reader bruce hall Comments for his calculation of the 18-month change using the non-seasonally adjusted CPI data, and then using the 6-month seasonally adjusted data to calculate the 18-month change:
…pgl sticks with 13.3%, which is what I wrote when I originally commented on this “price level change” based on unseasonally adjusted data. After he “revised my thinking,” I used the Fed’s seasonally adjusted data, which is 12.6% from January 2021 to June 2022. Of course, we all know that the end point of seasonally adjusted data is corrected, because the seasonal factor has to be recalculated over time. So maybe in a few months that number could be 12.4% or 12.8%. But the question is not the exact percentage change, but the general magnitude of that change and how it relates to subsequent (June-December) changes. June seems to be a discontinuity that doesn’t normally appear in year/year numbers, but has to be collected.
Well, I’ve pointed out the dangers of his 18 month changes to NSA data here; but what about adding nsa and sa data?
Well, here’s an example (from Scott Walker’s campaign, In a 2016 post) not only add different series, but also different seasonally adjusted (or unadjusted) data.
especially Walker race in trouble for Wisconsin governor when they tout Job creation figures obtained by adding together the non-seasonally adjusted employment figures (from the so-called Quarterly Census of Employment and Wages) and seasonally adjusted employment data (from establishment surveys) to obtain cumulative changes in employment. (They do this because the QCEW data lags many months, while the agency survey data is more timely). As shown in Figure 3.
image 3: Wisconsin nonfarm private employment from the Quarterly Employment and Wages Census, not seasonally adjusted (blue), private nonfarm private employment from the Survey of Establishments, seasonally adjusted (red). Black arrows indicate changes in QCEW and firm survey data; cyan arrows on institutional surveys. Source: U.S. Bureau of Labor Statistics.
Note that it is possible to calculate the change from December 2010 (just before Walker took office) to March 2012 (the latest QCEW data as of December 12, 2012), and then add the change from March 2012 to October 2012 Monthly change (latest available figures as of December 12, 2012). That is, adding 89.1 to 6.4 yields 95,500, which is close to the 100,000 figure cited by Governor Walker’s campaign. You can see why Gov. Walker’s campaign officials did this — the correct calculation using changes in agency surveys from December 2010 to October 2012 was only 61,100.
So, Bruce Hall, if not in “good” company, is in the company of the Scott Walker campaign.
