Captain Obvious weighing in: The number of charges decreased in every category on that chart. Although they're not pictured in the chart, the number of retaliation claims also dropped (to 37,955). The only charges that actually increased were GINA charges (to a whopping 333).
On Twitter, I wondered aloud, "I'd like to see statistical regression analysis comparing EEOC charges to economic indicators. I bet there are some strong correlations." Well guess what? I crunched some numbers and found a pretty strong correlation.
I used the total charges per year as reported by the EEOC (linked above), and the unemployment rate from January 1 of the following year as reported by the Bureau of Labor Statistics (i.e. I used the unemployment rate on January 1, 2015 as the unemployment rate for 2014 - we could quibble about whether a mid-year or aggregate rate would be better, but I only have so much time to devote to a blog post, ya know?).
The results? Excel calculated a correlation of 0.895379655 (0.0 being no correlation and 1.0 being a perfect correlation - in other words, there exists a pretty darned strong correlation). I charted the numbers here (you can see the correlation easily with the naked eye - but note they are plotted on different scales):
Pretty interesting stuff if you ask me. Raw data below: