In the search for more information many feel that real time data is better than batch. Just a few weeks ago I sat in a case study presentation by GE Aviation, at our recent US Enterprise Information and MDM Summit, where the speakers explained how real time data from monitoring devices fixed to engines provided new ways to predict failures. Apparently these ‘new ways’ had not been possible with the previous batch or snap-shot approach to collecting data every few minutes. So real time is better.
Not so fast. In the US print edition of the Wall Street journal today there is an article in the front page of the Money and Investing section titled, “Fed Fights Itself Over GDP Data“. It’s an interesting story. Basically GDP is an estimate of the growth of an economy. The Federal Reserve publishes this data point periodically for the US economy. As time passes the estimate for a given period is updated, based on more accurate data and, sometimes, improvements in the estimating process itself. However even this effort has challenges.
The GDP estimate is just that- it is not a national survey of actual production. It is also an estimate based on an amalgam of data from various sources including other parts of the Federal Reserve system of banks. In any given period should one source be late as part of the data collection cycle, the process proceeds without the data feed and an approximate data feed is substituted. So the estimates themselves are really estimates, and it is quite fair and acceptable that revisions take place. However some revisions have been significant, and at times the data, or a revision, can have significant impact.
When GDP itself is reported at such low numbers, as in close to zero, or if it were contracting or changing direction (as in negative to positive, or vise versa), national news, politicians and economists are all over place react. A change in direction or a monthly report that accentuates, or refutes, a negative (or positive) trend can be taken to mean a change in our collective fortunes going forward and policy changes like interest rate setting can take place. So we are not talking of small beer here.
The article in question reports on the fact that several federal reserve banks offer up additional, more timely estimates for GDP. The Federal Reserve Bank of Atlanta publishes its GDPNow whenever one of its data feeds is updated. So it is not a monthly update but more of a net-change or event-based update. The Federal Reserve Bank of New York plans to publish its own GDP results on a weekly basis, taking into account new data that week. These two reports are referred to as GDP trackers as they are not the official GDP estimate but are signals for the likely direction of the official report.
Being more real time such GDP trackers are likely to be more sensitive to the underlying data. Since revisions to the slowly complied GDP analytic are coming, revisions to the more real time trackers must also be expected. That is accepted. But what is at odds is that the current GDP trackers from the Federal Reserve Banks Of New York and Atlanta are completely at odds with the current official GDP report. As the article reported, New York’s Q1 GDP estimate (FRBNY Staff Nowcast) is at 1.1%, while Atlanta’s Q1 GDP estimate (GDPNow) is at 0.1%. The official estimate will not be published until June.
So what was already a nervous number with inherent perturbation from the estimating process itself, is fast becoming more nervous and even creating conflict in the data, let alone the conclusions drawn from it. The alarmist title of this blog might have gotten your attention, and I am not sure my alternative title of “Nervous Numbers, the Fed and You” would have grabbed you the same way. There clearly are no secrets or lies at the Fed. But there is inherent risk in the dynamics of data collection and the process for estimating. Given our use of this data point (GDP) and the credibility of our news reporting systems, I wonder if we can really exploit the data for increased value. Worse, when two parts of the official system produce two different versions of the truth when, in this case, they are the sole arbiter of truth, one wonders how anyone can get more value from the effort.
View Free, Relevant Gartner Research
Gartner's research helps you cut through the complexity and deliver the knowledge you need to make the right decisions quickly, and with confidence.Read Free Gartner Research
Comments or opinions expressed on this blog are those of the individual contributors only, and do not necessarily represent the views of Gartner, Inc. or its management. Readers may copy and redistribute blog postings on other blogs, or otherwise for private, non-commercial or journalistic purposes, with attribution to Gartner. This content may not be used for any other purposes in any other formats or media. The content on this blog is provided on an "as-is" basis. Gartner shall not be liable for any damages whatsoever arising out of the content or use of this blog.