Since the GDP report came out for First Quarter 2014 and there was much attention to this from many quarters, including in the press (see column by Los Angeles Times columnist Michael Hiltzik and column by Peter Orszag, to cite a couple examples).

In the last week or so, a couple more pieces have come out to weigh in on this.  For example:
As reflected in my previous post on this, the GDP report was based on preliminary data, and that has been confirmed from dialogue with analysts at the Bureau of Economic Analysis. In particular here are two documents that explain how the estimates are computed:

  • The first quarter 2014 estimates are explained in a document found here, and the document explains that "The acceleration in health care spending reflects additional spending associated with the implementation of the Affordable Care Act (ACA). Because the preliminary monthly source data used to estimate health care services do not reflect the effects of the ACA, additional information on Medicaid benefits and on ACA insurance exchange enrollments, as well as other related information was used to prepare the estimates of consumer spending on these services."
  • Furthermore the BEA explains their ACA estimates in another document found here. As this document explains: "to the extent that the ACA results in changes to demand for health insurance and health care services, these effects on consumer spending will be reflected in the final source data that underlie BEA’s estimates of personal consumption expenditures (PCE). For preliminary estimates made before the final source data are available, BEA will take account of available information on Medicaid benefits, ACA insurance exchange enrollments, and other related information in preparing the estimates of consumer spending on these services."
So it appears that the GDP numbers for health spending and utilization are based on estimates, and not actual utilization numbers or surveys.  So the suggests even more reasons to be cautious, and expect revisions as we move forward.