You cannot group the marginal cap ex (which has nothing to do with non-land) with the total cap ex. There is no doubt that all of cap ex makes a good return. That is not the issue. The issue is about the return on only the cash received from issuing shares: $15M from 30M shares at $.50.
Even fish farms pay back in 2 1/2 years according to Solomon. That's a 40% return -- very, very good.
So where exactly do you get 80%?
But let's accept that hypothetically. They have to make $12M on the $15M. Let's say they actually do that first year. It still dilutes earnings per share! Say they make $100M with 100m shares. That's $1.00 per share. With your 80% (unbelievable) return, they make $112M on 130m shares, or $.86.
And the return is not multiplicative. You can't just add the second year earnings, as the share count remains 130m. They must continue to make $12M on that investment to keep eps at $.86.
That's why the policy must change, as there is evidence that it is or will.
Of course, the math changes by a factor of 3 if they issue above $1.50.