3 July 2014
- April 30 — ADVANCE estimate US Real GDP +0.1% from the fourth quarter of 2013 to the first quarter of 2014
- May 29 – SECOND estimate US Real GDP -1.0%
- June 25 – THIRD estimate US Real GDP -2.9%
The nearly 3 percentage point change is significant compared to the typical change from advance to third estimate of about 0.5% (See “GDP all over again — where estimates come from” 11 March 2014).
Markets were surprised. The surprise was more important to traders who can gain/lose immediately from news of any type. It was more of a chin-stroker to investors with a longer view (which sometimes is as little as a few hours for markets to digest information).
Why were some people more surprised than others?
Because some people dig more into data than others.
Steps to better digging:
- Know what comprises GDP. Start with the press release. From where? http://www.bea.gov/newsreleases/national/gdp/2014/gdp1q14_3rd.htm, click on the “Interactive Data” link (right hand navigation bar) and explore.
- Know what source data the US Bureau of Economic Analysis (BEA) uses to estimate each line item and methods used for those estimates. From where? Primarily, the National Income and Product Accounts (NIPA) Guide, Concepts and Methods, Chapter 3. http://www.bea.gov/methodologies/index.htm#national_meth
- When updated source data is released before the next GDP release, plug the newer data into the GDP tables. Use your own estimates as needed. BEA even provides a template http://www.bea.gov/national/Index.htm, scroll down to “source data.”
There’s more, yet these 3 steps put you ahead of most people.
By happy circumstance, June was a good month for data miners because the BEA published an updated article on the GDP estimation process. “Gross Domestic Product and Gross Domestic Income, Revisions and Source Data” is a title only a data geek would love. Yet, Alyssa Holdren’s helpful article was a page-turner, knitting together materials from several BEA sources. Pie charts on Page 3 are a quick look at how estimates increasingly improve with more “Comprehensive” and “Direct Indicator” data. http://www.bea.gov/scb/pdf/2014/06%20June/0614_gross_domestic_product_and_gross_domestic_income.pdf
How would this have helped traders prepare for the third estimate?
To make this example easy, just look at “big stuff.”
- From prior releases, almost 70% of Real GDP is Personal Consumption Expenditures (PCE) and just under 70% of PCE is Services. Just over 25% of Services is Housing and Utilities. Another 25% is Health Care. Much has been written on housing and utilities, so let’s dig into health care.
- What’s the source data? From Ms Holdren’s article, it is Medical Expenditure Survey, part of the Quarterly Services Survey (QSS) from the Census Bureau. In the pie chart, this is “Direct Indicator” data. Click on the press release at http://www.census.gov/services/index.html.
- What does the data show? Spending was down 2.0 percent (± 0.8%). From Table 1a, hospitals are down 1.3%. Adjusting for inflation, both drop a bit more. A glance at other services shows widespread weakness.
Did you notice the timing? QSS released was 11 June – 2 weeks before the GDP third release.
Health Care’s influence is clear in the GDP release. In 4Q2013 it added nearly 1 percentage point to PCE; in 1Q2014 it subtracted nearly 0.25 point.