线上销售与未来股票收益

Online Sales and Expected Returns

本文通过分析相关上市公司在电商平台的线上销售数据,发现线上销售增长可以预测未来股票收益。根据线上销售增长率构建投资组合可以获得月均1.27%的超额收益,经三因子、五因子模型调整后收益率分别为1.40%和1.35%,并且该超额收益在较长时间内不会逆转。横截面回归结果显示,线上销售增长与未来股票收益显著正相关,并在控制其他市场异象因子后仍然显著。此外,本文还发现线上销售数据的预测能力主要集中在投资者关注有限、线上销售占比高以及套利成本高的公司,其投资价值来源于对公司未来基本面信息的预测能力。进一步研究表明,同时利用线上销售指标和营业收入指标进行投资可以获得更高的超额收益。在考虑业绩预告和业绩快报对线上销售指标预测能力的潜在影响后,结果依然稳健。

E-commerce has become an important component and driving force of China’s economic development. As a representative of the Internet economy, online sales show great potential. Against the background of the booming digital economy, this paper studies the value of online sales to predict future returns, using the sales data of the e-commerce platforms of listed companies.Why do online sales predict expected returns? First, online sales data provide incremental information to financial reports. With the rapid development of Internet technology, online sales have become an increasingly important business model. As the proportion of companies’ online sales continues to grow, online sales are increasingly relevant to the companies’ overall revenue. Online sales information is also instantly available. Therefore, it provides more timely and granular information on company operating performance than traditional financial statement data. Second, investors are not fully aware of the content of online sales information. According to the theory of limited investor attention, investors have limited time and energy, and they may not fully understand all of the available information in a timely manner, which creates a temporary pricing bias. The A-share market has a large proportion of individual investors and an intense, speculative atmosphere in which investors mainly pursue short-term interests and lack the non-financial information that reflects a firm’s fundamentals. Because online sales information is costly, this information may not be fully acquired and understood by investors in the A-share market.Using the online sales data of 275 companies from January 2015 to April 2020, this paper shows that online sales data predicts future returns. Hedge portfolios based on online sales growth earn a 1.27% monthly excess return. The three-factor and five-factor adjusted returns are 1.40% and 1.35%, respectively. The Fama and MacBeth(1973) regression results show that online sales growth is positively correlated with future stock returns when we control for other related factors.To further illustrate the value of online sales, we investigate the predictability of online sales over a longer period. The results show that online sales predict stock returns for the next two months. The cross-sectional analyses show that the predictability of online sales is more pronounced for firms with limited investors’ attention, a higher proportion of online sales and higher arbitrage cost. Furthermore, we find that the investment value of online sales stems from their ability to predict future firm fundamentals. Further analysis shows that combining online sales and operating revenue may earn higher expected returns. In the robustness test, the predictability of online sales is still significant when we consider the potential impact of management forecasts.Our study builds on and contributes to two strands of the literature. First, from the perspective of online sales, this paper shows the informational content of non-financial data, and it is supplemental to traditional financial statement information. Research has documented the investment value of non-financial information, but no study has examined the ability of online sales to predict stock returns. This paper expands the literature by investigating online sales data, which is an important form of non-financial information. Second, our study enriches the investing strategy research by showing that online sales information has investment value and that investors may obtain abnormal returns by constructing hedge portfolios based on online sales. From the theory perspective, the market anomaly based on online sales information challenges the efficient market theory. We take advantage of the unique big data on online sales in China. Our results have important practical value for improving the efficiency of the Chinese stock market and protecting minority investors.