This letter provides strong evidence of nonlinear mean reversion in the US
debt-GDP ratio using data since 1916 and shows that the difficulty to detec
t mean reversion encountered by a number of researchers may be due to the i
nvalid maintained hypothesis of linearity which is common to the testing pr
ocedures employed in this context. (C) 2001 Elsevier Science B.V. All right
s reserved.