We find that the composite leading index (CLI) is useful for forecasti
ng gross national product (GNP), both in sample and in an out-of-sampl
e real-time exercise. We propose a nonlinear specification in which cy
clical shifts of the CLI precede those in GNP. However, we find that b
etter forecasts are provided by a simple linear relation between curre
nt GNP growth and the growth rate of the CLI during the previous quart
er along with an error-correction term corresponding to the previous q
uarter's logarithmic difference between the level of the CLI and the l
evel of GNP.