Stock market as a reliable reflection of real sector’s performance: can it act as one?

(1)

Exchange

by Windy Dyah Indriantari 1713825

W.D.Indriantari@student.rug.nl

windy_dee@yahoo.com

Supervisors : Drs. H. Ritsema

Drs. J.J. Hotho

Referent : Prof. Dr. L.Karsten

Master Thesis

International Business and Management

Rijksuniversiteit Groningen

(2)

_______________________________________________ Windy Dyah Indriantari

Master Thesis International Business & Management Rijksuniversiteit Groningen 2009 1 TABLE OF COTETS ABSTRACT ……… 2 1. INTRODUCTION ……… 3 2. LITERATURE REVIEW ……… 5 2.1. Signaling Theory ……… 5 2.2. Types of Signals ……… 6 Internal Signals ………. 6 External Signals ……… 10 2.3. Conceptual Framework ……… 11 Hypotheses ……… 13 3. METHODOLOGY ……… 16 3.1. Case Background ……… 16

Indonesia Stock Exchange ………. 16

ew York Stock Exchange ………. 17

3.2. Analyses Steps ……… 18

3.3. Independent Variables ……… 19

Market Capitalizations ……….. 19

ews Feed Frequency ……… 21

Industry Categorizing ……… 21

3.4. Samples and Data Collection ……… 22

4. RESULTS AND ANALYSES 4.1. Indonesia Stock Exchange ………... 23

IDX Correlation Analysis ………. 23

IDX Regression Analysis ……….. 25

4.2. New York Stock Exchange ……… 28

YSE Correlation Analysis ……….. 28

YSE Regression Analysis ……….. 29

5. DISCUSSION ……… 32

6. CONCLUSION ……… 34

6.1. Limitations ……… 34 REFERENCES

(3)

Master Thesis International Business & Management Rijksuniversiteit Groningen 2009

2 ABSTRACT

(4)

3 1. ITRODUCTIO

The stock market nowadays has been considered as one of country's economic indicators. When the stock market’s index is going down, investor will see it as a sign that the business sector is in or will be in trouble. The investors who are planning to inject a direct investment to a certain company often postpone their plan until the economic indicator is back to normal.

Of course, the stock market is not the only indicator being considered by the investors. Nevertheless, it plays a great role in their investment decisions. The main question to be asked is, does stock market really able to represent the real sector's performance. If it does, how good is its ability to reflect the performance?

The main objective of this research is to investigate whether stock prices as quoted by stock exchange can be a relatively accurate reflection of the actual company’s performances or not. Accurate means when the company's performance is going up, the stock price should be having the same trend, and if the company’s performance is going down so is the stock price. This research measures the company’s performance by using earnings per share as an indicator.

Many previous researches have examined the relationship between stock price and company’s performance. However, the researches are focusing on causal relationships and mostly investigating the effect of company's performance announcements on the stock prices movement. Those observations can not give an answer to the research question presented in the previous paragraph.

In order to answer the question, this research observes a non causal relationship between stock prices and company's performance. The idea is only to find out if the stock prices are moving into the same direction as the company's earnings per share. If they are, the stock market has given an accurate reflection of the related company's actual performance.

(5)

4 example quarterly financial reports, give information about the current state of its performance and signal the company’s year end performance. The stock market reacted by quoting a higher or a lower stock price, or in some cases the market did not give a reaction and the stock price remained the same.

Signals from the company’s management are not just about company’s performance, but also many other things such as corporate actions. In addition, signals can also come from sources other than the company itself. This research is based on an assumption that the stock market absorbs all the signals and interpreted them into stock prices quotations. If the signals give sufficient information, and the market interpret it rationally, then the stock prices will correctly reflect the company’s actual performance.

However, there are also chances that the sources give false signals to the market. A good example for this is the Enron case. In August 2000, Enron's stock price hit its highest value of $90. At this point Enron executives, who possessed the inside information on the hidden losses, began to sell their stock. At the same time, the general public and Enron's investors were told to buy the stock.

Executives told the investors that the stock would continue to climb until it reached possibly the $130 to $140 range, while secretly unloading their shares. As executives sold their shares, the price began to drop. Investors were told to continue buying stock or hold steady if they already owned Enron because the stock price would rebound in the near future. The Enron’s management has misled the market by signaling the false information.

Taking a comparative case study as the method, this research involves two stock markets: Indonesia Stock Exchange and New York Stock Exchange. The observation results of the two distinct cases; one comes from a developing country and the other from a developed country, then be compared.

(6)

5 2. LITERATURE REVIEW

2.1. Signaling Theory

The signaling theory is based on an assumption that information is distributed unevenly across economic agents. Managers of the firm possess more information than outsiders regarding a project’s viability, expected profits, risk exposure, etc (Levy & Lazarovich-Porat, 1995). They can signal this information to outsiders by various means, e.g., choosing a particular dividend policy, selecting a particular capital structure, and so on. The signal could also take shape as the willingness of a company to buy its rivals (Snyder, 2009).

However, the signals not only come from the management itself but also from other sources, such as the government, domestic market, world market, etc. The changes in a certain government regulations for instance, can affect the company’s future performance.

The stock market absorbs all of these signals and translated into the stock price. In other words, stock prices can be considered as the reflections of stock market’s expectation on company’s future performance.

Dow & Gorton (1997) argue that the stock market indirectly guides investment by transferring two kinds of information: information about managers’ past decisions and information investment opportunities. With these kinds of information, the stock market has both monitoring role and prospective role.

The monitoring role involves evaluation of what the firm has done that the market considers can affect not only the firm’s current performance but also its future performance. The prospective role of stock prices arises because we allow the market to have information the manager does not already have. This potentially allows the current stock price to be of value in making current investment decisions.

(7)

6 price goes down, the manager is less likely to invest; if the price goes up, he/she is more likely to invest.

Nonetheless, there is also a possibility that financial analysts are overly optimistic (pessimistic) about their expectation on earnings per share, which can cause mispricing (Bartov & Kim, 2004). Mispriced stock can be misleading for the manager’s investment decision. If mispricing happens quite often, it is only logical that stock market should not be treated as a reliable indicator of what is happening in the real business sector.

Ferri and Chung-Ki (1996) provide the evidence of stock market's overreactions. Their research examines The Standard & Poor's 500 common stock index over a period of 1962 to 1991. Two criteria of overreactions and adjustments measurement were developed. First, the market changes direction in a substantial way and moves from a large positive return on one day to a similarly sizable negative return on the next. Second, the market's change in the direction is not explainable by the arrival of new information, but rather reflects a reaction to previous changes in equity values. The results suggest that the stock market overreacted in some occasions.

2.2. Types of Signals

Researches which study the variables that influence the stock price are quite numerous. Based on these researches, the signals--received by stock market, indicating company’s performance, can be divided into two categories, the internal signals and the external signals.

Internal Signals

(8)

7 The stock market will then receive the internal signals and give a reaction by raising or lowering the stock price. In some cases, the stock market may not give a reaction and therefore the stock price remains the same. Bae and Jo (1999), provide evidence that the release of new information regarding the state of a company affects both price volatility and volume of underlying stock.

Many researches which examined internal signal conducted the testing of dividend signaling theory (Rimbey & Officer, 1992; Garret & Priestley, 2000; Murhadi, 2008). The dividend signaling theory suggests that the announcement of dividend policy by the company’s management will affect the company’s stock price in the market.

The researches have not fully confirmed that dividend signaling theory is relevant. While the other two provide results showing a positive relationship, research by Rimbey & Officer (1992) suggest a more complex results. The positive relationship between the announcement of dividend policy and stock price is only relevant for a type of firms which investors believed will be able to sustain the dividend's payment.

Studies on the relationships between earnings announcements and stock prices have moved even further in several researches. The researches are not only showing that the announcements affect stock prices, but also suggest there is variability in stock market's response.

Chen and Lin (1997) attribute earnings surprises and subsequent stock price changes to the quality and quantity of available information. If a stock is followed by many financial analysts, the amount of information available to investors contributes to higher quality information. As the quality and quantity of information increase, stock prices adjust more quickly.

(9)

8 not. They also confirm market participants take different amounts of time to process the information conveyed by the earnings announcement.

Researches which study the effect of company’s corporate action on stock prices, whether the action is announced or made public by a different source, show results that match the prediction. These results follow the bad news and good news postulate. Actions which believed reflecting or will lead to a poor performance get a negative response. On the contrary, good news' actions will increase the stock prices.

Shahid and Sami (1994), for instance, examine the stock market reaction to equity-for-debt swap announcements. The results showed negative abnormal returns upon swap announcements. They suggest that swap conveys management information about the prospects of poor future economic condition of the firms.

Other research in the financial control's subject, studies the stock price reaction to news about corporate tax aggressiveness. Hanlon and Slemrod (2009) suggest, on average, a company's stock price declines when there is news about its involvement in tax shelters. They find the reaction is more negative for firms in the retail sector, suggesting that part of the reaction may be a consumer/taxpayer backlash.

A study on the British stock market reaction to the announcement of different types of capital expenditures, show that only joint venture is believed as good news (Burton and Lonie, 1999). Joint ventures appear to surprise the market and enhance market expectations on the future prospects of the participating firms, and therefore relate positively to the stock prices.

Table 1. Correlations between Internal-External Signals and Stock Prices

Type of Signals Subject Relationship Sign Author

positive Garret & Priestley

(2000), Murhadi

(2008)

Internal Performance Dividend-Stock

Price

partially positive; only when firm is believed will be able to sustain

(10)

9 the payment

positive; adjust more quickly when the quality

and quantity of

information are higher.

Chen & Lin (1997) Earnings-Stock

Price Changes

Variability

positive; depends on the degree of analysts'

pre-disclosure earnings

forecast dispersion

Lobo & Tung

(2000)

Equity-for-debt swap-Stock Price

negative Shahid & Sami

(1994) Tax shelter-Stock

Price

negative Hanlon & Slemrod

(2009) Corporate Actions Type of capital expenditure-Stock Price

positive for joint venture Burton & Lonie,

(1999) Exchange rate-Stock Price positive Dimitrova (2005), Tabak (2008) Interest rate-Stock Price differentiated according to Federal reserved

monetary policy and

motive behind the

discount rate change

Jensen & Johnson (1993)

External Macroeconomic

Indicators

Inflation-Stock Price

negative in short run; positive in ling run

Al-Khazali and

(11)

Master Thesis International Business & Management Rijksuniversiteit Groningen 2009 10 Government Regulations Stricter Tobacco Reg-Stock Price positive news External Signals

The second category is the external signals. They come from various sources outside the management, such as macroeconomic indicators, government regulations and industry. The signals are not part of company's corporate actions or directly related to company's performance.

Dimitrova (2005) found support for the hypothesis that a depreciation of the currency may depress the stock market—the stock market will react with a less than one percent decline to a one percent depreciation of the exchange rate. This also implies that an appreciating exchange rate boosts the stock market. Tabak (2006) confirms the suggestions, that exchange rate and stock price are indeed related. In addition he finds that one has predictive power to forecast the other.

There are also evidences stock prices follow the inflation and interest rate movement (Golob & Bishop,1996; Apergis & Eleftheriou, 2002). However, the market reaction does not inline with the strict definition that discount rate increases are considered bad news, while rate decreases are good news (Jensen & Johnson, 1993). The market reaction is differentiated according to Federal reserved monetary policy and motive behind the discount rate change.

(12)

11 memory associated with inflation shocks that make stock portfolios a reasonably good hedge against inflation in the long run.

Changes in or new government regulations can also affect company’s performance. Stricken regulation regarding tobacco products, for example, may cause revenues of companies in the related industry shrinking. When a panel of researches and policy experts issued a recommendation for the US Food and Drug Administration (FDA) which include stricter regulations of tobacco advertising, The Global Tobacco Index was down by 9.3% (Havrilla, 2009).

2.3 Conceptual Framework

Based on signaling theory and research’s findings presented in the previous paragraphs, we can build a conceptual model for a starting point of this research. The model explains the determinants of earnings per share and stock prices and how each variable interact with others.

Company’s productivity determines the outcome of company’s performance (represented by earnings per share) and at the same time sends internal signals to the stock market (see Firgure 1). The signals provide information for the stock market regarding the current state of the company's performance which in turn will affect its future performance. These signals can be clear information or just a hint. It is up to the stock market to evaluate and then interpret the information into stock prices. Nonetheless, the quality and quantity of the information play an important role in the interpretation.

In the mean time, stock market also receives external signals. The exchange rates, inflation and interest rates are indicators of level of economic growth in a macro level. Higher economic growth creates a more favorable business environment which is needed for the growth of companies.

(13)

12 the company's fiscal year. The same changes that send external signals to market create opportunity/threats for related company, which influence the corporate management.

As Dow and Gorton (1997) suggest the flow of information in a stock market may be bidirectional: the market may want to learn about the quality of the managers’ decisions. Nevertheless, the managers may also want to learn the market’s valuation of prospective investments.

Figure 1. Determinants of Stock Price and Earnings per Share

Stock Price (reflection of company’s performance)

Earnings per share (actual company’s performance) External signals (macroeconomics indicators, government regulations, industry-specific dynamics, etc)

Market opportunity and

threat Internal signals

(corporate actions, quarterly & annual

Company’s productivity

(14)

13 The stock price, although intrinsically irrelevant to the investment decision, may be useful indirectly because it conveys information about prospective investment projects and cash flows. For example, a high stock price may signal to the manager that the market believes the firm has profitable investment opportunities.

Not all relevant information is taken as a consideration by the manager, and, consequently, he can improve his decisions by using the prospective stock price. In other words, the stock market may have important information that the manager neglected to take account.

On the contrary, the market not always receives the proper internal signals from the managers. For many reasons, managers could hide the information concerning the firm’s current and/or future performance.

Furthermore, we should not neglect the fact that stock market can be overreacted. Bulkley and Harris (1997) found evidence that analysts’ expectations on earnings per share are irrational. Based on their study on large numbers of US companies trading the New York and American Stock Exchange, analysts’ growth forecasts are uncorrelated with actual growth.

Nevertheless, there is a possibility that the analysts’ forecasts are not reflected in the stock prices. A direct use of stock prices to measure market’s expectation might be able to provide a different insight.

Hypotheses

If the stock market is able to absorb all the signals and translated into stock prices accordingly, it is very likely that the stock prices movement will have the same direction as the actual company’s performance (represented by earnings per share). When this is happening, the stock market can be considered as one of indicators of company’s actual performance. Here, we can say that the stock prices’ quality is good.

(15)

14 information are higher. This also suggests the market feel confident with their evaluation about a company’s current state of performance and what might happen to the company’s future performance. And therefore, they should be able to generate the correct stock prices.

In the stock market level, there are at least two factors that can potentially influence the quality of stock prices. First, the institutional governance of the stock market itself. A stock market with good institutional governance makes sure that the flow of information can reach every investor equally. The rule of transparency is not only become the highest priority but also strictly implemented. Good institutional governance is usually built over a relatively long experience.

The second factor is stock market’s size. As applied in common market; the higher the number of player, the higher the level of competitiveness and the market become more efficient. Within an efficient system, stock market should be able to generate a high quality stock price.

Based on stock market-level characteristic’s argument, a hypothesis can be formulated, as follows:

H1. The larger and more established a stock market, the better its ability to absorb internal and external signals and then correctly interpreting the signals as indications of company’s performance.

Stock market only provides a minimum standard of transparency. Firms can just follow the standard to comply, or they can apply a higher degree of transparency.

(16)

15 Based on the above argument, two hypotheses can be formulated as follows: Figure 2. Correlation between Stock Price and Earnings per Share and Its Determinants

Stock price at the end

of fiscal year Stock price is able

to reflect company’s actual

performance accordingly Annual earnings per

share, which is announced 1-3 months after the end of fiscal year

correlate, same direction

Stock Market Characteristic

- Institution Governance

- Market Size

Firm Characteristic

- Firm Size

- Degree of Transparency

(17)

16 H2. The higher the degree of firm’s transparency, the better the quality of stock prices quoted by the stock market.

H3. The larger the firms, the higher the quantity and quality of information in internal signals and therefore the quality of stock prices is higher.

3. METHODOLOGY

3.1. Case Background

The research takes form as a comparative case study with two stock markets, Indonesia Stock Exchange and New York Stock Exchange (NYSE), as the target of observation. The Indonesian stock market is chosen because of its vulnerability. It is commonly believed that the stock market is very sensitive to the external changes that are not directly related to the company’s future performance. Furthermore, the Indonesian stock market is often overreacted in translating the market/management’s signals. Especially, the negative signals. The NYSE will be used as a comparison because the market is well-established and often viewed as one of global business indicators. By choosing two contrasting stock markets, this research tries to present a representative for each kind of market and see if they are really different in terms of the ability to reflect the company’s performances.

Indonesia Stock Exchange (IDX)

(18)

17 Exchange and Surabaya Stock Exchange merged and named Indonesian Stock Exchange (IDX) by Indonesian Minister of Finance.

Three of the primary stock market indices used to measure and report value changes in representative stock groupings are the JSX Composite, the Jakarta Islamic Index (JII) and LQ45. The JII was established in 2002 to act as a benchmark in measuring market activities based on Sharia (Islamic Law). Currently, there are approximately 30 corporate stocks listed on the JII. LQ45 is a group of 45 stocks with highest market capitalization and the most liquid during a certain period of time.

According to IDX annual report 2008, the stock exchange managed to raise market capitalization of 1,076.49 trillion rupiahs in the equity market. Until the end of 2008, there are 396 companies listed in the stock exchange.

ew York Stock Exchange (YSE)

The origin of the NYSE can be traced to May 17, 1792, when the Buttonwood Agreement was signed by 24 stock brokers outside of 68 Wall Street in New York under a buttonwood tree on Wall Street. On March 8, 1817, the organization drafted a constitution and renamed itself the "New York Stock & Exchange Board". Anthony Stockholm was elected the Exchange's first president.

Now, the stock exchange which is located at 11 Wall Street in lower Manhattan, New York, USA, has become the largest stock exchange in the world by dollar value of its listed companies' securities. The NYSE is operated by NYSE Euronext, which was formed by the NYSE's 2007 merger with the fully-electronic stock exchange Euronext.

(19)

18 3.2. Analyses Steps

This research conducted two stage analyses. First, correlations analyses over a set time series data of the stock prices and company’s earnings per share. The method used here is bivariate correlations analysis and the results are used to develop the dependent variable for the next stage.

In this stage we will see the rate of significant correlations in both IDX and NYSE cases. The higher the rate, the better the ability of stock price in reflecting the company’s performance. At this point of analysis step, the potential outcomes will be symmetry. According to Gill and Johnson (2003), symmetry of potential outcomes happens when the results will be equally valuable. If no correlation between stock price and earnings per share were found, it would be as interesting and important as if there were found to be a high correlation.

Regarding the Indonesia Stock Exchange analysis, the result can be considered as the follow up of previous research on Indonesia Stock Exchange. Murhadi (2008) has found evidence that dividend policy announcements affect the stock prices. The stock market absorbs the signal provided by managers, in which takes the dividend policy’s announcement as the channel, and therefore try to forecast the company’s future performance. How well the Indonesia stock market translates the signal (and a bunch other signals), we should be able to see it by observing the movement of the stock prices and companies’ earnings per share.

If we take the above common belief on Indonesia Stock Exchange, this research will be expected to get high rate of non significant correlation cases. It means the movement of stock prices mostly deviates from the movement of company’s earnings per share. The stock market failed to give a relatively good reflection of company’s performance. Therefore, the stock price in the stock market is not a reliable source of information for managers to decide the company’s financial investment.

(20)

19 can be a good source of information for further investment. . The confident level to measure significant results in this research is 0.05.

The same way of analysis will be applied to the NYSE case and then compare the result to the Indonesia Stock Exchange case. The comparison is necessary to build a result that can be generalized. The result on the NYSE case is expected to be largely significant because the market are considered well established with high level of transparency, therefore should be more rational and better in forecasting a company’s future performance. However, based on the research by Bulkley & Harris (1997), it is very likely the analysis on NYSE case will provide evidence that the stock prices and earnings per share are mostly not correlated.

The second stage is a regression analysis to investigate which factors influencing the appearance of significant correlations between stock prices and earnings per share. The method used here is the linear regression analysis with a dependent variable taken from results of the first stage of research analysis. A dummy variable is created by categorizing significant correlations cases and non significant correlations cases into two separate groups. The dummy variable acts as the dependent variable in the regression analysis.

3.3. Independent Variables

This research formulated one independent variables and a control variable for the second stage of analysis. Act as the independent variable, firm-level characteristics are represented by firms size, with market capitalization as the indicator, and the degree of transparency with frequency of news issuance by the company as the indicator. The industry-level characteristic used industry categorizing as an indicator and plays the role as a control variable.

Market Capitalizations

(21)

20 However, the distribution of the market capitalization’s values in both markets is too widely dispersed. Therefore, to simplify the data this research grouped the market capitalizations into three sizes: large, medium, small.

Table 2. Descriptive Statistics Market Capitalizations before Grouping

N Minimum Maximum Mean Std. Deviation

Market Cap IDX 42 0.95 trillion IDR 130.98 trillion _IDR 21.87 trillion _IDR 30.09 trillion _IDR

Market Cap NYSE

43 18.03 billion USD 335.92 billion USD 71.3544 billion USD 59.77966 billion USD Valid N (listwise) IDX 42 Valid N (litwise) NYSE 43

By dividing the market capitalizations into three groups and each group consists of an equal number of companies, the parameters are set based on the lowest value in large market capitalizations and medium. Since the market capitalizations differ from one stock market to the other, the grouping parameters also different between IDX and NYSE. Based on the three groups, a dummy variable is developed.

For IDX case, the companies with market capitalizations over 17 trillion rupiahs were in large size group. The medium size group consisted of companies with market capitalizations ranging between 4.5 trillion to 17 trillion rupiahs. The small group has market capitalizations less than 4.5 trillion. The market capitalizations data is dated on 31st January, 2009.

(22)

21 ews Feed Frequency

The news feed frequency variable is based on how many written information a company has sent to and published by the stock exchange authorities. This research assumed that the quantity of information flow determined the degree of transparency. Since the data is too widely dispersed, the data is simplified by grouping it into three categories: most frequent, medium and least frequent. These groups categorizing is used to create a dummy variable.

Table 3. Descriptive Statistics News Feed before Grouping

N Minimum Maximum Mean Std. Deviation

News Feed IDX 42 189 1206 480.8810 270.37149

News Feed NYSE 43 696 8613 2009.5814 1782.76848

Valid N (listwise)

IDX 42

Valid N (litwise) NYSE

43

The parameters for each group are set in the same way as market capitalizations’ grouping. The most frequent group in IDX market are cases with more than 465 information publications, the medium were set in range of 280-465 publications and the least frequent with less than 280 publications. Whereas for NYSE, the most frequent group has more than 1,849 publications, the medium group is ranging from 1,104 to 1,849 publications, and the least frequent group set in less than 1,104 publications.

Industry Categorizing

(23)

22 some adjustments to make it more compatible with the nature of Indonesian companies. ICB has 10 categories of industry, whereas IDX classification consists of 9 categories.

ICB industries are Oil & Gas, Basic Materials, Industrials, Consumer Goods, Health Care, Consumer Services, Telecommunications, Utilities, Financials, and Technology. In the mean time, IDX classifications are Agriculture, Mining, Basic Industry & Chemicals, Miscellaneous Industry, Consumer Goods Industry, Property & Real Estate; Infrastructure, Utilities & Transportation; Finance; and Trade, Service & Investment.

3.4. Samples and Data Collection

In the case of Indonesia stock market, the samples consist of 42 stocks. The 42 stocks are taken from the LQ45 board, a group of stocks which have the most active traded shares. This is necessary in order to avoid a bias result from the use of inactive traded share. Inactive kind of shares would have prices that remain stable over a relatively short period or even in a long period.

There should be 45 stock in LQ45 can be taken as samples. However, since the list of LQ45 was updated every six months, a company may not in be listed in some periods only, not in every period. Only companies that able to be in LQ45 board for at least six times (50%) over a period of August 2003 until January 2009 were selected. The time period is matched with the period of earnings per share and stock prices which were used for this research.

The IDX website published online annual financial reports of all its listed companies, but only available for a period of 2003-2008. Therefore the earnings per share data that was obtained form the online database comprised of six figures for each company.

(24)

23 Each stock price data is taken from the time of the end of the company’s fiscal year. Most of the companies have fiscal years which are ended on 31st December and therefore the stock price taken also from that date, or if not available, the day after. Stock prices quoted in the end of fiscal years are the result of a year-full of various signals absorbent. Therefore, if the prices are correctly interpreting the signals, they should be able to indicate the company’s fiscal-end performance. There always a few months delay after the fiscal year ended, for a company to announce its annual report. That is why we can use the word ‘indicate’ whereas the more correct word to define it should be ‘reflect’. The reason to choose a single date because it is the same stock price that usually presented by the corporate management when announcing the company’s annual reports. The end-year stock price represents the market value of the company.

4. RESULTS AD AALYSES

This section presents the results evaluation of Indonesian Stock Exchange (IDX) and New York Stock Exchange (NYSE) markets separately. The separation is necessary to give a thorough results’ presentation for each market before ending it with the comparison.

4.1. Indonesia Stock Exchange IDX Correlation Analysis

(25)

24 Tabel 4. Percentage of Significant Correlation’s Cases in Indonesia Stock Exchange

Frequency Percent Valid Percent Cumulative Percent Valid Significant* 6 14.3 14.3 14.3 Non Significant 36 85.7 85.7 100.0 Total 42 100.0 100.0 * significant level 0.05

However, in the cases that have shown significant results, the ability of stock exchange in reflecting the company’s performance is largely accurate. The quality of this ability is based on the observation on correlations coefficients, which in those cases the values are all more than 0.8. These values are indication of strong correlations.

One of the cases with significant correlations showed a negative relationship. It involved a communication company, Indosat Tbk. The stock prices quoted by IDX were heading to the opposite direction of the company’s realized performances. Although it can be categorized as a significant predictor of company’s performance, the negative direction made it a bad predictor. When the company’s performance is rising; indicated by the rising of net earnings per share; the stock price is decreased, in which it should be rising.

Table 5. Significant Correlations Cases in Indonesia Stock Exchange

No Code Company Correlation Coefficient*

1. TINS Timah 0.994

2. ANTM Aneka Tambang 0.941

3. SMGR Semen Gresik 0.922

4. BBCA Bank Central Asia 0.908

5. PGAS Perusahaan Gas Negara 0.896

6. ISAT Indosat -0.871

(26)

25 IDX Regression Analysis

Based on cross-tabulations descriptive statistics, all cases with significant correlation belong to groups of firms with large and medium market capitalizations. There are four cases with significant correlation involve large market capitalization firms. Two cases are in the medium market capitalization group. The similar cross-tab statistics also shown by news feed. None of the significant cases are in the least frequent group. There are four in frequent group and two belongs to medium group.

The cases with significant correlations and positive directions are mostly found in the mining industry. Three of the eight cases (37.5%) in the mining industry showed this

Tabel 6. Cross-Tabulation Statistics Significant Correlation Cases, Market Capitalizations and Industry in Indonesia Stock Exchange

Industry

Market Capitalization Total

Large Medium Small

Correlation Siginificant 1 2 3 Non Significant 3 2 5 Mining Total ₄ ₄ ₈

Correlation Non Significant 1 2 3

Agriculture

Total 1 2 3

Miscellaneous

_Total ₁ ₁ ₂

Correlation Non Significant 1 2 1 4

Trade, Service & Investment Total 1 2 1 4 Correlation Significant 1 0 0 1 Non Significant 2 4 1 7 Finance Total ₃ ₄ ₁ ₈ Correlation Significant 1 0 0 1 Non Significant 0 2 3 5

Basic Industry & Chemicals Total 1 2 3 6 Correlation Significant ₁ ₀ ₁ Non Significant 1 2 3 Infrastructure Total 2 2 4

Correlation Non Significant 4 4

Property

_Total

(27)

26

Correlation Non Significant ₁ ₂ ₃

Consumer Goods

Total 1 2 3

Tabel 6. Cross-Tabulation Statistics Significant Correlation Cases, News Feed and Industry in Indonesia Stock Exchange

News Feed Industry _Frequent _Medium Least Frequent Total Significant 2 1 3 Correlation _{Non Significant} ₂ ₃ ₅ Mining Total 4 4 8

Correlation Non Significant 1 1 1 3

Agriculture

_Total ₁ ₁ ₁ ₃

Miscellaneous

_Total ₁ ₁ ₂

Trade, Service & Investment Total 2 2 4 Significant 1 0 1 Correlation Non Significant 5 2 7 Finance Total 6 2 8 Significant 1 0 1 Correlation _{Non Significant} ₁ ₄ ₅

Basic Industry & Chemicals Total 2 4 6 Significant 1 0 1 Correlation _{Non Significant} ₀ ₃ ₃ Infrastructure Total 1 3 4

Property

_Total ₁ ₃ ₄

Consumer Goods

Correlation Non Significant

1 2 3

Total 1 2 3

(28)

27 Tests on the role of firm’s characteristic, with industry as control variable, have proven that none of the variables influence the correlation between stock prices and net earnings per share. Even though the cross-tab statistics shows that the appearance of significant cases seemed to follow the size of market capitalizations and news feed, the regression analysis suggests otherwise. The significant cases appear randomly across the three groups of market capitalization and news feed. In addition, the significant correlations also appear randomly throughout four industries: mining, finance, basic industry and infrastructure.

Table 7. Output Regression Analysis in Indonesia Stock Exchange Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .382(a) .146 .078 .34002

a Predictors: (Constant), Industry, MarkCap, NewsFeed

ANOVA(b)

Model

Sum of

Squares df Mean Square F Sig.

Regression .750 3 .250 2.161 .109(a)

Residual 4.393 38 .116

1

Total 5.143 41

a Predictors: (Constant), Industry, MarkCap, NewsFeed b Dependent Variable: SigCorr

Coefficients(a) Unstandardized

Coefficients

Standardized

Coefficients t Sig.

Model B Std. Error Beta B Std. Error

(Constant) 1.468 .163 9.014 .000

MarkCap .092 .077 .214 1.196 .239

NewsFeed .080 .082 .183 .975 .336

1

Industry .010 .022 .073 .446 .658

(29)

28 4.2. ew York Stock Exchange

YSE Correlation Analysis

Significant correlation appearances are more often to be found in this market, covering half of the 43 cases, or 51.2% to be exact. The computation showed that in those cases, stock prices quoted in the end of fiscal years indeed correlate with related company’s annual net earnings per share. The rate of significant cases suggested that NYSE can be considered as a moderate reflection of what is going on with the companies, especially in term of performance.

Tabel 8. Percentage of Significant Correlation’s Cases in New York Stock Exchange

Frequency Percent Valid Percent

Cumulative Percent Valid Significant * 22 51.2 51.2 51.2 Non Significant 21 48.8 48.8 100.0 Total 43 100.0 100.0 *significant level 0.05

Among the cases with significant correlations, one of them shows a negative beta coefficient. This result implies that in that case, the stock exchange has quoted rising stock prices while in fact the company’s performance was declining. In other occasions, the stock prices were declining when the company’s earnings per share were rising. The negative correlation involved a medical equipments company, Medtronic.

Table 9 New York Stock Exchange Significant Result Cases

No Code Company Correlation

Coefficient

No. Code Company Correlation

Coefficient

1. GS Goldman Sachs 0.991 12. GE General

Electric

0.698

2. BMY Bristol-Myers 0.930 13. XOM Exxon 0.651

3. MS Morgan Stanley 0.892 14. SLB Schlumberger 0.560 4. HPQ Hewlett-Packard 0.875 15. LMT Lockheed Martin 0.516

(30)

29

6. USB US BankCorp 0.846 17. UTX United

Technologies 0.506 7. AXP American Express 0.826 18. HAL Halliburton 0.500 8. BAC Bank of America 0.800 19. KMB Kimberly Clark 0.472

9. PEP Pepsi 0.773 20. LLY Eli Lily 0.465

10 SGP

Schering-Plough

0.759 21. BA Boeing 0.460

11. JPM JP Morgan 0.746 22. MDT Medtronic -0.517

YSE Regression Analysis

The number of significant case appearances seemed to follow the size of market capitalizations, but in the opposite direction. Based on the cross-tabulation statistics, there are six cases found in the large market capitalization groups, seven in medium and nine among small capitalization firms. The news feed cross-tab shows a different distribution. The number of significant cases is relatively equal in each group. There are seven cases in the highest news feed frequency group, eight in the medium and seven in the least frequent group.

Financial industry showed the highest appearances of cases with significant correlations between stock prices and companies earnings per share. The industry has got six out of eight cases (75%) indicating significant correlations.

There are three industries, in which significant correlation case was not found. These industries are basic materials, consumer services and telecommunications. However, two of the three industries, each only had one case as a representative.

(31)

30 Tabel 10. Cross-Tabulation Statistics Significant Correlation Cases, Market

Capitalizations and Industry in New York Stock Exchange

Industry MarketCap

Large Medium Small

Total

Industrial Correlation Significant 1 1 2 4

Non Significant 0 1 2 3

Total 1 2 4 7

Health Care Correlation Significant 0 3 1 4

Total 2 6 1 9

Consumer Goods Correlation Significant 1 1 2

Non Significant 2 2 4

Total 3 3 6

Financial Correlation Significant 1 2 3 6

Total 2 2 4 8

Oil & Gas Correlation Significant 1 1 1 3

Total 2 1 1 4

Utilities Correlation Significant 1 1

Total 1 1

Basic Materials Correlation Non Significant 1 1

Total 1 1

Technology Correlation Significant 2 2

Total 2 2

Consumer Service Correlation Non Significant 1 3 4

Total 1 3 4

Telecommunications Correlation Non Significant 1 1

Total 1 1

(32)

Master Thesis International Business & Management Rijksuniversiteit Groningen 2009 31 Total 2 2 3 7 Significant 0 1 3 4 Correlation _{Non Significant} ₁ ₂ ₂ ₅ Health Care Total 1 3 5 9 Significant 0 1 1 2 Correlation _{Non Significant} ₁ ₁ ₂ ₄ Consumer Goods Total 1 2 3 6 Significant 5 1 6 Correlation _{Non Significant} ₂ ₀ ₂ Financial Total 7 1 8 Significant 2 1 3 Correlation _{Non Significant} ₀ ₁ ₁ Infrastructure Total 2 2 4 Correlation Significant 1 1

Oil & Gas

_Total ₁ ₁

Basic Materials

_Total ₁ ₁

Correlation Significant 1 1 2

Technology

_Total ₁ ₁ ₂

Consumer Service

_Total ₂ ₂ ₄

Telecommunications

_Total ₁ ₁

The regression analysis suggests both market capitalizations and news feed failed to explain the development of significant correlation between stock prices and earnings per share. In addition, the significant correlation cases appearance happens randomly throughout seven industries: industrial, health care, consumer goods, financial, infrastructure, oil & gas and technology.

Table 12. Output Regression Analysis in New York Stock Exchange Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .186(a) .035 -.040 .51567

(33)

32 ANOVA(b)

Model

Sum of

Squares df Mean Square F Sig.

Regression .373 3 .124 .468 .706(a)

Residual 10.371 39 .266

1

Total 10.744 42

a Predictors: (Constant), Industry, NewsFeed, MarketCap b Dependent Variable: SigCorr

Coefficients(a) Unstandardized

Coefficients

Standardized

Coefficients t Sig.

Model _B _{Std. Error} _Beta _B _{Std. Error}

(Constant) 1.473 .351 4.199 .000

MarketCap -.068 .099 -.111 -.687 .496

NewsFeed .029 .097 .048 .299 .766

1

DummyIndustry .024 .032 .124 .753 .456

a Dependent Variable: SigCorr

5. DISCUSSIO

Results on correlations testing between stock prices, quoted in the end of fiscal years, and annual earnings per share have confirmed one of this research’s hypotheses. The results suggest that NYSE, as a more established stock exchange market, was able to reflect company’s performances better than IDX.

However, it is possible that the confirmed hypothesis is only valid in the two specific stock exchanges. Results from a case study with a very limited number of cases as samples may not be sufficient to be generalized.

(34)

33 performance was actually rising, and vise versa. Of course, lack of proper signals indicating the performance could also be the reason it happened.

In the attempt to determine the factors that influence significant stock price-earnings per share correlations, the results shows that they are remain unknown. Both markets generate the same results regarding the role of firm characteristics with industry as a control variable.

Table 13. Results Comparisons Indonesia Stock Exchange versus New York Stock Exchange

Independent Variables Control Variable

Market Sig.Correlations

Stock Price-EPS* Market

Capitalizations News Feed Frequency Industry Indonesia Stock Exchange

14.3% non significant non significant non significant

New York Stock Exchange

51.2% non significant non significant non significant

*EPS = earning per share

The firm’s size, as measured as market capitalizations, can not predict the quality of stock prices quoted by stock exchange. Argument developed in the Conceptual Framework subsection to formulate the firm’s size hypothesis can be partially right. Larger company tends to draw larger amount of news and information from other sources which sometimes the news forced the corporate management to release information.

Nevertheless, we should realize that there are researches which have proven that stock market can be overreacted. On the other hand, stock market also faces difficulties in trying to quote correct prices for stock of smaller company, due to the lack of information.

(35)

34 important details and therefore the quality of information is lower than it supposed to be. The problem is only the company knows about that and therefore stock market can only give an interpretation based on its own judgment. In this case it is very difficult to measure the quality of the information.

6. COCLUSIO

Stock exchange markets are not a reliable reflection of the real sector performances. Even the NYSE, which is considered the largest equities-based exchange in the world, the chances to produce the correct stock prices are no more than 52%. Correct, in the sense of reflecting companies’ performances. In this sense, correct prices should follow the trend of the companies’ performances.

The ability of a more well-established and larger stock market to produce correct stock prices seems to be stronger than a smaller and underdeveloped market. Nonetheless, the hypothesis still needs further investigations using a much larger number of markets as samples.

This research can not find factors that determine the forming of stock prices which correctly reflecting companies’ performances. Two possible factors, firm’s size and the degree of transparency, with industry categorizing as control variable, failed to give an explanation. Future research should investigate any other factors that might contribute to the producing of correct stock prices.

6.1. Limitations

(36)

35 The lack of previous research that examined the exact same variables causes a

problem on determining the correct formula of analyses. At first, I could not decide the perfect timing of stock prices to be used for the sample’s data. There is no reference that pointing the right timing. Therefore I have to develop my own argument which of course it is widely open to be debated.

The period of stock prices and earnings per share datasets which being used for IDX are shorter than NYSE. The IDX’s dataset starts from 2003, whereas the NYSE’s dataset starts from 1995. It is possible that this difference could affect the results of correlations testing.

REFERECES

Al-Khazali, Osamah M., and Chong, Soo Pyun. 2004. Stock Prices and Inflation : New Evidence from the Pacific-Basin Countries. Review of Quantitative Finance and Accounting 22 (2) :123-140.

Apergis, Nicholas & Eleftheriou, Sophia. 2002. Interest rates, inflation and stock prices: the case of the Athens Stock Exchange. Journal of Policy Modeling. 24(3):231-236. Bae, Sung C., Jo, Hoje. 1999. The Impact of Information Release on Stock Price Volatility and Trading Volume : The Rights Offering Casa. Review of Quantitative Finance and Accounting, 13 : 153-169.

Bartov, Eli., and Kim, Myungsun. 2004. Review of Quantitative Finance and Accounting, 23 : 353-376.

Bulkley, George and Harris, D.F. 1997. Irrational Analysts’ Expectations As A Cause of Excess Volatility in Stock Prices. The Economic Journal, 107: 359-371.

Burton, Bruce M. and Lonie, A. Alasdair. 1999. The stock market reaction to investment announcements: The case of individual expenditure projects. Journal of Business Finance & Accounting 26 (5/6 ): 681 (28).

(37)

36 Dimitrova, Desislava. 2005. The Relationship between Exchange Rates and Stock Prices: Studied in Multivariate Model. Political Economy Vol.14.

Dow, James., and Gorton, Gary. 1997. Stock Market Efficiency and Economic Efficiency : Is There a Connection? The Journal of Finance, 52(3): 1087-1129.

Ferri, Michael G., and Chung-Ki Min. 1996. Evidence that the stock market overreacts and adjusts. Journal of Portfolio Management 22 (3): 71-76

Garret, Ian & Priestley, Richard. 2000. Dividend Behavior and Dividend Signaling. The Journal of Financial and Quantitative Analysis. Vol 35 (2): 173-189

Gill, J. and Johnson, P. 2003. Research Methods For Managers. Third Edition. Sage Publications.

Golob, John E., Bishop, David G. 1996. Do stock prices follow interest rates or inflation? Research Working Paper 96-13. Federal Reserve Bank of Kansas City.

Havrilla, Mike. 2009. Regulations to Affect Tobacco Tracking Index.

http://seekingalpha.com/article/123487-regulations-to-affect-tobacco-tracking-index Hanlon, Michelle and Slemrod, Joel. 2009. What does tax aggressiveness signal? Evidence from stock price reactions to news about tax shelter involvement. Journal of public economics, ISSN 0047-2727 93 (1): 126-141.

Levy, Haim & Lazarovich-Porat, Esther. 1995. Signaling theory and risk perception: An experimental study, Journal of Economics and business. Vol 47 (1) : 39:56

Lobo, Gerald J., and Tung, Samuel S. 2000. Financial Analysts' Earnings Forecast Dispersion and Intraday Stock Price Variability Around Quarterly Earnings. Review of Quantitative Finance and Accounting 15 (2) : 137-151

Murhadi, Werner R. 2008. Studi kebijakan Dividen: Anteseden dan Dampaknya terhadap Harga Saham. Jurnal Manajemen dan Kewirausahaan 10 (1): 1-17.

Prager, Robin A. 1989. Using Stock Price data to Measure the Effects of Regulation: The Interstate Commerce Act and the Railroad Industry. RAND Journal of Economics. 20(2):280-290.

(38)

37 Shahid, Abdus and Sami, Heibatollah. 1994. Common stock returns on swaps

announcements and firms' specific variables. Atlantic Economic Journal 22(1): 91, 1/3p. Snyder, Andrew. 2009. How Pfizer Signaled A Difficult Year for Shareholders.

http://www.contrarianprofits.com/articles/how-pfizer-pfe-signaled-a-difficult-year-for-shareholders/10843

(39)

38 Appendix 1. Indonesia Stock Exchange Cases

Code Company Correlation Coefficient Industry Market Capitalization (Rp) News Feed Frequency

TINS Timah 0.994 Mining 5.50E+12 299

ANTM Aneka Tambang 0.941 Mining 1.10E+13 568

SMGR Semen Gresik 0.922

Basic Industry &

Chemicals 2.10E+13 389

BBCA Bank Central Asia 0.908 Finance 6.70E+13 1167

PGAS Perusahaan Gas Negara 0.896 Mining 5.10E+13 701

ISAT Indosat -0.871 Infrastructure 3.10E+13 940

BUMI Bumi Resources .non sig. Mining 1.30E+14 437

TLKM Telekomunikasi Indonesia .non sig. Infrastructure 1.30E+14 340

UNVR Unilever Indonesia .non sig. Consumer Goods 6.00E+13 189

BBRI Bank Rakyat Indonesia .non sig. Finance 5.60E+13 945

ASII Astra International .non sig. Miscellaneous 5.30E+13 437

BMRI Bank Mandiri .non sig. Finance 3.80E+13 895

BNBR Bakrie and Brothers .non sig.

Trade, Service &

Investment 3.70E+13 467

INCO International Nickel .non sig. Mining 2.50E+13 672

AALI Astra Agro Lestari .non sig. Agriculture 1.70E+13 649

PTBA PT Bukit Asam .non sig. Mining 1.70E+13 430

BNII Bank International Indonesia .non sig. Finance 1.70E+13 788

UNTR United Tractors .non sig.

Trade, Service

and Investment 1.70E+13 528

INTP Indocement Tunggal Perkasa .non sig.

Basic Industry

and Chemicals 1.70E+13 251

GGRM Gudang Garam .non sig. Consumer Goods 1.60E+13 196

ENRG Energi Mega Persada .non sig. Mining 1.20E+13 460

BBNI Bank Negara Indonesia .non sig. Finance 1.20E+13 434

BDMN Bank Danamon Indonesia .non sig. Finance 1.10E+13 934

BNGA Bank CIMB Niaga .non sig. Finance 1.00E+13 1206

INDF Indofood Sukses Makmur .non sig. Consumer Goods 8.60E+12 328

MEDC Medco International .non sig. Mining 5.70E+12 737

RALS Ramayana Lestari Sentosa .non sig.

Trade, service

and Investment 5.40E+12 262

INKP Indah Kiat Pulp & Paper .non sig.

Basic Industry

SMCB Holcim Indonesia .non sig.

Basic Industry

LSIP PP London Sumatera .non sig Agriculture 4.10E+12 268

CMNP Citra Marga Nusaphala .non sig. Infrastructure 3.80E+12 324

PNLF Panin Life .non sig Finance 3.80E+12 460

BRPT Barito Pasific Timber .non sig

Basic Industry

(40)

39

CTRS Ciputra Surya .non sig Property 2.40E+12 264

BLTA Berlian Laju Tanker .non sig Infrastructure 2.20E+12 420

JIHD Jakarta Int'l Hotel & Develop. .non sig

Trade, Service,

Investment 2.10E+12 273

KIJA Kawasan Industri Jababeka .non sig Property 1.80E+12 231

CTRA Ciputra Development .non sig Property 1.80E+12 261

TKIM Pabrik Kertas Tjiwi Kimia .non sig

Basic Industry

GJTL Gajah Tunggal .non sig Miscellaneous 1.50E+12 242

(41)

40 Appendix 2. New York Stock Exchange Cases

Code Company Correlations Coefficient Industry Market Capitalization (US$Trillion) News Feed Frequency

GS Goldman Sachs 0.991 Finance 68.97 3600

BMY Bristol-Myers 0.93 Consumer Goods 40.21 936

MS Morgan Stanley 0.892 Finance 28.81 8613

HPQ Hewlett Packard 0.875

Trade, Servicee and

Investment 90.06 1576

DUK Duke Energy 0.849 Infrastructure 18.86 1610

USB US Bankcorp 0.846 Finance 30.04 1915

AXP American Express 0.826 Finance 27.12 1515

BAC Bank of America 0.8 Finance 75.97 5293

PEP Pepsi 0.773 Consumer Goods 83.27 1336

SGP Schering-Plough 0.759 Consumer Goods 38.01 830

JPM JP Morgan 0.746 Finance 123.56 8548

GE GE 0.698 Miscellaneous 121.99 1887

XOM Exxon 0.651 Mining 335.92 1136

SLB Schlumberger 0.56 Miscellaneous 64.64 696

LMT Lockheed Martin 0.516 Miscellaneous 31.69 935

IBM IBM 0.51

Trade, Service and

UTX United Technologies 0.506

Trade, Service and

HAL Halliburton 0.5 Miscellaneous 18.03 1509

KMB Kimberly Clark 0.472 Consumer Goods 21.15 812

LLY Eli Lily 0.465 Consumer Goods 38.17 1091

BA Boeing 0.46 Miscellaneous 34.06 1588

MDT Medtronic -0.517 Miscellaneous 37.15 1150

WFC Wells Fargo .non sig. Finance 95.98 4223

VZ Verizon .non sig Infrastructure 89.09 3360

HON Honeywell International .non sig. Miscellaneous 23.49 3032

PFE Pfizer .non sig. Consumer Goods 99.8 2251

MO Altria .non sig. Consumer Goods 33.81 2137

DD EI Dupont .non sig.

Basic Industry and

Chemicals 21.75 1907

BK Bank of New York .non sig. Finance 32.26 1857

CAT Caterpillar .non sig. Miscellaneous 19.47 1849

PG P & G .non sig. Consumer Goods 147.37 1769

JNJ Johnson & Johnson .non. sig Consumer Goods 152.91 1319

HD Home Depot .non sig.

Trade, Service and

MRK Merck .non sig. Consumer Goods 53.1 1276

MCD McDonald's .non sig. Consumer Goods 63.11 1183

CL Colgate .non sig. Consumer Goods 35.24 1104

WMT WalMart .non sig.

Trade, Service and

(42)

41 Investment

ABT Abbott Laboratories .non sig Consumer Goods 72.19 992

DIS Walt Disney .non sig

Trade, Service and

WYE Wyeth .non sig Consumer Goods 59.5 841

CVX Chevron .non sig Mining 131.83 834