Investor Sentiment Explanation - Essay Example

Introduction Closed-end funds can serve as a perfect illustration for limitations of many theoretical assumptions. Concerns of many researchers fromUS and UK trying to explain closed-end funds, which differ from open-end funds with a fixed limited number of shares on the stock market, deserve serious attention. But before an explanation of closed-end fund phenomena from the behavioral perspective will be given, clear frames of the subject question should be defined. There are some differences and similarities between closed-end funds in the US and the UK. One of the most important differences is much greater participation of institutional investors in the UK market than in the US (Bleaney and Smith 2003, p. 19). Nevertheless the US and the UK markets are more similar than different, that is why we can extend results acquired from one market to another, but with some corrections made. This paper consists of two parts. The first part provides an explanation for the closed-end fund puzzle from the behavioral perspective. The second part deals with further tests, which can be used to verify the investor sentiment explanation. Investor Sentiment Explanation Unlike economic theories trying to explain the closed-end fund puzzle with rational assumptions (e.g. tax liabilities), behavioral theories imply that there is some amount of irrationality on the market. It is called “noise trader sentiment” — the component of expectations about asset returns not warranted by fundamentals. Basic assumptions of investor sentiment explanations are as follows:some of the investors are not fully rational, and their demand for risky assets is affected by their beliefs or sentiments; arbitrage is risky and therefore limited. However, researchers advise not to fall into opposite extreme: “It is absolutely not true that introducing a degree of irrationality of some investors into models of financial markets eliminates all discipline and can explain everything” (Shleifer and Summers 1990, pp. 19-20). Thus a moderate approach should be implemented. Closed-end fund puzzle is often divided into four parts: 1. Trusts are typically launched at a premium of almost 10%. 2. Trusts typically soon move to a discount, although some funds do at times trade at a small premium. 3. Discounts fluctuate widely over the life of a fund. 4. Upon termination, discounts narrow as the share price rises to meet NAV. (Lee et al 1991, pp. 76-77) Let us observe each of these issues from the sentiment perspective. Traditionally, premium is supposed to be strongly connected to returns. “Premia have an economically strong ability to predict returns” (Pontiff 1995, p. 366). However, at a start of a closed-end fund there is little can be said about returns, but most of them are launched at a premium of 5-10%. How can we explain it? Behavioral theory assumes that entrepreneurs gain profit from selling their assets gathered in a closed-end fund to noise traders, which are baselessly optimistic about returns. “It seems necessary to introduce some type of irrational investor to explain why anyone buys the fund shares at the start when the expected return over the next few months is negative” (Lee et al 1991, p. 84). This theory argues that there is no efficiency reason explaining that part of the puzzle. The main issue of the closed-end fund is moving to a discount after a short time period. Investor sentiment theory implies that holding the fund entails larger risks than holding its portfolio: The evidence that funds, on average, sell at a discount does not rely on the average pessimism of noise traders, but stems completely from the risk aversion of rational investors who are willing to buy closed-end fund shares only if they are compensated for the noise trader risk, which means buying the fund at a discount. (Dimson and Minio-Kozerski 1999, pp. 19-20) However, some researchers have doubts about the fact that investor sentiment is able to explain discounts. For example Busse et al suggests that the reason lies in the managerial performance rather than in sentiments: “The discounts observed are too small to compensate for the negative [management performance indexes] and the greater risk… Furthermore, including sentiment risk causes the [management performance index] to move in the wrong direction. Examining equilibrium returns does not support a sentiment story” (1998, pp. 498-499). That is why further researches implemented a more balanced approach including some other factors, influencing pricing (i.e. arbitrage): “The interaction between noise and arbitrage, the former moving the price and the latter constraining its movement to a particular channel, leads to the existence of a discount” (Gemill and Thomas 2000, p. 19). Such approach has proven to be better at explaining movement towards discounts in the closed-end funds puzzle. The third “piece” of the puzzle is about frequent fluctuations that occur during lifetime of the closed-end fund. A survey conducted by the UK researchers Levis and Thomas shows that it also can be explained with the investor sentiment: “Empirical results, using a new set of UK data, provide strong support to the notion that changes in the level of discount reflect retail-investor sentiment, proxied by the volume of (unexpected) retail-investor money flowing into/out of the mutual funds” (1998, p. 15). Moreover it is stated that discount exists exactly due to constant changes in pricing. Finally, it has been supported with lots of empirical data that discounts narrow upon termination or open-ending as the share prices rise to meet NAV. This occurs because when it becomes clear that the fund will be liquidated or open ended than noise trade risk is reduced or even completely eliminated. “The liquidation costs of underlying assets are quite small and the NAV represents a correct estimate of the value of the underlying assets.” (Kadapakkam et al 2003, p. 17). After the risk is eliminated the discount also disappears. Sometimes a small discount may remain. Although the investor sentiment theory provides rather solid explanation of the closed-end fund phenomena, it is still a subject of discussions. For instance, Chen et al (1993) criticize the investor sentiment to be a deciding factor for small firms. If fund discounts are a sentiment index for small firms with little institutional ownership then companies with higher level of institutional ownership ought to behave differently. For low-institutional firms the connection between fund discounts and returns should be strong, for others it should be weak. However the data retrieved shows opposite results: “in fact no difference in the response of discounts on closed-end funds to small firms with less than or with more than 10 percent institutional ownership can reliably be detected” (Chen et al 1993, p. 796). Advocating the behavioral explanation Chopra et al (1993) have presented the important data in support of their concept. “Consistent with the hypothesis of Lee et al., the evidence clearly shows that for utilities, lower institutional ownership firms’ returns are more negatively correlated with discount changes” (Chopra et al. 1993, p. 802). It should be noted that investor sentiment theory provides clear explanation, however more evidence is required before putting this theory out of discussion. Further Tests of Investor Sentiment Explanation Due to the fact that investor sentiment explanation of the closed-end puzzle still requires more evidences some assumptions that are critical for this theory should be tested to verify it. There are three issues, verification of which will bring more confidence into behavioral theory: correlation of discounts across funds; behavior of discounts during initial public offering periods (IPO); and systematic nature of investor sentiment (Lee et al 1991). Let us examine each of these issues more explicitly. Investor sentiment explanation speculates that discounts on closed-end funds reflect widespread changes in investor sentiment. Therefore, levels and changes in discounts should be highly correlated across funds.” Since the same sentiment drives discounts on all funds as well as on other securities, changes in this sentiment should determine changes in discounts” (Lee et al 1991, p. 85). That is why the first suggested test should look into correlation between discounts across funds. Finding evidences of such correlation would serve as a support for the investor sentiment explanation of the closed-end puzzle. The second fact about investor sentiment is optimistic expectations on returns of noise traders, which is stated as the necessary condition for a closed-end fund to emerge. In fact, “the combination of hidden IPO costs (paid via an inflated price) and the after-market price support looks anticompetitive. Do small investors (who are the vast majority of CEF IPO subscribers) understand what they are getting into?” (Cherkes 2003, p. 35) Moreover, Lee et al (1991) suggests that new funds should emerge when old funds are sold at a premium or at a small discount (i.e. before their liquidation). To test this issue we should look into behaviour and influence of discounts on old funds when new funds are started. Finally, the influence of investor sentiment on return generation should also be tested. Theory implies that the investor sentiment should be systematic. Therefore “if the investor sentiment is systematic, we need to compare its importance in the return generating process to a set of factors that most researchers believe that are not priced.” (Doukas and Milonas 2002, p. 9). The investor sentiment explanation does not specify, which securities will be influenced by the same sentiment. Nevertheless the test on influencing the return generating process should be conducted. Conclusion Of course, the investor sentiment explanation is not perfect. It generates a lot of questions, and finding answers will bring us closer to understanding of the closed-end fund phenomenon. Nevertheless, considering the fact that other economic theories are less helpful at explaining the closed-end fund puzzle, behavioral theory should be recognized for providing the most integral explanation from all theories. References Bleaney, M. and Smith, R.T. (2003). Prior Performance and Closed-End Fund Discounts. Discussion paper, University of Nottingham. Busse, J.A. Gruber, M.J. Elton, E.J. (1998). Do Investors Care About Sentiment? The Journal of Business, Vol. 71, No.4, pp.477-500. Chen, N.-F. Kan, R. Miller, M.H. (1993). Are the Discounts on Closed-End Funds a Sentiment Index? The Journal of Finance, Vol. 48, No. 2, pp. 795-800. Cherkes, M. (2003). A Positive Theory of Closed-End Funds as an Investment Vehicle. Princeton University. Chopra, N. Lee, C. Shleifer, A. Thaler, R. (1993). Yes, Discounts on Closed-End Funds Are a Sentiment Index. The Journal of Finance, Vol. 48, No. 2, pp. 801-808. Dimson, E. and Minio-Kozerski, C. (1999). Closed-End Funds: A Survey. Financial Markets, Institutions & Instruments, Vol. 8, No.2. Doukas, J. and Milonas, N. (2002). Investor Sentiment and Closed-End Puzzle: Out-Of-Sample Evidence. Discussion paper, Aristotle University of Thessaloniki. Gemill, G. and Thomas, D. (2000). Noise-Trading, Costly Arbitrage, and Asset Prices: Evidence from Closed End Funds. Faculty of Finance, City University Business School, London. Kadapakkam, P.-R. Misra, L. Yildirim, S. (2003). Open-Ending of Closed-End Funds: An Analysis of Discount Reduction. Preliminary draft paper, University of Texas, San Antonio. Lee, C. Shleifer, A. Thaler, R. (1991). Investor Sentiment and Closed-End Fund Puzzle. The Journal of Finance, Vol. 46, No.1, pp.75-109. Levis, M. and Thomas, D. (1998). Country Funds and Investor Sentiment: UK and US Evidence. Working paper. City University Business School. Pontiff, J. (1995). Closed-end Fund Premia and Returns. Implications for Financial Market Equilibrium. Journal of Financial Economics, Vol. 37, pp. 341-370. Shleifer, A. and Summers, L. (1990). The Noise Trader Approach to Finance. Journal of Economic Perspectives, Vol.4, No.2, pp. 19-33. Read More