Don't Compare Absolute Levels of GDP from Different Integrated Assessment Models

This article on Climate Progress claims that because world GDP is higher under the RCP 2.6 emissions scenario than under RCP scenarios with higher levels of carbon emissions that means that cutting emissions is good for the economy:

Nothing could be further from the truth. The reason GDP varies so much under the different scenarios is because each of the four scenarios was produced by a different modelling team using a different integrated assessment model. The RCP 2.6 team used the IMAGE model, which is a model that is relatively optimistic about the amount of economic growth that will occur under business as usual. In the EMF22 exercise it had one of the highest levels of GDP under the reference scenario. The other modelling teams used models with lower baseline scenarios for economic growth. This explains most of the difference. Our current research on modelling the costs of climate change shows that it's likely that models that find economic growth to be harder to achieve anyway also find that cutting emissions is harder. So probably they had to use a model like this to be able to model the RCP 2.6 scenario. In EMF22 most models couldn't model the more extreme scenarios.