Regardless of the units you use, you won't get an exponential curve. You can fit an exponential curve to the data. You can also fit a logistic curve to the date.
Heck, you can even fit a linear curve to the data.
This line of thinking can lead to catastrophic consequences if the long-term implications are not well understood. Short-term exponential growth can lead to a catastrophic crash down the road if the debt, resources, or social fractures are not sustainable in the long run.
Things that appear exponential locally usually turn out to be logistic or cyclic.
I would like to say that I'm surprised that an economist doesn't understand math, but I can't.