Equity Market Reactions to Financial Information: A Survey of Stock Price Impact Exposure Through the Presentation of Annual Reports

Andreas H. Sørensen & Frank B. Nielsen

Student thesis: Master thesis


The main object of this master thesis is to identify and analyze which quantitative components in an annual report, which gives most information value to investors. In order to try and illuminate this area the theory of return regression is used.
In order to use return regression for investigating which components in the annual report that gives the best information to investors, all data must be gathered as quantitative figures. In order to find the best quantitative components in the annual report, accounting theory with focus on fortune-oriented accounting theory and performance oriented accounting theory, was reviewed. In addition, theory about valuations models are used, based on the assumption that input to these models must include figures that could be used as quantitative data for the regression testing.
Key figures from the income statement, balance sheet and cash flow statement data were selected based on the theory of valuation
Before the key figures could be tested, all other assumptions for using return regression must be in place. Regression uses a dependent and an independent variable. For the dependent variable, return regressions is characterized by having the change in stock prices as the dependent variable. To be able to explain any variation between the change in stock prices (the dependent variable) and the independent variables, such as the change in revenue, net income etc., the Danish stock market most be analyzed to determent if the market is efficient according to Fama’s hypothesis from 1970.
To prove that the Danish stock market react as an efficient market, when it receives new information, hypothesis testing was prepared based on data from the Danish stock market.
The result of the hypothesis testing was that the null hypothesis was rejected and therefore the evidence suggests that the stock market reacts more on the day that the annual report was made public, compared to other days closed to the announcement date. Given these findings, it is therefore assumed that collated evidence was sufficient for the use of stock prices as the depended variable. Other testing also showed that most of the financial information was recognized in the stock prices during the financial year, as only 11-12 % of the total change in stock prices for the period, was identified on the date of the announcement of the annual report.
After all assumption for return regression was analyzed and tested, it was possible to perform regression testing. The result of the testing was that only a few components had evidence of correlation with the change stock prices. The few components, which did have a correlation (change in return of equity and change in revenue), the degree of exploitation was around that 17- 18 %.
After some analyzing on the preliminary output and exclusion of outliers in the dataset, the degree of explanation could be significantly improved, when the sample was decomposed into its individual segments such as Large, Mid and Small Cap. While Mid and Small Cap companies only have a low degree of correlation with the change in stock prices, the Large Cap companies had a degree of explanation around 40 % albeit only with the use of the change in return of equity and change in revenue.
The performed testing did not give any evidence that accounting components from the cash-flow statement or the fortune-oriented accounting theory for the measurement of assets gave significant information to investors. On the other hand, some evidence suggested performance oriented accounting theory did give some valuable information for especially the measurement of revenue and net income for companies located in the large cap index.
The last trial in the master thesis was to perform a multiple regression test, with the purpose of creating a straight-line estimation with many independent variables but only with observations from the Large Cap index. By doing so a regressions model was estimated with a degree of explanation of 67 % whereas return of equity, revenue and net income were proven the only correlated components, with the change in stock prices.

EducationsMSc in Auditing, (Graduate Programme) Final Thesis
Publication date2016
Number of pages164