ARE LOSS AVERSION AFFECT THE INVESTMENT DECISION OF THE STOCK EXCHANGE OF THAILAND S EMPLOYEES?

ARE LOSS AVERSION AFFECT THE INVESTMENT DECISION OF THE STOCK EXCHANGE OF THAILAND S EMPLOYEES? by San Phuachan Doctor of Business Administration Program, School of Business, University of the Thai Chamber of Commerce, Bangkok, 10400, Thailand Senior Analyst Investigation, Market Surveillance Department, The Stock Exchange of Thailand, Bangkok, Thailand E-mail: san@set.or.th ABSTRACT The behavioral finance concept has emerged as an alternative theory that incorporates the investor psychology and economics principles for explaining the behavior of a financial market. Since investors may behave differently in different markets, it is worthwhile to separately investigate the behavioral finance concept for each economy. This work conducting a survey to study one of the factors of behavioral finance concept that is the loss aversion. Questionnaires and non-parametric tests are used to examine the proposed hypotheses in this work. The questionnaire is designed to capture the loss aversion biases among the Stock Exchange of Thailand s employee (SET s employee). The results of this study show that SET s employees mostly obtain investment information for their decisions on stock trades from the media. Most SET s employees also believe that dynamic investment environments and their own mistaken investment decisions are the major causes when they miss their profit targets. The fundamental analysis is found to be a main analysis tool that most SET s employees rely on when considering to buy or sell common stocks in the stock markets. It is discovered that personal factors including gender, education level, and investment experience are related to the loss aversion behavior. In addition, SET employees have a tendency to exhibit loss aversion when facing investment losses, but would likely show risk aversion when experiencing investment profits. The finding supports the behavioral finance concept rather than the traditional finance theories. KEY WORDS Loss Aversion, Behavioral Finance, Stock Exchange INTRODUCTION Research in behavioral finance is relatively new. Within behavioral finance it is assumed that information structure and the characteristics of market participants systematically influence individuals investment decisions as well as market outcomes. According to behavioral finance, investor market behavior derives from psychological principles of decision making to explain why people buy or sell stocks. Behavioral finance focuses upon how investors interpret and act on information to make investment decisions. In addition, behavioral finance places an emphasis upon investor behavior leading to various market anomalies. Behavioral finance is defined by Shefrin(1999) as a rapidly growing area that deals with the influence of psychology on the behavior of financial practitioners. Behavioral finance research is developing rapidly and now beginning to answer such questions as (see Taffler (2002)): Why, when all the evidence shows investors cannot beat the market on any systematic basis, they still resolutely do; how can we explain the stock market bubbles ; why is the volume of trading in financial markets so excessive and why is the stock market so volatile; why do investment analysts have so much difficulty in identifying under- and over-valued stocks; why do stock prices appear to under-react to bad news; why do acquisitions on average turn out to be unsuccessful; why should new issues exhibit short-run stock market out-performance and then long-run under-performance. Loss aversion refers to the tendency for people to strongly prefer avoiding losses than acquiring gains. Some studies suggest that losses are as much as twice as psychologically powerful as gains. Loss aversion was first convincingly demonstrated by Amos Tversky and Daniel Kahneman. Loss aversion is one of the most important concepts in behavioral finance. It is consistent with a wide range of empirical findings such as the endowment effect (Thaler (1980); Kahneman et al. (1990)), status quo bias (Samuelson and Zeckhauser (1988)). Loss aversion is traditionally defined in the context of lotteries over monetary payoffs (Kahneman and Tversky (1979)). However, people

often incur losses that may not be measurable in monetary terms (e.g. loss of a close friend or a relative, loss of faith, reputation or prestige, loss of a sports title, loss of animal species etc). For example, imagine that your country is preparing for an outbreak of a disease which is expected to kill 600 people. Given the choice between two vaccination schedules, Program A which will save 200 and Program B which will save all 600 with probability 1/3, most will choose Program A. However, if the question is framed as: imagine that your country is preparing for an outbreak of a disease which is expected to kill 600 people. Given the choice between two vaccination schedules, Program C which will allow 400 people to die and Program D which will let no one die with probability 1/3 and all 600 will die with probability 2/3, most people will choose option D. This is an example of loss aversion: the two situations are identical in quantitative terms, but in the second one the decision maker is losing instead of saving lives, thus setting 0 lives lost as the status quo from which losses are measured, making the sure loss of 400 people more loathsome than the probable loss of 600. We can conclude that the most common behavior that most investors do when making investment decision are (1) Investors often do not participate in all asset and security categories, (2) Individual investors exhibit loss-averse behavior, (3) Investors use past performance as an indicator of future performance in stock purchase decisions, (4) Investors trade too aggressively, (5) Investors behave on status quo, (6) Investors do not always form efficient portfolios, (7) Investors behave parallel to each other, and (8) Investors are influenced by historical high or low trading stocks. Population and Sample Group DATA Population in this study is formulated from the Stock Exchange of Thailand (SET) employee. There are 804 employees in the SET (as of year 2009). This study uses a sampling process based on a convenient sampling technique mentioned in Zikmund (2003) and set the size of targeted samples according to the suggestion of Wanitbancha (2003). The required size This study set a targeted sample size at 95% of confidence level and 5% statistical variance. Hence, the targeted sample size is required to be 260 employees. Data Collection This study uses primary data in a combination from close-end questionnaire as suggested in Zikmund (2003). For the close-ended questionnaire part, dichotomous questions and checklist questions are employed to collect information related to personal information, decision making under different situation and attitude, which show the psychological factors information. The process of data collection could be summarized as follows: 1. Distribute 350 questionnaires to each department in the Stock Exchange of Thailand 2. In total, There are 164 copies of the completed questionnaire returned are use for research analysis in this study. 3. Enter completed questionnaires into a common statistical package program named Statistical Package for Social Science: SPSS Version 17 Data Analysis Raw data drawn from responded questionnaires was processed by the selected statistical package program, SPSS for Windows, to calculate statistical scores as well as to test a series of hypotheses. Steps of data analysis can be described as follows: 1. Calculate descriptive statistics such as frequencies, mode, percentage and mean for describing an overall characteristic of Thai retail investors. 2. Verify a series of hypotheses by using the non-parametric test like Chi-square test for investigating the relationship between personal factors and psychological factors influencing investment decision-making of SET s employees.

Scope of the Study This study covers only the employee of the Stock Exchange of Thailand. The period of data collection is in March and April 2010. THEORETICAL FRAMEWORK We have theoretical framework for this study as follow: Dependent Variables Loss Aversion Behavior Independent Variables Personal Factors: - Gender - Age - Education Level - Experience RESULTS AND ANALYSIS Descriptive Statistics Results The descriptive statistics obtained from the proposed survey for studying behaviors of Stock Exchange of Thailand s employees (SET s Employees) are presented in this section. The majority of the SET s employees (sample group) in this survey are female (59.76%). In this study, mostly age range is in the range of 31-40 years old (35.98%), 41 50 years old (30.49%), respectively. It can be show that most of the sampling groups are in the middle age of work. Most of SET s employees sample graduated bachelor degree (60.98%). experience that including invest in equity instrument debt instrument derivatives instrument and mutual funds. From the survey we find that most of the SET s employees are new investor because they have investment experience below one year (31.71%), between 1 3 years (30.49%), respectively. We imply that SET s employees are likely to have little investing experience since most have just entered to an investment past 1-3 years. TABLE 1 THE SOURCES OF INVESTMENT INFORMATION THAT USED BY THE SAMPLE Sources of Information Listed Company Stock Exchange News Analyst and Marketing Medias Friends Your Expected Other Sources 11.09% 12.90% 21.98% 27.62% 8.06% 16.94% 1.41% Table 1 indicates that the most popular sources of investment information that greatly influence SET s employees investing decisions are from medias (27.62%) such as television, radio, newspaper, and the Internet. Other popular information sources of information include the advice from marketing officers of brokerage firms and stock analysts (21.98%). Surprisingly, there are 16.94% of SET s employee used own expected (decision) in their investment decision. Next, in table 2, show the sources of investment data that used in an employee investment.

TABLE 2 THE SOURCES OF INVESTMENT DATA THAT USED BY THE SAMPLE Sources of Data Percent (%) Most Active Stock 10.40 Most Gainer 5.71 Most Loser 0.88 Most Volume 8.35 Bank Interest Rate 5.71 News 9.81 Foreign Buy Sell Volume 7.91 Foreign Institution Buy Sell Volume 5.56 Expected Net Income 13.46 Dividend Payout Ratio 15.81 P/E Ratio 15.08 Other 1.32 From table 2 the SET s employees mostly used accounting data in their investment, that is, dividend payout ratio (15.81%), P/E ratio (15.08%), and expected listing company net income (13.46%) Table 3 shows that SET s employee believe the most common causes of their underachieving investments are the environment (27.29%) changes in political, social, and economic situations both domestically and internationally. In addition, the mistaken decisions made by SET s employees themselves (24.47%) and the market trends (18.82%) are among other less acknowledged causes of employees missing their investment targets. TABLE 3 THE MOST COMMON CAUSES OF UNDERACHIEVING INVESTMENTS Causes of underachieving investment Percent (%) Marketing Officer and Analyst 14.12 Friends / Other Investors 4.47 Market Direction 18.82 Own Mistake 24.47 Medias 5.88 Listed company 3.06 Environment (economics, political) 27.29 Other 1.88 Next, table 4 indicates that the majority of individual employees (43.95%) analyze stocks based on the fundamental factors. Basically, the fundamental analysis is the most widely-used technique because it is easy to understand, convenient to adopt, and it is easy to obtain information needed for this analysis. Other popular analysis methods used by employees include the technical analysis (28.61%) and the market sentiment analysis (18.58%). TABLE 4 METHODS OF SECURITIES ANALYSIS Methods Percent (%) Fundamental Analysis 43.95 Technical Analysis 28.61 Market Sentiment Analysis 18.58 Rumor Analysis 7.08 Other 1.77 Regarding the behavior of investment or investment style we find that SET s employees would be preferred to sell winning stocks to gain profits, but willing to hold onto the losing stocks, as show in table 5.

TABLE 5 INVESTMENT STYLE style Percent (%) Profit Sell, Loss Not Sell 51.30 Profit Not Sell, Loss Sell 3.10 Profit or Loss Not Sell 2.50 Profit or Loss Sell 13.10 Not Sure 30.00 When the employees face with the uncertainty situation such as assume that the securities price of employees are decrease (today, employees are loss status). There are equally chances to make profit (price increase) or loss (price decrease) in the next month. We find that most of employees are holding their securities (57.30%) more than sell their securities in today (cut loss), as show in table 6 TABLE 6 UNCERTAINTY SITUATION Behavior of the employees Percent (%) Cut Loss 12.10 Hold 57.30 Buy 12.70 Not Sure 17.80 Not Sure 30.00 INFERENTIAL STATISTICS RESULTS In this study, hypothesis tests are derived to examine the relationship among SET s employees personal factors associated with the psychological factors associated with the loss aversion behavior. For each test, the personal factors being examined include the gender, age, education, and investing experience. The proposed hypotheses are described below. 1. SET s employees investment style. H o : A given personal factor has no relationship with the loss aversion behavior of investment style. H a : A given personal factor affects the loss aversion behavior of style. 2. SET s employees reactions when holding losing stocks and facing with uncertainty about the losing stocks that may either break even or decline further. H o : A given personal factor has no relationship with the loss aversion behavior of SET s employees. H a : A given personal factor affects the loss aversion behavior of SET s employees. 3. SET s employees reactions when the securities price decrease. H o : A given personal factor has no relationship with the loss aversion behavior of SET s employees when the securities price decrease. H a : A given personal factor affects the loss aversion behavior of SET s employees when the securities price decrease. Tables 7-9 display the calculated statistics for the tests on relationship between SET employees personal factors and the loss aversion behaviors. Based on the proposed hypothesis testing, it is found that Gender and experience of the SET s employees correlate with their investment style (question number 6), which can referred to the loss aversion behavior in their investment style. Furthermore, simulations in this study (question number 7) also suggest that most SET s employees tend to choose to hold onto the losing stocks when facing with the situations where the stocks they are holding have lower values than when they were bought. This happens even when the chance that the price would go up is equal to the chance that the price would fall down for the same amount. That is, most SET s employees prefer hold onto the stocks and refuse to cut losses. Alternatively, this finding can be explained as the loss aversion behavior according to the behavioral finance concept. Basically, investors (in this case is SET s employees) are likely to exaggerate pain when

losing money, although the amount of losses may equal profits. When their stocks are losing values, investors are reluctant to realize the fact that they are suffering losses by not getting rid of the losing stocks at lower prices. Instead, they would opt to hold onto the losing stocks. In summary, it can be concluded that SET s employees use their past experience to predict future stock prices, hoping that what happened in the past will repeat again. This finding and the fact that investors are not willing to accept the truth about their losses if they actually sell the losing stocks at low prices, motivate most of them to hold onto the losing stocks in the hope that the values of the losing stocks will bounce back to the same level or exceed the costs when purchased. Surprisingly in the hypothesis 3, from the situation of the stock (named PTTT) price decrease about 20% (question 9(2)). We find that only the education level of the SET s employees correlated with the concept of loss aversion behavior. TABLE 7 CALCULATED STATISTICS FOR HYPOTHESIS TEST 1 Personal Pearson Hypothesis Test Result 1 Factors Chi-Square Asymp. Sig. 2 Contingency Approx. Sig. 3 Coefficient Gender * 16.673 0.002 0.307 0.002 Age 19.336 0.252 0.328 0.252 Education Level 4.858 0.773 0.172 0.773 Experience * 23.051 0.027 0.355 0.027 * Rejected Ho 1 Confident level 95 % (Alpha = 0.05) 2 Asymptotic Significant 3 Approximate Significant TABLE 8 CALCULATED STATISTICS FOR HYPOTHESIS TEST 2 Personal Pearson Hypothesis Test Result 1 Factors Chi-Square Asymp. Sig. 2 Contingency Approx. Sig. 3 Coefficient Gender * 13.766 0.003 0.284 0.003 Age 13.691 0.321 0.283 0.321 Education Level* 18.639 0.005 0.326 0.005 Experience* 19.191 0.024 0.330 0.024 * Rejected Ho 1 Confident level 95 % (Alpha = 0.05) 2 Asymptotic Significant 3 Approximate Significant TABLE 9 CALCULATED STATISTICS FOR HYPOTHESIS TEST 3 Personal Pearson Hypothesis Test Result 1 Factors Chi-Square Asymp. Sig. 2 Contingency Approx. Sig. 3 Coefficient Gender 3.990 0.263 0.159 0.263 Age 15.996 0.191 0.308 0.191 Education Level* 30.653 0.000 0.409 0.000 Experience 8.234 0.511 0.226 0.511 * Rejected Ho 1 Confident level 95 % (Alpha = 0.05) 2 Asymptotic Significant 3 Approximate Significant

CONCLUSION This work studies the behavior of the Stock Exchange of Thailand s employee based on the behavioral finance framework, which utilizes the psychology concepts together with financial economics concepts in explaining the behavior of investors in making investment decisions. The non-parametric and Chi-square statistics tests are used to examine the relationship among personal information, the psychological factors, and the investment decisions of investors. It is found that the SET s employees used the investment information obtained from the media including television and radio programs, newspaper, and the internet and tend to believe the advice given by experts including stock analysts and brokerage marketing officers. Most of them tend to used the accounting data (dividend payout ratio, P/E ratio, expected listed company net income) to make an investment decision. When their performance in the market is not as good as expected, SET s employees tend to believe that the most common causes of their investment failure are influenced by an environment such as the changes in political, social, and economic situations both domestically and internationally. Interestingly, they are blaming themselves for their underperformance. Regarding the behavior of investors, it is found that most SET s employees tend to sell the winning stocks, but hold onto the losing stocks. That is, they can be a risk averse in the situation where their stocks are making profits, yet they may become a loss averse when facing with losing environments. This controversial argument is not consistent with the rational investor concept of the traditional finance theories, which suggested that rational investors would have favored in risk aversion behaviors by cutting losses rather than holding onto the losing stocks. In addition, there are evident to show that the personal factors of SET s employee correlated with the loss aversion behavior, that is, most SET s employees are likely irrational investors since their investment decisions are often influenced by emotions and other distractions rather than rationality. This finding does not agree with the traditional finance principles but can be explained by the behavioral finance concept. The results from this study have useful merits for both educational purposes and practical applications. The results can also be used together with the traditional finance concepts for better explaining any financial phenomena. The SET s employees may learn from the results of this study, and accordingly adjust their investment strategies to improve their future performance. REFERENCES Kahneman, D., J. L. Knetsch and R. H. Thaler (1990) Experimental Tests of the Endowment Effect and the Coase Theorem Journal of Political Economy 98(6), 1325-1348 Kahneman, D. and Tversky, A. (1991) Loss Aversion in Riskless Choice: A reference Dependent model. Quarterly Journal of Economics, 1039-1061. Kahneman, D. and Tversky, A. (1979). Prospect Theory: An Analysis of Decision Making under Risk. Econometrica, 47(2):263 291. Samuelson, W.F. and R.J. Zeckhauser. (1988) Status Quo Bias in Decision Making Journal of Risk and Uncertainty 1, 7-59 Shefrin, H. (1999) Beyond Greed and Fear: Understanding Behavioral Finance and the Psychology of Investing. Boston, MA: Harvard Business School Press. Taffler,R.J. (2002) What can we learn from behavioral finance?, Credit Control, Vol.23. Thaler, R. (1980) Toward a Positive Theory of Consumer Choice. Journal of Economic Behavior and Organization 1(1), 39-60 Wanitbancha, Kanya. (2003). Usage of SPSS for Windows in Data Analysis. Bangkok:Thammasarn Publishing Zikmund, W. G. (2003). Business Research Methods. 7th ed. South-western, Thomson Learning.