You can't compare across 4 year periods (General economic noise will swamp the signal), but instead look at declines in the run up. Basically take the first der

You can't compare across 4 year periods (General economic noise will swamp the signal), but instead look at declines in the run up. Basically take the first derivative.