SETI bioastro: Could It Happen Again? Long-Term Growth As A Sequence of Exponential Modes

From: Larry Klaes (
Date: Fri Apr 14 2000 - 11:12:20 PDT

From: Robin Hanson <>

I have substantially revised my paper predicting an economic "singularity."


                        Could It Happen Again?
           Long-Term Growth As A Sequence of Exponential Modes

                           by Robin Hanson

                First Version 9/11/98, Revised 4/12/00

Brad De Long has interpolated standard world product time series with older
population time series, to construct a history of world product from one
million B.C. to today. After modifying De Long's series to reflect more
recent estimates of prehistoric population, I model world product history as
both a sum of exponentials, and as a constant elasticity of substitution
(CES) combination of exponentials.

World product history since two million B.C. is nicely described as a CES
combination of three distinct exponential growth modes: "hunting,"
"farming," and "industry." If we count the slow exponential growth of animal
brains as an earlier mode, then each mode grew over one hundred times faster
than its predecessor. The CES parameters describe how much each mode
substituted or complemented its neighbors. "Farming" seems to have strongly
substituted for "hunting," making for a sharp transition, while "industry"
more complemented "farming," making for a smoother transition.

Perhaps it is not crazy to wonder if we might see yet another transition to
a faster mode. If so, and if modes are comparable in terms of how much the
economy grows when they dominate and how much faster new modes are, then
within the next century we are likely to see a new economic growth mode, one
where the doubling time is measured in weeks, not years. This is a fantastic
and radical prediction, to be sure, but perhaps worthy of further study.

Robin Hanson
Asst. Prof. Economics, George Mason University
MSN 1D3, Carow Hall, Fairfax VA 22030
703-993-2326 FAX: 703-993-2323

This archive was generated by hypermail 2b30 : Wed Mar 28 2001 - 16:07:53 PST