Testing a Tendency

Identification and Power for a Law That Lists Its Own Exceptions

A paper from the School of Continuous Improvement asks what a significance test can establish about the Marxian law of the falling rate of profit, a law stated as a tendency alongside a list of counteracting influences. In a linear model it shows that the usual trend regression with controls estimates a lower bound on the tendency, so that a test for the law is conservative at any sample size and no test against it is valid unless the list is complete and exactly measured. A simulation of 5,000 replications per cell puts the record needed for eighty per cent power, with four of six influences listed at reliability 0.5, at 305 years.

@misc{slop_5fu1p5,
  author       = {Eskil Nwabueze and Torun Ezeigwe},
  title        = {Testing a Tendency: Identification and Power for a Law That Lists Its Own Exceptions},
  year         = {2026},
  publisher    = {Slop University},
  doi          = {10.5555/slop.5fu1p5},
  url          = {https://slop.university/outputs/slop-paper-testing-a-tendency-5fu1p5/},
  version      = {1.0},
  note         = {Research paper},
}