Measuring Economic Growth with a Fully Identified Three-Signal Model

Working Paper: NBER ID: w31517

Authors: Andrea Civelli; Arya Gaduh; Ahmed Sadek Yousuf

Abstract: We augment Henderson et al. (2012)’s two-signal model of true GDP growth with a third signal to overcome its under-identification problem. The additional moment conditions from the third signal help fully identify all model parameters without ad-hoc calibrations of the GDP’s signal-to-noise ratio. We characterize the necessary properties of the third signal. Using the model, we recover the optimal weight of official GDP in the composite true GDP growth estimates, which varies with the quality of the national statistics. The model improves on existing methodologies that use signals to measure true income.

Keywords: Economic Growth; Three-Signal Model; Urban Land Cover

JEL Codes: E01; O11; O47; O57


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
Third signal (urban land cover data) (R14)Identification of model parameters (C51)
Urban land cover data (R14)Economic growth (O49)
Reliability of GDP data (E01)Relative weight of GDP data in economic growth estimates (E01)
Third signal (urban land cover data) (R14)True income growth estimates (O49)

Back to index