Economist here. To defend conventional GDP calculations for a moment, it’s worth keeping in mind that when GDP measures were first invented the data necessary

Economist here. To defend conventional GDP calculations for a moment, it’s worth keeping in mind that when GDP measures were first invented the data necessary to compute something like what is proposed in the article wasn’t even a pipe dream. (The original ideas predate the 20th century.) The data simply did not exist and was not even foreseen in anyone’s wildest imagination. Even computing conventional GDP measures 75 years ago took a lot of work, a lot of it on paper. (In fact one of the earlier innovators in GDP calculations warned about treating it as a welfare measure way back in the 1930’s! The critique here is not new.) You can sell a measure with a lot of individual heterogeneity like the one proposed in this note to economists. (You don’t need to dress it up in superfluous mathematical machinery like a geometric Brownian motion either.) But this is stuff that has become possible only recently thanks to advances in data availability and computing power. It’s also worth noting that the data itself is only available to a tiny fraction of researchers. If you are the IRS and have a good picture of everyone’s income at a very precise level, you can compute this measure. (It is not easy once you get into the details!) If you are the Bureau of Economic Analysis, which computes GDP, you probably need permission from Congress to access that data for that purpose.