The Impact Of Investor Attention On Stock Pricing

Classic finance theories state that market information is incorporated into stock prices timely and accurately.The implied assumption under this statement is that investors can obtain and process infinite information at the same time.However,Investor attention is a scarce cognitive resource.Because of the information explosion brought by the internet,investors pay attention to information selectively and ignore other information.For this reason,investors' attention allocation affects the speed of information diffusion,and in turn affects asset pricing.Therefore,studies on the impact of investors' attention on asset pricing is helpful for understanding the real information diffusion process and investors' asset allocation,and is meaningful for the decision making of market participants such as market regulators,analysts and managers etc.Though there is rich literature in the field of investor attention,questions like how would investors' attention to stocks be affected by geography distances and firm operate characteristics,how would investors' attention affect stock pricing at industry level and aggregate market level are still open to be studied.For this reason,this thesis analyses the impact of investors' attention on China A-Share stock pricing at firm level,industry level and aggregate market level respectively,and tries to explain market phenomenon such as local bias,post earnings announcement drift,price momentum,market volatility,aggregate market risk aversion and so forth from the perspective of investor attention.To be more specific,the main content of this thesis is as follow:First,this thesis uses the percentage of local investors' internet search volume on stock names to that of national investors to measure at what extent investors over-allocate their attention and assets to local stocks.With this measurement,this thesis empirically tests the impact of investors' local attention on A-share stock prices,and exams the causes of investors' local bias on asset allocation.It's found that on average the percentage of internet search volume initiated by local investors is 5.73 times of their population percentage,which indicates that investors in A-Share markets over-allocate their attention to local stocks notably.Investors in less developed provinces pay more attention to local stocks.Stocks with smaller sizes,lower book-to-market ratios,lower turnover,lower debt-to-asset ratios,less earnings,fewer shareholders and fewer staff are more attractive to local investors.It's also found that investors' local attention increases individual stocks' risk premium,enhances their price co-movement with stocks in the same market,the same industry or the same geographic region,and decreases their pricing efficiency.All these findings are consistent with the irrational behavior hypothesis,and prove that local investors in A-Share markets do not have significant information advantages.Second,this thesis measures firm transparency by a composite index constructed with earnings quality,related transactions,audit quality and information disclosure ratings,and finds that less transparent firms have lower analyst coverage,lower media coverage,less institutional ownership and less internet search volume.In other words,less attention is paid to less transparent stocks.This thesis then tests the effects that firm transparency impact on individual stocks pricing through investor attention by examining the relationship between firm transparency and post earnings announcement drift.It's found that less transparent firms have fewer abnormal trading volume,higher earnings delay response ratio and stronger drifts.When multiple firms release earnings announcement on the same day,more transparent firms show weaker drifts.To sum,this thesis shows that firm transparency would affect asset pricing through investors' attention allocation rather than information uncertainty.Third,with a theoretical model,this thesis derives that investors' inattention to information causes industry information only diffuse gradually.Because of the gradual information diffusion within industry,returns of stocks with less attention are correlated with past returns of stocks which are from the same industry but are with more attention.Investors receive information originating from relatively neglected industries only with a lag.As a result,the returns of industry portfolios that are informative about macroeconomic fundamentals will lead the aggregate market.The slow react to industry information would also cause price momentum at industry level.The empirical study in this thesis supports these theoretical hypothesizes.In A-Share markets,the returns on the top 30% attentive stocks of an industry would lead the returns on the top 30% inattentive stocks in the same industry,but not vice versa.Returns on mining industry portfolio and other three industry portfolios are informative about the aggregate market return.The more these industries are neglected,the more their stock returns are informative.Controlling market premium,size premium and value premium,a long-short portfolio which longs industry winners and shorts industry losers obtains significant yearly positive excess returns of 6%-7%.Fourth,this thesis measure investors' aggregate market attention with the Baidu search volume on representative stock market indices.It's found that investors' aggregate market attention correlates with aggregate market volatility,trading pattern and risk aversion level.A two-way causality exists between aggregate market attention and aggregate market volatility.Including the proxy of aggregate market attention reduces up to 36% market volatility in-sample forecast error.Individual investors make attention-driven trades at aggregate market level.The more attention individual investors pay to the stock market,the more shares they buy on the next trading day.This effect is more pronounced when the market reaches a new high.Using the difference between implied variance and expected realized variance of market indices,known as variance risk premium,as market-level measures of risk aversion,this thesis finds that individual investors are more risk aversion when they pay more attention to the market.The reason is that they hold more shares when they are more attentive because of attention-driven trades.But this effect could be weakened by investors' optimistic sentiment during high attention period.