Unexpectedly, Part 1

[Note: this post has been updated to reflect a critique posted on social media.]

The monthly unemployment numbers issued by Minnesota’s Department of Employment and Economic Development (DEED) are much anticipated as a barometer of the state’s economic health.

This month's report contained the disappointing news that Minnesota lost 4,200 jobs during the month of July and the further bad news that the June number had been revised downward by 3,600.  In summary, we now believe that 7,800 fewer Minnesotans are employed than we thought were employed a month ago.

For its part, the agency was quick to distance itself from its own numbers,

According to DEED officials, recent divergences due to seasonal adjustment lessen confidence that Minnesota’s seasonally adjusted estimates are accurately reflecting recent trends in job growth.

Funny, last month DEED officials expressed no concerns about the data behind the headline “State Gains 8,500 Jobs in June.”

The latest downward revision fulfills a prediction I made last month and marks the fifth consecutive downward revision, as shown on this table,

Reporting	Reported	Later	Actual
Month	Gain	Correction	Gain
July	(4,200)	tbd	(4,200)
June	8,500	(3,600)	4,900
May	10,300	(3,100)	7,200
April	(4,200)	(1,100)	(5,300)
March	2,600	(700)	1,900
February	(100)	(1,100)	(1,200)
January	600	200	800
Total	13,500	(9,400)	4,100

The odds of randomly getting five consecutive downgrades works out to 1 in 32, or only a 3.125 percent chance.  I don’t know about you, but in my world, those odds work out to be a long shot.

[Note:  Twitter commenter Paul (@PJGJ03) suggests that these downward revisions are not random but follow a regular annual pattern, with the consecutive downward revisions ending in August, before turning positive in September.  Which means:  brace for two more months of downward revisions.]

In the last twelve months, DEED reports that Minnesota created 68,344 jobs—with only 4,100 created in the seven months of calendar year 2014, after initial revisions.  This recent trend would seem to indicate something other than seasonal adjustments going on.  The Minneapolis Star Tribune is reporting that the fully revised figure for the year so far works out to only 2,900 jobs gained.

[If Paul is right, The 2,900 figure for 2014 jobs may be overly optimistic.  If the expected downward revision of July's job loss is as big as either of the last two monthly revisions, then I don't even want to think about it.  For his part, Paul thinks the July revision will come in at -1,500 and could range from -500 to -3,000.]

If that wasn’t bad enough, the Star Tribune reports that the state’s workforce participation rate has fallen to 70.1 percent, the lowest level in more than 30 years.

As he runs for re-election, Democrat Gov. Mark Dayton’s political allies have been quick to attribute the state’s alleged low unemployment rate to Dayton policies.  Now that the numbers are heading in the opposite direction, we’ll see if the tune changes.

In Part 2, I dive deeper into the numbers.