Traditional asset pricing models often overlook the impact of cost factors on stock prices. This study addresses this gap by examining the relationship between cost stickiness, a phenomenon where costs adjust less to downward than to upward changes in activity levels, and the informativeness of stock prices regarding future earnings. Cost stickiness is prevalent across various countries and cost categories, impacting operational efficiency and risk. It also significantly influences analysts' forecasting behavior, a key factor shaping investor perceptions and capital market efficiency. Analysts, as representatives of rational investors, play a crucial role in informing market participants about the implications of current earnings on future performance. Accurate analyst forecasts enhance stock price information content, promoting market efficiency and effective resource allocation. Conversely, cost stickiness increases earnings volatility and reduces forecast accuracy, potentially distorting investment decisions and market valuation. The research hypothesizes that lower cost stickiness increases stock price information content by reducing earnings volatility, facilitating better future earnings forecasts, and enabling more accurate valuation of company worth. The study uses the future earnings response coefficient (FERC) to measure the information content of stock prices, linking current returns with future earnings. The FERC model offers several advantages, such as comprehensively reflecting investor confidence in disclosed earnings information, capturing both short-term and long-term perspectives on earnings and stock return relationships, and effectively expressing stock price information content, even with considerable market noise. As a complementary measure, the paper also analyzes the relationship between cost stickiness and stock price synchronicity, which reflects the extent to which firm-level information is incorporated into stock prices. Lower synchronicity indicates a greater incorporation of firm-specific information, reflecting higher market efficiency. The choice of the Chinese capital market for empirical testing is driven by factors including investor reliance on analysts, relatively stable economic and financial conditions leading to predictable future profits, and the distinct cost management practices between state-owned and non-state-owned enterprises, offering opportunities for comparative analysis. The study employs annual data from 15,070 A-share companies in Shanghai and Shenzhen between 2009 and 2021 to investigate the relationships between cost stickiness, earnings response coefficients, and stock price synchronicity.

The paper draws upon a substantial body of literature on cost stickiness, earnings response coefficients (ERCs), and stock price synchronicity. Studies by Anderson et al. (2003), Banker et al. (2013), and Kama and Weiss (2013) provide evidence of cost stickiness, demonstrating that costs are less responsive to decreases than to increases in activity levels. This "stickiness" has been shown to affect operational efficiency and operational risk (Anderson et al., 2003; Yao, 2018) as well as analysts' forecasting behavior (Weiss, 2010; Ciftci et al., 2016). The influence of cost stickiness on the accuracy of analysts' earnings forecasts is widely recognized (Balakrishnan et al., 2004; Banker and Chen, 2006; Weiss, 2010). The authors note that analysts' forecasts greatly influence investor decisions and market efficiency (Gleason and Lee, 2003; Jegadeesh and Kim, 2006; Cheng et al., 2016; Huang et al., 2016). Prior research highlights the impact of earnings response coefficients on investor forecasting and confidence (Hayn, 1995), capturing the value relevance of earnings (Choi et al., 2019). Stock price synchronicity, often used to gauge market efficiency (Roll, 1988; Morck et al., 2000; Durnev et al., 2004; Qiu et al., 2020; Zeineb et al., 2022), reflects the incorporation of firm-level information into stock prices. High synchronicity indicates limited firm-specific information and lower market efficiency. The paper leverages existing models for measuring cost stickiness (Weiss, 2010), stock price synchronicity (Morck et al., 2000), and ERCs (Ayers and Freeman, 2003; Piotroski and Roulstone, 2004; Choi et al., 2019) to test its hypotheses.

The study utilizes a quantitative approach, employing data from the CSMAR and iFinD databases. The sample comprises 15,070 A-share companies in China's Shanghai and Shenzhen stock exchanges from 2009 to 2021, focusing on manufacturing firms to minimize measurement errors due to potential pricing effects. Cost stickiness is measured using Weiss's (2010) method, which calculates the difference between cost reduction rates during periods of declining activity and cost increase rates during periods of rising activity. This method employs changes in sales revenue as a proxy for activity changes. The measure, denoted as ACostSticky, considers total costs, accounting for changes in sales revenue and earnings. Outliers were removed by a 1% bilateral tailing treatment. Stock price synchronicity (Syn) is measured using the logarithmic transformation of R-squared from a market model regression (Morck et al., 2000). Two measures are used: Syn1 using weekly data and Syn2 using daily data. The study employs a FERC model to assess the relationship between current stock returns and future earnings. The basic model includes past (Xit−1), current (Xit), and future (Xit+1) earnings, as well as future returns (Rit+1), controlling for various firm-level characteristics. To test hypotheses, the model incorporates interaction terms between cost stickiness and earnings variables. Control variables include firm size, market-to-book ratio, earnings volatility, stock return volatility, and analyst coverage. Industry and year fixed effects are also controlled. Hypothesis testing is conducted using OLS regression for H1 and H2, and OLS and dynamic panel GMM regression for H3. Robustness tests involve using alternative cost stickiness measures, limiting the sample to competitive firms, including additional control variables, and employing PSM to match high and low cost-sticky firms. A mediation analysis, using the Sobel test, examines the role of earnings forecast accuracy in the relationships between cost stickiness and both the earnings response coefficient and stock price synchronicity. Earnings forecast accuracy is measured using the absolute value of the difference between analysts' consensus forecasts and actual earnings, scaled by the beginning-of-year stock price. Control variables include financial characteristics and measures of income uncertainty. The study controls for industry and year effects, and uses clustered standard errors. GMM dynamic panel method is also used to mitigate omitted variable issues.

The empirical analysis reveals several significant findings. First, consistent with H1, cost stickiness significantly reduces both the current and future earnings response coefficients (ERCs and FERCs). Higher cost stickiness implies lower information content regarding future earnings as reflected in current stock prices. Specifically, the coefficient of the interaction term between cost stickiness and current earnings (Sticky × Xit) is positive and significant at the 1% level, indicating a reduced ERC. Similarly, the interaction term between cost stickiness and future earnings (Sticky × Xit+1) is positive and significant, suggesting a lower FERC. This supports the hypothesis that cost stickiness decreases the value relevance of accounting information. Secondly, supporting H2, the negative impact of cost stickiness on FERC is significantly weaker for state-owned enterprises compared to non-state-owned enterprises. This highlights the difference in investor behavior concerning cost stickiness across different ownership structures. In line with H3, cost stickiness is significantly positively correlated with stock price synchronicity, suggesting it reduces the incorporation of firm-specific information into stock prices and lowers market efficiency. This holds true for both weekly and daily data, and remains robust across different model specifications. The mediation analysis confirms that analysts' earnings forecast accuracy plays a significant mediating role in the relationship between cost stickiness and both the ERC and stock price synchronicity. Higher cost stickiness is associated with lower forecast accuracy, which in turn reduces ERC and increases stock price synchronicity. Robustness checks, using alternative measures of cost stickiness, sample restrictions, additional control variables, and GMM dynamic panel estimation, confirm the main findings. PSM further enhances the robustness, mitigating potential selection bias.

The findings offer valuable insights into the impact of cost stickiness on the information content of stock prices. The reduced ERC and FERC in the presence of higher cost stickiness suggest that investors perceive firms with sticky costs as less predictable and less valuable. This is consistent with the idea that cost stickiness increases earnings volatility and hinders accurate future earnings forecasting. The weaker effect of cost stickiness on FERC for state-owned enterprises likely reflects the influence of factors such as political connections, soft budget constraints, and government support, which make their future performance less predictable based on current earnings. The positive relationship between cost stickiness and stock price synchronicity aligns with the notion that information asymmetry and higher cost of information processing for firms with high cost stickiness reduce the ability of the market to incorporate firm-specific information into stock prices. The significant mediating role of earnings forecast accuracy highlights the importance of analysts' forecasts in the valuation process. The results highlight the importance of considering cost stickiness in evaluating firm value and market efficiency.

This study contributes significantly to the literature by examining the impact of cost stickiness on stock price informativeness, showing that higher cost stickiness reduces the information content of stock prices about future earnings. The study also reveals differences in investor behavior concerning cost stickiness based on enterprise property rights. Future research might explore the impact of specific cost components (e.g., labor costs, materials costs) on stock price information content or investigate the effectiveness of various disclosure mechanisms in mitigating the negative effects of cost stickiness on market efficiency. Policy implications suggest promoting greater corporate transparency regarding cost behavior to increase the accuracy of analysts' earnings forecasts and improve market efficiency.

The study focuses on Chinese A-share companies, limiting the generalizability to other markets. The use of sales revenue as a proxy for activity levels may introduce some measurement error. While the study controls for many variables, the possibility of omitted variable bias cannot be entirely excluded. Furthermore, the study relies on analysts' consensus forecasts, which may not perfectly reflect all relevant information.