Although Columbia County is technically part of the Portland-Vancouver MSSA, the reality is that most residents don't commute into Portland to work, and it certainly doesn't resemble any definition of an "Exurban" County, although there is a population of South County residents around Scappoose-Warren and St Helens that commute to Portland or Vancouver.
Instead, there are many residents that commute to work in Longview, Washingon (Cowlitz County) where there was a major ILWU strike against an international consortium that was hiring Non-Union cheap subcontractors from out of the area, despite receiving $200 Million in Tax breaks as part of "sweetheart economic development deal". It's a complicated story, but this resonated in Cowlitz and Columbia Counties, which had some of the biggest county level swings in Washington & Oregon respectively....
Obama outperformed his national average in both of these deeply Union counties in both '08 and '12, although in statewide elections, Republicans have been successful for lower down-ballot election races over the past few decades.
I read the entirety of your post, but I cut it down so we don't have massively long quotes.
Anyway, I do understand what you're saying. The swing is representative of a larger regional swing among white working-class voters to Trump. The question I have is how indicative is that of future trends. In other words, is the swing in that larger region just due to Trump or is it turning into Southwestern Oregon(/Far Northern California) that once voted Democratic years ago?
It might be noteworthy that Kate Brown did about the same this year as Kitzhaber did in 2014 in Columbia County and the statewide margin was only marginally better for Brown. I hope that's not a sign of things to come.