as shown here. The timing seems about right for a bounce-back from the effects of the recession (late 2013), and then there is the additional jump in early 2014 attributable to the newly insured.

Spending Chart

However, the continuation of very low health care price inflation through March 2014 (about 1%) means that the acceleration in spending must be driven by utilization. While information on current health care utilization is scarce, most reports emphasize stagnant utilization rather than growth. Furthermore, there has been no acceleration in health employment. My colleague Ani Turner has noted that ACA-expanded coverage could drive spending up without any change in prices or utilization via shift from uncompensated to compensated care. Also, there is some evidence of utilization growth in states expanding Medicaid coverage. So I guess that the jury is still out. BEA estimates for Q1 will be importantly updated in mid-June with data from the quarterly services survey. The date has been marked on our calendar.

Speaking of dates, our annual symposium on sustainable health spending will be held on July 15. Among other things, the data revisions will be viewed and discussed.  For information on registration, contact Paul Hughes-Cromwick.


[1] The most widely quoted figure has been 9.9%, but this refers to “real” spending that has been adjusted downward to remove the effects of inflation (

[2] Health services spending in Q4 2013 was about $2 trillion. This is not to be confused with national health expenditures, which are currently running at about $3 trillion and which include categories of spending beyond health care services.

[3] This is not my personal forecast. The main point here is to convince you that it makes no sense to interpret the 10% annual growth in Q1 as a prediction for the full year.

[4] Peter Orszag makes this point in a recent post.

All postings to the Health Policy Forum (whether from employees or those outside the Institute) represent the views of the individual authors and/or organizations and do not necessarily represent the position, interests, strategy, or opinions of Altarum Institute. Altarum is a nonprofit, nonpartisan organization. No posting should be considered an endorsement by Altarum of individual candidates, political parties, opinions or policy positions.