But you're still missing my point. You are looking at domestic production completely separate from total market (domestic + import).
I'd hazard a guess that with increased interdiction in the 80s due to the rise of cocaine a lot of the marijuana production moved to the US because it was easier to hide a grow op domestically than smuggle bales of pot across the border.
No need to imagine an export market for US grown marijuana to explain faulty interpretation of government guesstimates under that scenario.
OK. Fine. I can agree with that. But I don't think that was the author's point.
Let's just assign some numbers, illustrating your (our) point. Let's say in 1981 we were producing 1000 tons and importing 10,000 tons. Today, we are producing 10,000 tons and importing 1000 tons.
Adjust those number how you will, it still doesn't support the author's conclusion, "Gettman said the 10-fold increase in U.S. marijuana production, from 1,000 metric tons in 1981 to 10,000 metric tons in 2006, showed the country was failing to control marijuana by making its cultivation and use illegal."