I think that fudging the numbers to bolster your pet theory is not an acceptable way of looking at this data.
> ~18% of their working age people *do not have jobs*
Which is a correct interpretation of participation rate. His theory on the causes may be off, but his numbers weren't
In general, either is fine by me as long we are consistent: they are both proxies for percentage of people needing work and should correlate to a large extent.