Analysts’ stock recommendations’ value: a comparison of the US and European markets

Going further to neutral recommendations, according to received results presented on the Table 8, the picture strictly corresponds to the findings from testing the first hypothesis. As it can be noticed, the level of a preceding recommendation does not matter for investors when the neutral recommendations are given.

As for the negative recommendations (please, see the Table 9), the picture seems to be enough clear, and switching to a negative level from the neutral one, as well as it repetition, leads to strong negative abnormal returns on the event day. The most durable reaction is captured in a situation of downgrading from positive to negative recommendations as it is expected to cause significant negative abnormal returns on an event day and one day after it.

On balance, as for the hypothesis H21, we have to partly reject it because of, according to Tables 7-9, only reiterations of neutral recommendations do not provide with significant abnormal returns. On the contrary, repetition of positive recommendations cause positive abnormal returns on the days and . In addition, subsequent negative abnormal returns on the day are captured for maintenance of negative recommendations.

In a framework of the hypothesis H22, we can only partly accept it because the following findings are quite ambiguous. Firstly, the conclusion is such due to the fact that, although, switching from negative to positive recommendations shows the largest abnormal returns, they are less durable than in cases of changes from neutral and positive recommendations. Secondly, switching from negative and positive to neutral recommendations brings insignificant abnormal returns to investors. However, in a case of negative levels, switching from neutral and positive recommendations causes sharper negative abnormal returns and longer reactions, respectively.

Table 7. Results from testing the second hypothesis for changes to positive recommendations.

Table 8. Results from testing the second hypothesis for changes to neutral recommendations.

Table 9. Results from testing the second hypothesis for changes to negative recommendations.

5.3 Results of determining other factors affecting the markets' reaction

On the Table 10, the results from the regression model are presented. The first reasonable thing to mention is that although the model do not have big explanatory power due to its low R2 parameter, this value is still around those similar models applied in the papers of Murg, et al (2014) and Suliga (2016), and the model is enough significant because of its around-zero F-statistics' value.

Table 10. Results from the regression model

From the table, it can be noticed that the variable does not overcome the 90% level of significance in this model. One of the reasons of this parameter's insignificance might be that these target prices that appear in analysts' reports often relate to different time horizons. At the same time, in these reports, there might be both target prices that are expected to be reached in a couple of days or, for instance, in half a year. Thus, investors differently consider such target prices. In addition, the variable appeared not to bring enough explanation to the cumulative abnormal returns as well, and this corresponds to the findings from testing the first and the second hypotheses (considering preceding recommendations' levels does not strikingly change the picture of reactions to pure recommendations). One of the possible reasons to this issue is that investors are more likely to pay attention to a detached recommendation coupled with the time-adjusted target prices.

.As for the multiplied variables, and are significant factors for the CAR in this model, and the connection between the explained variable and these regressors is inverse. This boils down to a suggestion that investors react stronger to negative and neutral recommendations for larger companies. However, the size of a company does not significantly influence on CAR in a framework of positive recommendations.

Conclusion

This study was aimed to investigate the recommendations' value for investors that are published by analysts and are related to stocks traded on the US and the UK markets, and to compare where the analysts have more predictive power.

In order to reach the stated objective, such methodological tools as an event study analysis and a linear regression model were applied. The analysis was conducted in the context of the NYSE and the LSE that are outstanding stock exchanges for the analysis due to their huge variation of stocks, trading volumes, and their higher liquidity. The research was carried out in the framework of the recommendations' three-staged levels: positive, neutral and negative ones. This study has met the author's expectations about the possible markets' reaction to such recommendations. It was found that the both stock exchanges react directly to the recommendations' levels. Thus, positive recommendations are likely to cause significant positive abnormal returns on the day of a publication and one day after it, whereas negative recommendations lead to negative abnormal returns on the date of the event on the markets. Although there is a some skewness in a total number of positive recommendations, and there was an expectation about sharper response to negative forecasts due to their rarity, the analysis of duration of the reactions boils down to a conclusion that these markets are more susceptible exactly to positive recommendations. Analysts' suggestions `to hold' stocks do not bring significant abnormal returns to investors.

From the point of a comparison of the US and the UK stock markets, the analysis of average cumulative abnormal returns does not show sufficient differences in such reactions to these recommendations. Therefore, it can be concluded that analysts and brokerage houses that operate on these stock exchanges have similar level of awareness in comparison with ordinary investors.

Another concern of this paper was to identify how changes in the recommendation levels affect the stock prices and, thus, abnormal returns. It was found that the markets more sharply react when a positive recommendation comes after a negative one rather than to a separate positive recommendation but the reaction is less durable. Although in this case there is the strongest reaction among other preceding to positive recommendations, the market response lasts for only one day, whereas in two other cases (when positive level changes the neutral or previous positive ones) the stocks remains to show significant abnormal returns the day after the event. In a situation with negative recommendations, the markets expectably react more significantly to downgrades to a negative level from the neutral one. One noticeable thing is that the markets' response to negative recommendations' reiterations is sharper than when they come after positive ones, but, frankly speaking, in that case, the reaction lasts for two days instead of one. The significant markets' response to negative recommendations' reiterations might be explained by a suggestion that investors on such large markets, as well as the analysts, are confident in general market's growth, and are averse to selling the stocks from their portfolios. Thus, some of the investors start to act only when the `sell' recommendation is supported by further analysts' report. If to talk about neutral recommendations, there is no significant reaction even if to consider the preceding levels.

Identification of other factors that might influence the market response implied conduction a regression model analysis with cumulative abnormal returns from an event window as an explained variable, and such factors as a company's size, extent of the change in a recommendation level, and a target price as the regressors. It was found that cumulative abnormal returns do not significantly depend on changes in recommendation levels as well as pure target prices. This boils down to a suggestion for future research that such target prices need to be time-adjusted in order to get more reliable results because investors are likely to consider both a forecasted target price and a period over which this price is expected to be reached. If to have a look at the effect of the size of a company, the results are ambivalent to the expected ones. It was identified that if a negative or a neutral recommendation is given to a larger company, this leads to stronger reaction and more sufficient cumulative negative abnormal returns. One of the possible explanations might be that huge powerful brokerage houses and analysts are more likely to concentrate on the bigger companies rather than smaller ones. Investors are more likely to rely on this information and trust analysts that are more reputable rather than those who are more independent from their names and, thus, are risk-prone. As for positive recommendations, a company's size does not significantly influence the cumulative abnormal returns.

To summarize the study and to match it with the mentioned objective, we are to conclude that analysts' recommendations can bring a significant value to ordinary investors, as their forecasts and opinions whether to buy or to sell a particular stock lead to subsequent significant abnormal returns. However, the investors need to decide by themselves if the expected value from obtaining this privileged information exceeds the price of subscription to such analysts' reports.

Although, due to the aforementioned findings, both American and British markets are feasible to be outperformed by the analysts, we have to state that it is impossible to highlight any of the markets as they behave almost identically around positive, neutral and negative recommendations. Though it was not a direct goal of the research, since stock prices significantly react when recommendations that sometimes are based on privileged information spread between the markets' participants, this might be evaluated as a piece of evidence that the markets are not strongly efficient. However, in order to speak about the issue more confidently and to determine the extent of the markets' efficiency, more attention should be paid to the information, which such recommendations are based on, because some of them might be enough late or be based on a just-published information that the market have not accepted yet.

As for other possible suggestions for improvement of this research, it might be interesting to consider types of analysts that give such recommendations on stocks from the point of their size, reputation and independence. Actually, one of the reasons of not enough durable and sharp markets' reactions to changes in recommendation levels might be the difference between the analysts. There is an expectation that some of the analysts are likely to influence the markets more than the others do. Thus, it could be an interesting point to cover in future research.

Annotated bibliography

1. Banz, R. (1981) The relationship between return and market value of common stocks. Journal of Financial Economics. 9 (1), pp.3-18.

2. Beneish, M. (1991) Stock Prices and the Dissemination of Analysts' Recommendations. Journal of Business. 64 (3), pp.129-153.

3. Davies, P.L., Canes, M. (1978) Stock Prices and the Publication of Second-Hand Information. Journal of Business. 64 (3), pp.393-416.

4. Fama, E. (1991) Efficient Capital Markets: II. Journal of Finance. 46 (5), pp.1575-1618.

5. Fama, E., French, K. (1993) Common risk factors in the returns on stocks and bonds. Journal of Financial Economics. 33, pp.3-56.

6. Hansen, J.D. (1995) London Calling?: A Comparison of London and US Stock Exchange Listing Requirements For Foreign Equity Securities. Duke Journal of Comparative & International Law. 6, pp.197-228.

7. Ishigami, S., Takeda, F. (2018) Market reactions to stock rating and target price changes in analyst reports: Evidence from Japan. Journal of International Financial Markets, Institutions and Money. 52, pp.134-151.

8. Jegadeesh, N., Kim, W. (2006) Value of analyst recommendations: International evidence. Journal of Financial Market. 9 pp.274-309.

9. Keasler, T., McNeil, C. (2010) Mad Money stock recommendations: market reaction and performance. Journal of Economics and Finance. 34, pp.1-22

10. Kothari, S., Warner, J. (2006) Econometrics of Event Studies. Handbook of Corporate Finance: Empirical Corporate Finance. A, pp.3-36.

11. Kraemer, M. (2016) The Economic Value of Stock Recommendations: Evidence From An Online Stock Advice Platform. Bachelor of Science. Erasmus School of Economics. Erasmus University.

12. Malkiel, B. (1992) Efficient Market Hypothesis. New Palgrave Dictionary of Money and Finance. London: Macmillan.

13. Murg, M., et al. (2014) The impact of analyst recommendations on stock prices in Austria (2000-2014): evidence from a small and thinly traded market. Springer. 24 (3) pp.595-616.

14. Murg, M., Zeitlberger, A. (2014) Impacts of analyst recommendations on Austrian and German stocks: Information leaks, overreaction, and the influence of firm size. Journal of Banking and Financial Research. 62, pp.1-19.

15. Okulov, V. (2010) Issledovanie ehffektivnosti rossijskogo rynka akcij: reakciya rynka na publikaciyu prognozov analitikov. Vestnik Sankt-Peterburgskogo universiteta. Seriya menedzhment. 3, pp.3-22

16. Palmon, D., et al. (2009) The value of columnists' stock recommendations: an event study approach. Review of Quantitative Finance & Accounting. 33 (3), pp.209-232.

17. Palmon, O., Sun, H., Tang. A. (1994) The Impact of Publication of Analysts' Recommendations on Returns and Trading Volume. The Financial Review. 29, pp.395-417.

18. Pogozheva, A. (2013) Ispolzovanie sobytijnogo analiza dlya ocenki informacionnoj znachimosto rekomendacij analitikov po rossijskim ehmitentam. Journal of Corporate Finance Research. 26 (2), pp.35-49.

19. Rounaghi, M., Zadeh, F. (2016) Investigation of market efficiency and Financial Stability between S&P500 and London Stock Exchange: Monthly and yearly Forecasting of Time Series Stock Returns using ARMA model. Physica A: Statistical Mechanics and its Applications. 456, pp.10-21.

20. Shen W. (2014) Reactions of the China Stock Market to Analysts' Stock Recommendations. Master of Science. University of Groningen.

21. Suliga, M. (2016) The reaction of investors to analyst recommendations of stocks listed on the WIG20 index. Managerial Economics. 17 (1), pp.123-148.

22. Womack, K. (1996) Do Brokerage Analysts' Recommendations Have Investment Value? Journal of Finance. 51 (1), pp.137-167.

Appendix

Размещено на Allbest.ru