Aggregate Earnings Information And Discount Rate News

Accounting earnings,as one of the core contents in company information disclosure,is important reference for decisions made by market participants.For a long time,the academic circles mainly focused on the information content of the accounting earnings of an individual firm at the micro level,and paid less attention to the aggregate accounting earnings,that is,the accounting earnings of all firms was aggregated by a certain weight.Recently,the study found that there is a weak or even negative correlation between the aggregate earnings and the stock market returns,which is quite different from the conclusion that the accounting earnings and stock returns are significantly positively correlated at the firm level.It shows that the aggregate earnings has different information content from the earnings of individual firm,and the aggregate earnings may have the “macro-economic” information content.Exploring whether the aggregate earnings convey the discount rate news is exactly what this paper tries to answer.Firstly,according to the decomposition model of Campbell(1991),the paper analyzes the theoretical logic that how the discount rate news and cash flow news impact the rate of return,and then analyzes the theoretical logic that how the discounted rate news conveyed by the aggregate earnings affects the stock return.The analysis found that the change in stock return is impacted by the two shocks of cash flow news and discount rate news.However,since the impact of the two kinds of shocks on stock return is different,the relationship between aggregate earnings and stock market returns may be different.It could be positively correlated,negatively correlated or weakly correlated,depending on the relative strength of this two information components.The paper then uses the Earnings Response Coefficient Model(ERC Model)for empirical testing.The basic ideas are as follows.(1)Calculate the aggregate earnings change of the market as an proxy for the aggregate earnings news of the market as a whole.The sample selects the quarterly earnings data disclosed by the Shanghai and Shenzhen A-share listed companies from the first quarter of 2003 to the fourth quarter of 2016.We calculates the earning change of each company,and then use the market value,circulation market value,etc.as the weight to make a weighted average.(2)Process the basic ERC regression,that is,the regression of the stock market return on the aggregate earnings.The regression coefficient of the aggregate earnings is significantly positive,which is 0.2382(the market-value-weighted aggregate earnings),indicating that the cash flow news component may be higher than the discount rate news component in the news conveyed by the aggregate earnings.(3)Extend the ERC regression,add the proxy variables of the discount rate(including real risk-free interest rate,inflation rate and risk premium)in the basic ERC model to separate the impact of cash flow news and discount rate news.It is found that the regression coefficient of the aggregate earnings significantly increases(relative to the coefficient of basic ERC regression),from which 0.2382 to 0.4461(the market-value-weighted aggregate earnings),and the level of significance also increases.Besides,the three discount rate proxies are also significantly negative.The result shows that the discount rate proxy variable significantly separates the discount rate news and makes the cash flow news more "prominent",which proves empirically that the aggregate earnings convey the discount rate news.At the end of the paper,a supplementary test of the relationship between the aggregate earnings and the discount rate was carried out.The result shows that the aggregate earnings is significantly positively correlated with the inflation rate and the risk premium,which demonstrates the significant correlation between the aggregate earnings and discount rate proxies.The main contribution of the paper is that it provides direct evidence that the aggregate earnings convey the discount rate news,which has a significant marginal contribution to the study of the “macro-economic” information content of earnings.