A glance over the last two years of statistics reported by the Administrative Office may lead you to think that the chances of getting a “pure” reversal are going down. Of all terminations on the merits across all the regional circuit courts, statistics for the “percent reversed” went from 8.9% in FY 2011 to 6.8% in FY 2012.  See Judicial Business 2012, Table B-5 and compare the information there to Judicial Business 2011, Table B-5. In the Third Circuit, those percentages went from 9.4% in FY 2011 to 6.4% in FY 2012.  The numbers for the Fifth Circuit went from 8.0% in FY 2011 to 5.4% in FY 2012 and the Ninth Circuit went from 11.8% to 7.7% in the same period. In fact, all but two of the 12 regional circuits saw declines of some sort; only the Seventh Circuit’s numbers went up, and then did so by a small amount.  The numbers for the Sixth Circuit were unchanged.

I suppose that a change in the mix of cases could explain part of these declines.  It’s not unusual to see the numbers bounce over time in a single circuit court, going up one year and going down the next, likely because of the mix of cases that court terminates.  But these bounces tend to cancel each other out in the overall numbers because some circuits see declines while others see upticks.  This explains why the percent reversed across all circuit courts hovered around 9.6% (plus or minus .6%) between FY 1997 and FY2011.  So it is surprising to see 10 of 12 circuit courts move in the same direction and to see the overall percent reversed to go down by over two percentage points.

So what gives?  What I found is one of the frustrations of dealing with reported statistics – unexplained changes in the way data are handled.  Bottom line – the percent reversed figures reported for FY 2012 are not comparable to the earlier percentages.

If you go through the online version of the 2012 Judicial Business report, you’ll find a notation to changes the Administrative Office made to the way its data are reported. The percent reversed numbers above come from Table B-5, and the 2012 Report explains that Table B-5 now presents “data on cases disposed by consolidation” whereas in prior years Table B-5 “did not provide such data.”

Ok, but that change doesn’t explain what happened to the percent-reversed figures.  That’s because the “consolidation” change affected neither of the numbers used to calculate those percentages.

What the Administrative Office doesn’t explain (at least I couldn’t find it) is that another change in the way it did things in the 2012 Report was to now treat certificate of appealability terminations as terminations on the merits. See that column labeled “Cert. Appealability” under the overall heading of “Terminations on the Merits” in Table B-5 of the 2012 Report? That column does not appear in Table B-5 in the 2011 or 2010 reports. In those reports – and in fact in all reports going back to 1997 – the “Cert. Appealability” column is in Table B-5A because those dispositions were considered procedural terminations, not merit terminations.  Why are certificate of appealability terminations all of a sudden merit terminations?  No clue.

The effect of the change, however, is to make the number of cases terminated on the merits larger, and as a result, to reduce the percentage of cases reversed.  So the 3 percentage point decline in the Third Circuit probably has more to do with the fact that certificate of appealability terminations represented over 14.5% of the total number of dispositions on the merits than anything having to do with how judges see things. Taking certificate of appealability terminations out of the base leads to a “revised” percent reversed of 8.3% for all circuits, and of 7.5% for the Third Circuit – both still smaller than in FY 2011, but not by huge margins.  The revised percent reversed for the Ninth Circuit is 11.25%, a mere .5% decline from the FY 2011 figure.  Although the change in the way certificate of appealability terminations are treated doesn’t explain all the declines in the reported statistics (e.g., the percent reversed in the D.C. Circuit went from 15.0% to 8.9%, but that court had only 7 certificate of appealability terminations in 2012), it does explain a sizeable portion of them.