A study to measure quality of earnings in financial statements of companies listed at the Nairobi Stock Exchange

Abstract

Concerns about the quality of earnings intensify as economies turn down, companies founder, and investors lose. With the bursting of the recent stock market bubble, business powerhouses like Enron-Andersen in the US and Uchumi in Kenya collapsing, the quality of reported earnings is again under scrutiny. One role of responsible accounting is to anchor investors on fundamentals. It has been extensively argued in (Helsingfors, 2005) that it would be useful if the value of the firm could be read directly from the balance sheet. This would be the case if assets and liabilities would reflect proper estimates for expected net present values and the firm's all future cash flows However, the estimation of fair values of assets without observable market prices would be dependent on managers' competence and discretion and would thus be unreliable (Helsingfors, 2005). It is against this background that current study sought to establish the quality of earnings in financial statements of the 48 companies listed at the Nairobi Stock Exchange for the period covering 2000 to 2004. The study was a cross sectional census survey of all the 48 firms quoted at the NSE from 2000 to 2004. The monthly stock returns were computed using the closing prices of stocks. Earnings per share were adjusted for annual bonuses as appropriate. The monthly return on stocks for each listed company was averaged over a 12-month period to obtain a representative stock return for each year. In analyzing the data, a time series auto correlation of earnings per share and return on stocks was conducted to check for consistency and data reliability. The autocorrelarions for stock returns were consistently positive for all the 16 lags while the autocorrelations for earnings per share (EPS) were both positive and negative. The consistency of the autocorrelation coefficients for stock returns (SR) depicts data reliability while the inconsistency of the autocorrelations for earnings per share depicts low earnings quality. The paired sample t-statistics revealed a substantial difference between earnings per share and stock returns, a pointer to the fact that the earnings per share might have been over specified in most of the firms to blindfold investors It can therefore be concluded that quality of earnings In financial statements are compromised and therefore cannot be relied on by investors, lenders, government authorities, customers, suppliers, and ernployees of these listed organizations in their decisions for investment, taxes and trading. . This is a pointer to the low earnings quality that is characteristic of the reporting in the financial statements of companies listed at the 'Nairobi Stock Exchange. The results of the study is an awakening call on the Capital Markets Authority (CMA) to come into play and ensure reliability, quality and value-relevance of reported earnings in financial statements of companies listed at the NSE. In the meantime, investors are urged to consult investment analysts in identifying firms with different degrees of valuerelevant earnings rather than on reported - earnings. This research is in tandem and concurrence with earlier researches and supports the findings and conclusions.