The Impact of Effective Investor Relations on Market Value Vineet Agarwal (Cranfield School of Management) Angel Liao (The Management School, University of Edinburgh) Elly A. Nash (Independent) Richard J. Taffler* (The Management School, University of Edinburgh) *Corresponding author: Richard Taffler Martin Currie Professor of Finance and Investment The Management School University of Edinburgh William Robertson Building 50 George Square Edinburgh EH8 9JY Tel: + 44 (0) 131 651 1375 Fax: + 44 (0) 131 651 3068 E-mail: [email protected] The Impact of Effective Investor Relations on Market Value Abstract In this first study to test formally the market value of investor relations (IR) activity, we employ the annual Investor Relations Magazine Investor Relations Awards from 2000 to 2002 to proxy for the quality of firm investor relations. We find firms perceived to have the most effective IR strategies earn superior abnormal returns, both before and after the nominations. This shows that while the nominations themselves may be influenced by past performance to some extent, they are nonetheless also associated with subsequent positive abnormal returns. We also find that, not surprisingly, higher analyst following is associated with more nominations suggesting analysts tend to favor the stocks they follow. Consistent with effective IR leading to lower information risk, liquidity of nominated firms increases in the year subsequent to the nominations. Overall, our evidence is consistent with effective IR successfully reducing risks associated with high information asymmetry, as predicted by information risk and agency theories. The Impact of Effective Investor Relations on Market Value 1. Introduction Well functioning capital markets require a free flow of relevant information to enable efficient asset pricing. The investor relations (IR) industry has developed substantially over the past few decades, primarily driven by a growing demand for firms to provide a higher degree of information transparency and accountability to multiple stakeholders. The National Investor Relations Institute (NIRI) defines IR as "A corporate marketing activity, combining the disciplines of communications and finance, providing current and potential investors with an accurate portrayal of a firm's performance and prospects, therefore having a positive effect on total value relative to the overall market and the firm's cost of capital." However, despite the substantial increase in importance firms now place on IR activities, until recently, little attention was paid in the literature as to whether an effective IR strategy adds to shareholder value. This is the first study in the literature, as far as we are aware, to seek to test formally whether effective investor relations (IR) increases firm market value. Specifically, we address this issue by working with firms perceived to have the most effective IR strategies as they are nominated by security analysts and fund managers for ‘Best Overall IR’ in the annual Investor Relations Magazine IR Awards for 2000 to 2002. We explore the relation between firm rankings in the IR awards, and their stock returns, liquidity, and analyst coverage surrounding the date of their nomination for these key IR industry awards. We show that in the year prior to being nominated for ‘Best Overall IR’ in the IR Magazine awards firms earn significant abnormal stock returns, and firms with more award nominations have higher analyst coverage. These findings suggest that analysts and fund managers may be influenced to nominate firms with high prior stock returns, and firms with which they are more familiar as evidenced by higher analyst following. This supports the behavioral finance literature, which predicts these firm characteristics will appeal to the psychological preferences and biases of the respondents to the IR survey, and can influence which firms they nominate (representativeness bias). The results are consistent with similar empirical findings of prior characteristics of firms that are rated in Association of Investment Management and Research (AIMR) surveys (e.g. Lang and Lundholm, 1993). Over the year following the IR awards, we find that nominated firms earn superior abnormal returns suggesting the market does not fully impound the implications of better IR. Our results are consistent with the literature that finds superior abnormal returns for highly rated companies in Fortune’s ‘America’s Most Admired Companies’ survey (Filbeck et al., 1997; Filbeck and Preece, 2003; Antunovich et al., 2000; Anderson and Smith, 2006). Consistent with the predictions of information risk and agency theories, which together propose that enhanced corporate communications will reduce information risk or agency problems caused by high information asymmetry, we find that nominated firms experience an increase in stock liquidity. A seminal paper by Brennan and Tamaronski (2000) demonstrates a chain of relationships that together establish a “direct link between a firm’s investor relations policy and its stock price”. The first link in this chain is an increase in analyst following that can result from a good corporate IR strategy that operates primarily by reducing analysts’ research costs (Bhushan, 1989; Lang and Lundholm, 1996; Francis et al., 1997; Holland, 1998). Secondly, there is empirical support that higher analyst coverage has a significant positive impact on liquidity, both directly due to reduced trading costs, and also indirectly through a consequent effect on equity trading volumes 2 (Brennan and Subrahmanyan, 1996). Finally, Amihud et al. (1997) find that increased stock liquidity is a direct determinant of a firm’s cost of capital and therefore directly impacts stock prices, thus completing the final link in a putative chain of causation from effective IR to shareholder value. However, there is limited empirical evidence of a direct link between a firm’s disclosure policy and market pricing. Botosan (1997) constructs a subjective disclosure quality index based on annual report disclosures, which are treated implicitly as a proxy for the effectiveness of the firm’s overall communication policy. She finds a direct negative relationship between firms’ disclosure index scores and their cost of equity, but only for firms with low analyst coverage. However, these findings may not be generalizable since the study is based on a small sample of firms in a single industry sector in 1991. Crucially though, the role of IR is much more than just the mechanics of conveying formal financial information, hence Botosan’s findings make only a tangential direct contribution to the IR literature (see Marcus and Wallace, 1997). Healy, Hutton and Palepu (1999) test the stock performance of the 97 firms with 3- year consecutive increases in AIMR disclosure ratings in the 1990s and find that on average these firms’ stocks earned excess risk-adjusted returns of approximately 5% over this period. Their sample consists only of firms with a sustained improvement in overall AIMR disclosure rating, and is thus not representative of a typical listed firm. Botosan and Plumlee (2002) use the AIMR survey of corporate communications ratings from 1986-1996 based on a survey of analysts and fund managers. They find no significant relation between firms’ IR ratings, and their cost of equity capital. However, both Healy et al. (1999) and Botosan and Plumlee (2002) use composite AIMR ratings which do not provide a ‘pure’ measure of the value of a firm’s IR activities, since a 3 firm’s IR performance receives only a maximum of 30% weighting in these ratings. Their results can thus only at best be a reflection of a relation with a firm’s market communications more generally defined. Finally, Bushee and Miller (2005) test 184 small and mid-cap firms that initiate IR programs between 1999 and 2004 by hiring professional IR agencies. They find that these companies significantly increase their level of disclosure and press coverage, stock trading activity, institutional ownership, analyst following, and market valuations after hiring a new IR agency. They suggest that IR activities play a significant role in helping small and mid-cap companies to overcome their low visibility because they do not generally trade on a major exchange, to attract a wider following by investors and information intermediaries and to improve their market valuation. Our study differs to Bushee and Miller (2005) because our sample firms are likely to have more established IR programs because they are nominated for IR industry awards. The rest of the paper is organized as follows: section 2 presents our hypotheses, data and method, section 3 presents our results, and section 4 summarizes our findings. 2. Hypotheses, Data and Method 2.1. Hypotheses The IR function of a firm is a dedicated channel of information from senior management to external stakeholders, hence IR performance, in theory, should have significant impact on information asymmetry between insiders and outsiders. Effective IR should reduce the risk premium associated with information asymmetry and thereby lead to lower cost of equity. It should also lower the cost of analysts’ information gathering for, and raise profile, with investors thereby creating higher demand for 4 analyst coverage of firms with better IR. Higher analyst coverage combined with lower information asymmetry should increase trading volumes and liquidity leading to lower liquidity premium and therefore higher stock returns. Information risk theory and agency theory thus together provide a framework in which an effective IR policy can influence both stock prices and stock liquidity by reducing risks associated with high information asymmetry. Effective IR thus should reduce the perceived risks that investors associate with high information asymmetry and lead to higher stock valuations. McGuire et al. (1990) find prior financial performance drives ratings in Fortune’s annual survey of ‘America’s Most Admired Companies’. Similarly, Lang and Lundholm (1993) find that firms with superior past performance and higher analyst following are more likely to receive a higher rating in the AIMR surveys. We therefore establish our first two null hypotheses: H1 : There is no significant relation between effective IR and prior abnormal equity 0 returns. H2 : There is no significant relation between effective IR and prior high levels of 0 analyst coverage. While effective IR can reduce information asymmetry, if the market is efficient with respect to impounding the implications of effective IR, firms that are nominated for these awards should not earn significant abnormal returns over the year following the nomination. We thus establish our third null hypothesis: H3 : There is no significant relation between effective IR and future excess equity 0 returns. 5 Effective IR is also associated with increased analyst coverage, primarily because it reduces the time and costs for analysts in searching for, and analyzing information about a firm, and because it reduces information asymmetry between the firm and investors and analysts, leading to increased demand for analysts’ services. We therefore establish our fourth null hypothesis: H4 : There is no significant relation between effective IR and future increased levels 0 of analyst coverage. Information risk and agency theories together predict that effective IR will reduce perceived risks that investors associate with high information asymmetry with firms in which they have invested, and lower information asymmetry will lead to increased liquidity. We thus establish our fifth null hypothesis: H5 : There is no significant relation between effective IR and future increased stock 0 liquidity. 2.2. Data For over a decade, IR Magazine has annually commissioned an independent research firm to obtain nominations from investors and analysts for firms that have performed the ‘best’ in distinct categories of IR over the previous 12-months. Nominations are collected from a large sample of fund managers and sell and buy-side analysts listed in the Thomson Financial I/B/E/S database and Barron’s and WILink databases, covering a wide range of industry sectors and investment specializations, although all respondents are encouraged to nominate firms outside their specialities. The nomination-collection 6 process takes place during March and is finalized March 31 each year, but nominations should only relate to IR performance over the past 12 months. Table 1 panel A presents the number of firms nominated in each category for each of the three years in the sample. Table 1 here Stock returns and prices, trading volumes, and industry codes are extracted from the Centre for Research in Share Prices (CRSP) database. Analyst coverage is obtained from the Thomson Financial I/B/E/S database. 2.3. Method Each year from 2000 to 2002, firms nominated in the ‘large firms’ category are sorted by the number of nominations received and divided into three portfolios formed at the rank percentage breakpoints of award nominations, portfolio 1 with firms in the bottom 33%, portfolio 2 with firms in the middle 34%, and portfolio 3 with firms in the top 33% by nominations received. Similarly, firms that are nominated in the ‘small firms’ category are also sorted into three portfolios. Finally, we form three pooled portfolios, portfolio 1 is formed by pooling together portfolio 1 firms from large and small categories, and portfolios 2 and 3 are formed by pooling together portfolio 2 and 3 firms respectively from large and small categories. Panel B of table 1 presents the number of firms in each portfolio, pooled across the three award years.1 To test whether the firms nominated for IR awards earn superior risk-adjusted stock returns, we employ the following Carhart (1997) four factor model: R – R = a + b RMRF + s SMB + h HML + m MOM + e (1) Pt Ft t t t t t 1 Since the portfolios are formed using percentiles of votes, the numbers of stocks in the portfolios are not equal. 7 where R = the average of the returns of the firms in portfolio P during month t, Pt R = the risk free rate (US long bond rate) at the start of month t, Ft RMRF = excess return on the market factor in month t, t SMB = return on the mimicking portfolio for the size factor in month t, t HML = return on the mimicking portfolio for the book-to-market factor in month t, t and MOM = return on the mimicking portfolio for the momentum factor in month t. t RMRF, SMB, HML and MOM factors are from the Kenneth French web site2. To test the average level of analyst coverage of the firms over one year prior (T-1) and one year subsequent (T+1) to the nomination year (T), we pool our sample firms across award years and run the following regression which controls for firm market value at each year-end: AF = a + b IR + b ln(MV)+ e (2) i IR i MV i i Where AF = number of analysts publishing forecasts in I/B/E/S database for firm i, i IR = IR rating of firm i, and i MV = market value of firm i at March 31 of the award year. i To test whether the stocks’ liquidity increases after the nominations, we use the turnover ratio as a measure of liquidity. Monthly turnover ratio for each stock is defined as (see e.g. Korajczyk and Sadka, 2008): 2 (http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/). 8