A swedish kind of death:

I am guessing (and I would appreciate being corrected if I am wrong) that GDP emerged or at least became important around the time that computerised records of transactionbased taxes came into existence.

GDP emerged out of the work of Simon Kuznets in the 1920's, I believe. Back then it was called National Income. In a way, from the point of view of tax base, replacing the hard to measure National Wealth with the easier to measure (transaction-based) National Income (which measures not wealth but economic activity) makes it easier to tax income than to tax wealth.

Kuznets is credited with revolutionising econometrics, and this work is credited with fueling the so-called Keynesian "revolution". An important book of his is National Income and Its Composition, 1919-1938. Published in 1941, it contains a historically significant work on Gross National Product. His work on the business cycle and disequilibrium aspects of economic growth helped launch development economics. He also studied inequality over time, and his results formed the Kuznets Curve.

...

There are two developments at Kuznets time: the emergence of econometrics and the Keynesian Revolution, both of which found in Kuznets's data an important resource for their advancement. Kuznets, however, was neither a Keynesian nor an econometrician - he took his cues from Mitchell's Institutionalism - as exemplified in his 1930 methodological pieces. Whereas Mitchell devoted his life to the study of business cycles, Kuznets turned to other fluctuations - seasonal ones and secular movements - then to national income estimation, and later to studies of economic growth. As a result, his initial work was on the empirical analysis of business cycles (1930) - a 15-20 year cycle he identified was later attached to his name, the "Kuznets Cycle".

It'd be nice if the battle were only against the right wingers, not half of the left on top of that — François in Paris