Working Paper: CEPR ID: DP15503
Authors: Bernard J. Dumas; Tymur Gabuniya; Richard C. Marston
Abstract: The distinction between domicile and place of business is becoming more and more relevant as a growing number of firms have activities abroad. In most statistical studies of international stock returns, a firm is included in a country's index if its headquarters are located in that country. This classification scheme ignores the operations of the firm. We propose, instead, to measure the firm's exposures to "geographic zones" according to the place where they conduct business. As a representation of "geographic risks", we synthesize zone factors from all firms in the dataset, be they domestic firms or multinationals. And we show the properties of the exposures to the zone factors.
Keywords: stock return indexes; stock return exposures; geographic investing; factor models; country factors; expectations-maximization algorithm
JEL Codes: C4; F3; F6; G1
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| geographic zones of operation (R32) | stock returns (G12) |
| revenue shares in geographic zones (R12) | risk exposures (G22) |
| geographic exposure of firms (F23) | revenue distributions across different zones (D39) |
| traditional methods of measuring business risks (G32) | misleading conclusions about risk exposure (G41) |