Offshore Sales Networks

Offshore Sales Networks and Stock Return Predictability

AuthorsJohn Bai; Priya Garg; Chi Wan

Paper linkhttps://papers.ssrn.com/sol3/papers.cfm?abstract_id=3455426

The strategy invests in NYSE/Amex/Nasdaq-listed securities, excluding financial firms and those with stock prices under $1. To identify firms’ foreign sales activities, the paper follows the Hoberg and Moon (HM) methodology. This involves analyzing 10-K reports to capture mentions of foreign countries and examining the context of words near each mention (such as “export” or “customer”) to identify offshore sales activities. The firm’s output market distribution is derived from the frequency of mentions, and proximity between firms is calculated using cosine similarity. The strategy links offshore sales network returns to the proximity-weighted average monthly return of industry peers. Stocks are sorted into deciles based on lagged returns from their industry peers, going long on the top decile and short on the bottom. The strategy is value-weighted and rebalanced monthly, with offshore and accounting data from year t matched to stock returns from the following year.

The strategy leverages the fact that industry peers with overlapping overseas sales destinations are exposed to similar regional economic and political shocks, impacting their stock prices. A key reason for this inefficiency is the insufficient regulatory requirement for reporting detailed foreign operations. The complexity of dynamic offshore networks makes it harder for investors to identify offshore sales network (OSN) relations between firms, leading to slow information incorporation. Research shows strong return predictability across OSN-linked firms. Additionally, abnormal returns from the strategy cannot be explained by asset pricing models (CAPM, Carhart, FF3, FF5), and the alpha of the long-short strategy remains significant and stable.