Share Split and Market Performance - Case of Polish Stock Exchange

Table of Contents

Abstract

Introduction

1. Share splits hypothesis
1.1 Hypothesis about signaling good perspectives
1.2 Hypothesis of attracting investors
1.3 Hypothesis of optimal price range
1.4 Hypothesis of increase in liquidity
1.5 Hypothesis of split as a marketing tool

2. Analysis of share splits in Poland

3. Methodology
3.1 Zero order models
3.2 First order models
3.3 Shifts and Pulses

4. Share splits on Warsaw Stock Exchange – estimation results

5. Conclusions

References

APPENDIX

Abstract

A purpose of this paper is to analyze an influence of shares splits on share performance and on investor’s behavior. To identify changes in the security prices and the volume of trade, an intervention analysis and modified CAPM are applied. To conclude on behavior of Polish investors the results of estimation are compared with theoretical hypothesis. Paradoxically, Polish stock exchange proves to be an effective market. Investor’s reaction on share splits was in generally indifferent or, unexpectedly, negative. There is no manipulation of investor’s mood, what is common on the other more developed markets. Despite share splits are very rare they are a neutral instrument of price correction.

Introduction

The purpose shares split is to divide nominal value in given proportion eg. 2:1, 5:4 etc, what hat involves the proportional reduction of the price Splits are conducted on shares, both traded public and in private turnover. From the rational point of view split should have no other consequences for investors and for share price and should be treated as a cosmetic correction. For shareholders, split changes a nominal amount of shares, but it does not change the value of portfolio. Shareholders rights, Cash Flow and equity of company stay unchanged.

In the history of the world stock exchanges the first shares splits took place in 1927. The earliest research is dated 1933 (Doodley 1933), but the top of popularity splits reached recently, since the ’90. Easley, O”Hara, Saar (2001) report that the most of splits took place in 1997 on NYSE, when about 235 companies made a split. On average about 10% of firms decide each year to lower the share price. Some of companies made a split for many times as e.g. Johnson & Johnson (6 splits since 1970), Bouygues (12 splits since 1979)^[1] etc.

The main question of researchers, investors and companies is ” why to make a split?”. Still the explicit motive of split is unknown. There are many hypothesis and case studies in the literature. All appearing hypothesis are clustered around some periods of time. In history of splits there are three main groups to be distinguished:

- Since ’30 up to ’70 – from splits appearance to the market transformation during fuel crisis
- Since ‘70 up to 1998 – form the development of modern financial instruments to the crisis of new technology companies
- Since 1998 up to now and more

Given above classification concerns mainly U.S. market, the most developed one in the world. But one should remember, that this market serves as a indicator of investments mood. It gives a direction of development and influences trends all over the world. So, everything what happens on the U.S. market is also about the rest of the world, but with some time lag.

1. Share splits hypothesis

Researchers did not established explicit one main reason of conducting share splits although in world literature there are many hypotheses. Some of the most surveyed hypotheses are presented below.

1.1 Hypothesis about signaling good perspectives

In the recent years, one of the well-known hypotheses binds split and the process of signaling good perspectives (Fernando, Krishnamurthy, Spindt, 1999). When a company lowers its security prices, it can mean that the growth in share price should be expected. It is a way to give investors some good news (Desai, Jain, 1997), which are often private and even confident (Brennan, Copeland, 1988). This information may concern the expected abnormal profits (Copeland, 1979). In some research such correlation between abnormal returns and splits was proved (Fama, Fisher, Jensen, Roll (1969), Asquith, Healy, Palepu (1989)). Announcement of good sights is not idle. Ikenberry, Rankine, Stice (1996) proved, that companies making split were better off at about 7% in one year after split and 12% in three years after split than companies without split. That all means that it is not a split that increases stock price, but is serves as a information instrument.

In the literature there are also some research, which shake this theory. Fama, Fisher, Jensen, Roll (1969) presented the results of 940 splits in 1927-1959 on base of U.S. monthly data. This survey showed a correlation between split and abnormal returns. The point is the sequence it appeared. There were abnormal returns of about 34% in the period of 30 months before split, but there were no abnormal returns after the split. Similar results obtained Lakonishok, Lev (1987). On base of daily data for 1015 splits in 1963-1982 they proved that for 5 years before split abnormal profits were about 53%, but they disappeared after split. It may be interpreted, that splits are executed after good times and have no influence on share price. On this base there were made another hypothesis about an optimal price range. Abnormal profits causes an increase in security price. Split is an instrument to reduce price to its optimal level. This is also called benchmark price hypothesis.

1.2 Hypothesis of attracting investors

A hypothesis of attracting investors is of great caution in the recent surveys. In Easley, O’Hara, Saar (2001) split is made to attract two kinds of investors. First, the noisy one, taking their decisions according to other market players and under impulses and second, informed investors, who invest with deliberation. As Black (1986), Baker (1956) and Stovall (1995) show, price reduction attracts small investors, who can afford buying cheaper shares. They can built smaller portfolio and better diversify risk then. It is impossible to buy a half of share, so cheaper shares give more investment possibilities. It has also a taste of sale, what encourages to buying. Investors take their decisions in two ways: before split, on base of expected low price, and after the split, while observing prices. In the first case, investors are well-informed and understand the issue of split. In the second case, players are small, an accidental one. The proof of this hypothesis give Desai, Nimalendran, Venkataraman (1998), who pointed out, that after split transactions are more often, but smaller. Attracting small investors gives some benefits to both market and company. Small investors do not seek confident information, are not as nervous as others and react slower for any market anomalies, what makes market more stable. In firms, small investors have not enough power to control a board and to execute their rights. After all, splits have magic that attracts gamblers. In Easley, O’Hara, Saar (2001), Desai, Nimalendran, Venkataraman (1998), Ohlson, Penman (1985) or Dubofsky (1991) splits causes an increase in shares volatility. This means it is an ideal situation for day traders, who love great volatility and fast market movements.

1.3 Hypothesis of optimal price range

There were also undertook some surveys on optimal price range (Angel (1997), Baker, Kent, Gallagher (1980), Baker, Kent, Powell (1993), Fernando, Krishnamurthy, Spindt (1999)). There are different reasons to keep prices on the given, fixed level. Baker, Kent, Powell (1993) argues that thanks to price correction investors can buy shares cheaper, what increases and diversifies players. Another reason for split is price correction to the optimal level, what follows the period of abnormal profits. This was proved by Fama, Fisher, Jensen, Roll (1969), Lakonishok, Lev (1987), Fernando, Krishnamurthy, Spindt (1999), Muscarella, Vetsuypens (1996), Conroy, Harris (1999). This means that shares have their benchmark prices and firms use different methods to achieve this level. Split can also serve as a counterweight for shares buyback. After the increase in price, due to buyback, split is an easy instrument to come back to optimal price level (Fama, French (2000), Asquith, Healy, Palepu (1989)). Other reason for keeping price on benchmark level is to increase shares liquidity (Fama, French (2000)). Unfortunately, because of great amount of liquidity measures, there are no explicit results.

An interesting reason for split is to correct price before IPO (Initial Public Offering) or seasoned offering. On U.S. market underwriters prefer to underwrite shares from price range $10-$20 (Fernando, Krishnamurthy, Spindt (1999)). Lower price is a signal for low quality of shares, but a higher price suggests prestige of shares. Split helps to be a common, not prestige company, in order to attract small investors.

1.4 Hypothesis of increase in liquidity

Hypothesis of increase in liquidity was examined by Baker, Kent, Gallagher (1980), Baker, Kent, Powell (1993), Dolly (1933). The increased liquidity of shares may be obtained through rise in the amount of investors^[2], diversification and lower price of shares. There are many measures of liquidity like: bid/ask spread^[3], trading volume^[4], volatility^[5], amount of transactions^[6] etc. Michayluk, Kofman (2001) pointed, that this measures are not adjusted to the situation of split. They argue, that split causes decrease in liquidity. As Michayluk, Kofman (2001) show, in 1997 there was a structural change on the U.S. market. It resulted from development of new economy firms and many IPOs in high-tech industry. In Poland there was no structural shift, so the results from U.S. market are not transferable.

1.5 Hypothesis of split as a marketing tool

An analysis of splits on the developed markets shows, that in the recent years companies make their share splits in smaller proportion than before. As a example, there are splits of Novartis (on 6th November 2000 in ratio 81:86 on Xetra), Preussag (on 20th April 1999 in ratio 12:17 in Frankfurt), Bouygues (on 09th March 2000 in ratio 81:81 in Paris), Diageo (on 02nd February 1998 in ratio 85:83 in London), AstraZeneca (on 13th November 2000 in ration 12:18 in Frankfurt), Alstom (on 04th June 2002 in ratio 61:74 in Paris) and many more. Such low ratio of split does not change the price significantly, what stays in opposition to fundamental assumption of the process. In those cases the goal of split is not to correct a share price. When investors treat a split as a signal of good news, other companies can use psychological effect of split to remind about themselves without any important changes. Many internet sites on split are very helpful.

Marketing effect of split can be confirmed in survey by Baker, Kent, Powell (1993). They found out that companies with cheaper shares are more intensively examined. There are more rapports and analysis of that firms and investors pay more attention to them. It is very important on the large stock exchanges, with hundreds or thousands of companies. The effect is that after split cheaper shares are more analyzed.

* * *

There are some more general conclusions. The character of splits, its application, results and reception by investors depends on moment of time they were made. Stock exchanges in each of three periods mentioned above can be characterized by other structure of shares and companies due to their quality, price, industry etc. This determines different investors behavior. In each of these periods different factors where the motives for splits. In the first period, share splits confirm the hypothesis of benchmark price. Security price was kept in the proper range because of splits. Split was not a signal for good performance, but it was an ex-post tool for price correction. Together with the market development in the second period, share split was used as an instrument for signaling of good perspectives and abnormal profits. This changed an investors perception of the share splits. Splits announcement were treated as strong buy signal and were as precious as analyst’s recommendation. Good opinion about splits and conviction of profitable firms perspectives were a key to investors consciousness. Both good and weak companies made many splits. But a misuse of investors trust, especially by weak firms, made the splits reputation worse. Many internet sites on splits made it fashionable to divide shares. “To be among the best” was a new habit of companies. This was also relatively cheap way to remind investors about company. Beside of informing, split role was to advertise. Those above one can sum up in few words: to attract investors at all costs, with advertisements and promises of abnormal profits. Now, market begins third phase. It is extremely difficult to asses the new role of splits.

[...]

^[1] Data from the companies sites

^[2] in Easley, O’Hara, Saar (2001) greater amount of investors increases the liquidity, but because of many measures of liquidity this is not explicit

^[3] In Copeland (1979), Conroy, Harris, Benet (1990), Desai, Nimaldendran, Venkataraman (1998) bid/ask spread rises, but in Murray (1985) stays without changes.

^[4] In Copeland (1979), Conroy, Harris, Benet (1990), Lamoureus, Poon (1987) split decreases trading volume, but in Murray (1985) it stays without changes

^[5] Volatility rises in Ohlson, Panneman (1985), Dravid (1987), Lamoureux, Poon (1987), Conroz, Harris, Benet (1990), Dubofsky (1991), Desai, Nimaldendran, Venkataraman (1998), Koski (1998)

^[6] An amount of transakction increases in Muscarella, Vetsuzpens (1996), Krzyzanowski, Zhang (1996), Desai, Nimaldendran, Venkataraman (1998),