Firm Location, Analyst Following, and Bias in Analysts' Earnings Forecasts
Asian Tax Journal Vol. 11 No. 4 (2010), pp. 275-304
Abstract
This study investigates the association between firm location and analysts following as well as direction of the errors included in the earnings forecasts issued by financial analysts. Firm location is measured by an indicator variable representing whether the firm's headquarter is located in urban (i.e., urban firms) or rural (i.e., rural firms) areas. The error in the earnings forecasts indicates the difference between financial analysts' forecasts issued and corresponding actual ex post earnings of the firms. Up to now, there has been voluminous research on the effect of location of various organizations or individuals with respect to diverse issues. For example, Coval and Moskowitz(1999), Loughran and Schultz(2004), Ivkovic and Weisbenner(2005), Kang and Kim(2008), and Teo(2009) report that individual or institutional investors are more likely to invest in firms located nearby. Because most of investors live in urban areas and headquarters of the firms are also more likely to be located in the urban areas, this kind of close distance between investors (and other stakeholders)and firms can influence the decision making of the investors with respect to the firm(Loughran and Schultz 2004). Similarly, Kedia and Rajgopal(2006) reports that the distance between regulators and firms are closely related to the degree of monitoring and thus the probability that a firm commit accounting fraud. Choi et al.(2006) also argues that the distance between auditors and the client firms are also related to both audit quality and audit fees. These are clear evidence supporting the argument that geographical location is an important factor on management decision as well as investor's decision. In Korean context, the firms located in Seoul or Kyunggi (including Incheon) areas are surrounded by majority of investors who live nearby. The population in this area enjoy greater income and spending than the population live in other areas in Korea. These urban investors and stakeholders pay close attention to the behaviors of the firms and are likely to invest in the firms located nearby. In contrast, there are not many people who live in rural areas and they are less likely to invest in stock market than the people who live in urban areas do(Nam and Choi 2009). As a result, one may say that urban firms are more likely to communicate closely with urban population than rural firms do so with rural population. Because urban investors have great interest in urban firms, they may demand more information on the urban firms than rural firms. Lang and Lundholm(1996) and Barth et al.(2001) suggest that financial analysts are more likely to follow a firms that investors have greater interests in and update their forecast for those firms. Thus, if a firm's urban location induce greater interests of investors, it is expected that the number of analysts following the firm increases for the urban firms. We empirically test this prediction as the first hypothesis. Next, we examine whether there exist any association between the firm location and errors in analysts' earnings forecasts. Prior studies in analysts' forecast argue that firms that receive greater attention from outside parties may have incentives to provide more affirmative and optimistic information to please the outsiders(Graham et al. 2005). In addition, from the view point of financial analysts, they compete with other analysts and closely observed by outside stakeholders who want more information on the firm. Thus, they have incentives to seek for more accurate information and thus need to maintain close relationship with firms to receive up－to－date information. Das et al.(9998) suggests that one way for a financial analyst to maintain a close relationship with the firm is to issue optimistic forecasts in the long run. If this explanation is true, it is expected that analysts' earnings forecasts for urban firms are more likely to be biased optimistically compared with those for rural firms. We test this prediction as the second research hypothesis. For the empirical analyses, we collect a total of 2,816 firm－year observations for the seven year period, starting from 2001 (up to 2007). For urban versus rural classification, we use an indicator variable separating Seoul and Kyunggi(including Incheon) versus others (URBAN variable). However, as sensitivity analyses, we also use Seoul versus others (URBAN_1) and Seoul and other major cities versus others (URBAN_2). But the results are qualitatively consistent irrespective of the use of different measures. The empirical results are performed by both univariate and multivariate regression analyses. The results largely support the predictions of the study. Specifically, analysts are more likely to follow urban firms than rural firms and analyst's forecasts tend to be optimistic for urban firms. These findings are interesting and provide a lot of important implications to regulators, academics, as well as investment and business communities.
Keywords
- firm location
- analysts following
- optimistic earnings forecasts
- analysts' earnings forecast error
Related Articles
The Effect of Firm Location on Dividend Policy
19(3) 215-235
The Effect of Book-Tax Difference, Temporary Difference, and Changes in Net Deferred Tax Assets and Liabilities on Analysts' Earnings Forecasts
10(2) 225-260
The Effect of the Information Content of Earnings after Mandatory IFRS Adoption on Financial Analysts’ Earnings Forecast Accuracy
19(3) 119-163
Group－affiliated Analysts’Forecasts and Product Market Competition
18(2) 129-156
A Study on National Institutional Characteristics and Corporate Social Responsibility
27(2) 43-65