Model Behavior Analysis of Stock Market Indicators and Listed Companies : Evidence from the Ghana Stock Exchange : Automated versus Floor Trading

This article studies the model behavior analysis of stock market indicators and listed companies on the Ghana Stock Exchange (GSE) over a five year period. The Ghana Stock Exchange transferred from floor trading to automation in June 2009. The GSE operates an up to date market for recurrently traded securities and an auction call for rarely traded securities. Data for this study was extracted from GSE’s profile of listed companies (fact book) for the period under investigation. The study results show a significant progress on stock market indicators within 2007 to 2011 in dividend yield, volume traded, share price, market capitalization and market returns following the automation in mid 2009, equity premium however, decreased and no significant effects on liquidity were detected. The study further revealed that, the introduction of the electronic trading system has significantly increased the tempo of trading activities on the Ghanaian Stock Market. The study recommends the GSE to accommodate all quoted securities as this will promote and improve fund raising for investors.


Introduction
The development of technology in telecommunications and the internet are modernizing the conventional business model in all spheres of trade.The decade of the 2000s witnessed an incredible increase in the use of fully automated or computerized trading, clearing and settlement of financial instruments on the markets around the world.Generally, the introduction of computerized trading systems provides major advantages such as speed and accuracy of operations, and perhaps most importantly, the ability to acquire an up-to-date information and see real-time state of the companies' financial position for all stakeholders.
The issue of market-trading systems has gained increasing attention in recent years, especially for emerging markets where a need exists to build a financial infrastructure.Advances in technology have led to the development of highly sophisticated computerized trading systems, which can both improve liquidity and reduce trading costs.The technological advance like electronic trading has trickled down from exchanges in developed countries to stock exchanges in emerging markets, where major efforts to improve structure and liquidity have been undertaken.
In recent years technology has increased tremendously in almost all spheres of business, and the automation of Stock Exchanges globally is no exception.The effects of automation recorded on many Stock Exchanges globally cannot be over emphasized (Jain, 2008).Empirical study indicates that, the use of electronic trading systems has grown rapidly due to advancement in technology and share market deregulation, to the extent that the fully automated trading systems' share in securities trading has expanded widely and plays a significant role in determining corporate liquidity (Brockman & Chung, 2000).
Contemporary empirical evidence from Mailafia, (2011) in a study using the key capital market indicators revealed that the performance in capitalization, turnover by volume and value of shares has significantly improved with computerization.Despite the extensive move towards the automation of trading systems, the issue is far from resolved.There is ongoing controversy (Venkataraman, 2001;Madhavan, 2001;Jain, 2005) concerning the relative merits of floor versus electronic trading systems, and to what extent these merits affect market characteristics such as liquidity, equity premium and price efficiency.
Using data from the Ghana Stock Exchange (GSE), this study examines the effects of the adoption of an electronic trading system on market indicators such as dividend yield, volume traded, equity premium, share price change, market capitalization, risk free and market return.In July 2009, the Ghana Stock Exchange was fully automated.The exchange completely computerized its trading system, from order entry to order execution, and traded alongside the traditional trading floor on which brokers exchange securities.Securities listed on the GSE were gradually transferred from the traditional system (quotation on panels) to the new electronic trading system.
The main features of the new trading system adopted by the GSE include: (i) trading automation; (ii) market opening with a call auction based on temporal consolidation of orders; (iii) two trading methods: call 'fixing' and continuous auction systems; and (iv) enhanced transparency.Regarding (i) and (ii), firms transferred to the 'fixing' system represent 90% of our sample and the five best limit selling and buying orders in the order book are public information.
The proponents of floor trading argue that these systems are better than electronic systems.For example, professional relationships may evolve during floor trading because of repeated trading and result in the sharing of information about order flows, thus reducing the level of information asymmetry and increasing liquidity (Venkataraman, 2001;Jain, 2005).Also, the role of human intermediaries such as specialists and brokers in floor trading systems could provide certain benefits for the trading process, through their quick reaction to different market conditions and the execution of sophisticated trading strategies, and thus reduce trading costs and market impact (Venkataraman, 2001).
However, there are arguments based on the anonymity of electronic trading systems which suggest that adverse selection is a more severe problem in electronic trading systems and, therefore, the bid-ask spreads may be higher.Theissen (1999) addressed the issue of transaction costs in floor and computerized trading systems empirically in the German Stock exchange, where floor and screen trading for the same stocks exist in parallel.Both markets were liquid and operate simultaneously for several hours each day.An analysis of the market shares of the competing systems reveals that the electronic trading system was relatively less attractive for less liquid stocks and in the presence of higher return volatility.Also, an analysis of the bid-ask spreads reveals that the electronic trading system was relatively less attractive for less liquid stocks.The market share of the electronic trading system was negatively related to the total trading volume of the stock, was positively related to the difference between spreads on the floor and in the screen trading system and is at least partially negatively related to assess return.They further documented that spreads in the electronic trading system respond more heavily to changes in assess return and that the adverse selection component of the spread was larger.
On the other hand, proponents of electronic trading systems argue that the latter is more efficient and may reduce problems associated with human error.Earlier, Weber, (1999) examined the design of one screen-based futures market on the Cartor Financial Futures Exchange and its capabilities relative to the rival, floor-based market in Chicago and found that electronic "order matching" leads to faster completion of desired trades and about a one-third reduction in transaction cost.Electronic trading systems do offer lower operating costs and the possibility of remote access to the market.In addition, these systems are able to attract new pools of liquidity by providing remote access to investors (Freund & Pagano, 2002;Venkataraman, 2001;Jain, 2005) The Member survey by the International Securities Association for Institutional Trade Communication [ISITC] (2012), confirmed that, many investment firms on both the buying side and selling side are increasing their spending on technology for electronic trading, consequently, removing many floor traders and brokers from the trading process.
These claim and others on the benefits of automation need to be substantiated by worldwide empirical studies.This study seeks to empirically assess the performance of listed companies before and after automation of the Ghana Stock Exchange using key stock market performance indicators.
Hence, the study seeks to answer the following questions: Does electronic trading improve stock market liquidity and performance?Does electronic trading reduce the cost of transactions for listed companies?Is there a positive price reaction when stocks move from floor to electronic trading for listed companies?
To answer these questions, this study contributes to the growing literature that examines the impact of automation on stock exchanges.The main hypothesis tested in this paper is that the automation of the trading process leads to improved liquidity, positive price reaction and a reduction in equity premium which investors require.The study gathers information on listed companies from the GSE and includes all listed companies two years data before automation, during the automation and two years after the automation for consistency and uniformity of the analysis.
The academic literature often focuses on technical questions of trading (e.g.appropriate trading method) without paying attention to the potential influence of automation on certain key indicators for profitability of the exchange (Naidu & Rozeff, 1994).Finally, greater knowledge of the effects of automation on trading provides valuable information for exchanges contemplating changes in trading methods.

Literature Review
A Stock Exchange is a place where securities of companies are traded.Berk & DeMarzo (2008) define Stock Exchanges as organized markets on which the shares of many corporations are traded.Domowitz (1996), explained that an exchange provides ways by which financing is raised by the sale of shares to outside investors.It provides a mechanism for the valuation of companies through the process of price discovery and a means by which such information is disseminated.
The Ghana Stock Exchange (GSE) is the principal stock exchange of Ghana.The GSE was incorporated in July, 1989 as a private company limited by guarantee under Ghana's companies Act, 1963 (Act 179).The Exchange was given recognition as an authorized stock exchange under the Stock Exchange Act of 1971 and commenced trading on November 12, 1990.In April 1994, the Exchange became a public company limited by guarantee.The GSE was a private sector initiative; not funded by government but has enjoyed the support of the Government of Ghana.The objectives of the GSE include; providing the facilities and framework to the public for the purchase and sale of bonds, shares and other securities, coordinating stock dealing activities of members and facilitate the exchange of information including prices of securities listed and also co-operate with associations of stockbrokers and Stock Exchanges in other countries; to obtain and make available to members information and facilities likely to be useful to all market participants.Similar to most stock exchanges around the world, for a company to be listed on the GSE, the listing requirements on capital adequacy, profitability, spread of shares, years of existence and management efficiency has to be met before listing.The GSE until October 2008 used the manual system of trading which involved the use of floor clerks and other floor traders who negotiate among themselves to buy or sell securities that is on offer from various brokers and brokerage firms.
The floor trading is essentially a system where traders and brokers meet at a specific venue referred to as a floor trading or pit to buy and sell financial instruments using open outcry method to communicate with each other.These venues are typically stock exchanges where transactions are executed by members of such an exchange using specific language or hand signals.Open outcry allows for floor traders to understand the emotions of the other traders on the floor when orders are being called in.This intangible information is essential for many of the traders on the floor, as being able to see a trader's greed or fear offers much more than watching a chart on the computer.Many traders will key off the "noise" in the pits to determine the volatility in the markets at a specific price point.Comparatively, there are a few services on the web that offer this live feed from the pits.One key drawback to the open outcry system or any floor based trading system is that traders are not privy to the limit order book which will give the trader an insight into the depth of the market place.This is especially useful during periods of low volatility when pit noise is not useful.
During the 1980s and 1990s electronic trading replaced physical floor and phone trading in most exchanges around the world.Historically, stock markets were physical locations where buyers and sellers met and negotiated.With the improvement in communications technology in the late 20th century, the need for a physical location became less important, as traders could transact from remote locations (Bowley, 2011).Electronic trading, sometimes called e-trading, is a method of trading securities (such as stocks, and bonds), foreign exchange or financial derivatives electronically.Information technology is used to bring together buyers and sellers through electronic trading platform and networks to create virtual market places such as National Association of Securities Dealers Automated Quotation Systems (NASDAQ), New York Stock Exchange (NYSE) Arca and Globex which are also known as Electronic Communications Networks (ECNs).
Several stock exchange floors, on which brokers manually matched orders using an open-cry system, have been replaced by or are being converted to electronic trading systems.In these markets, information technology is also being adopted to globalize trading and settlement activities, enabling investors to trade in different markets regardless of time and location.Amihud and Mendelson (1989) established in their study that market liquidity could be enhanced through the proper use of information technology.
The increase of electronic trading in contrast to floor and phone trading has had some important implications including: Reduced cost of transactions, Greater competition, increased transparency, Tighter spreads, Improve liquidity and significant increase in the amount of publicly available information in the secondary markets.
Empirically, Automation of capital markets has also been found to have instilled or enhanced transparency in market activities to an extent, with respect to price and trade information (Picot, Bortenlaeger & Roehrl, 1997).Brailsford et al. (1999) found that the automation of the stock market resulted in an increase of around 77% in contemporaneous public information transfer.With the entire bid-and-ask schedule and course of sales available on an almost real-time basis through e-trading directly to all market participants, investors do not have to rely on brokers for this information.On the part of brokers, they do not have to undertake a physical search for the appropriate party to fulfill a sell or buy order raised by their clients in these markets.
The types of trading systems are sometimes differentiated by the form of market intermediation provided by entities with direct access to the system.The nature of competition between exchanges is a defining feature, since exchanges may adopt varying market structures in order to compete in different fashions, for instance, according to Zhong (2002), the China Securities Regulation Commission's (CSRS) implementation of new rules in 2002 was to lower the price cap on brokerage fees, introduce price competition among securities brokerage firms thus prompting brokerage services to adopt online trading as a mean of lowering transaction or operation costs in the face keen market competition.
According to Brailsford, T. J., Frino, A., Hodgson, A., and West, A., (1999), automation of equities trading has a three-fold effect on the market including: The transparency of the market with respect to price and trade information improves radically with participants being able view the entire bid and ask schedule and course of sales in real time; Reporting lags and errors reduced due to instantaneous dissemination of all market activity via electronic signals thereby producing a cleaner price feed for the market; and Faster trade execution is achieved for large portfolios of shares as parties to transactions no longer had to physically search each other out.Biais, Glosten and Spatt (2005), in their study noted that automation can lead liquidity to decrease because it does not allow the direct negotiation between traders for transactions and does not allow them to preserve a certain control on trading conditions.Contrarily, an argument raised in a study by Pirrong (1996) has shown that automated stock exchanges can be deeper and more liquid than manual stock exchanges.
In this regard, Naidu and Rozeff (1994) also observed an increase of volatility and liquidity as well as an improvement in efficiency following the automation of the Singapore stock exchange.They observed that automation speeds up the dissemination of prices, forcing volatility to increase, as well as potentially altering trading volumes especially when information is hitting the market.Moreso, Sioud and Hmaied (2008), in their work established an increase in trading volume after the transfer of shares to the new trading system.Conversely, they found that the new trading mechanism does not reduce pricing error and hence does not improve market efficiency.Jain (2008) however concluded that automation appears to reduce transaction cost and make emerging markets more accessible to foreign traders hence increasing the liquidity and market capitalization of the firms and the economy.
Empirical and theoretical researches have also found the effects of the trading mechanism on market characteristics.Schwartz and Steil (1991) suggests that call auctions permitting the determination of a single price for all transactions permit investors to post limit orders, which improves market liquidity.Mendelson & Tunca (2004) shows that increasing the number of participants in an auction increases price precision.However, he suggests that beyond some minimal number of traders the benefits of concentrating trade in an auction are practically exhausted and a second stage of continuous trading can be employed to further increase traders' opportunities.Madhavan (2001) established that in the presence of asymmetric information, continuous markets fail.Nevertheless, call-auction markets continue to function because of the averaging effect of all traders' prices rather than bilateral trading.Pagano and Roell (1996) compared liquidity and the price-formation processes in several trading systems with different degrees of transparency.They suggest that a greater transparency in the trading process improves market liquidity by increasing opportunities for less-informed traders to participate in a system with reduced spreads, volatility and pricing errors.Nevertheless, investors with private information tend to prefer less-transparent systems to take advantage of their situation.
The adoption of electronic trading has become one of the major changes that have occurred in the market design and structure in recent years.The change (moving from floor to electronic trading system) is an important issue in market microstructure, and deserves thorough investigation for several reasons.For investors, financial services on the web offer great crucial benefits such as reduced cost of transactions for all concerned as well as the ease and the convenience in trading.

The Stock Market Description
For a long time, insufficient financing retarded economic growth in Ghana.The production sectors could not meet the developing needs of firms facing severe international opposition due to market liberalization.Since 2001, the Ghanaian government implemented several reforms to stimulate sectors like the financial market and favor the creation of new sources of financing.
As a result of these reforms, the Ghanaian stock exchange increased its order flow.However, due to the inadequate financial information and the dearth of enhanced market activity, the increased demand has not been accompanied by a similar increase in supply.As a result, stock prices increased steadily, sustained by the dominating group behavior of small investors.For five successive years, the Ghanaian stock market displayed a rise in the GSE index from 18 in 2007 to 34 in 2011.By the end of 2009, investors realized that stock prices did not reflect true intrinsic values and they became reluctant to buy at the offered prices, which ensured a progressive fall in stock demand.
Fearing a fast downfall of the stock market, the authorities decided to adopt corrective measures.From June 2009 until May 2010, minimum transaction amounts were required to allow a stock price change.Unfortunately, this proposed remedy did not work out as expected.To address the crisis, the GSE decided to eliminate the traditional trading floor on which brokers exchange securities and introduced an automated trading system that was expected to enhance market liquidity and efficiency.Securities listed on the GSE were gradually transferred from the traditional system to the new electronic trading system.

The E-Trading System
From 2001 until June 2009, a system based on an electronic compromise gradually replaced the physical quotation on panels.A replication point was initiated from March to June 2010 to enable stock operators to be familiar with the new trading system.From July to September 2010, the required infrastructure was set up.
Since the transfer, in 2011, the Ghana Stock Exchange implemented some major changes in its trading activities.To complement the automated trading regime, the Exchange extended its trading hours to afford dealers increased contact hours with their clients during the trading day and also to afford non-resident investors in time zones different from Ghana, greater opportunity to reach out to their local brokers.The new trading hours become 09.30hours GMT to 15.00 hours GMT from the existing 09.30hoursGMT to 13.00hours GMT.This was expected to also help improve liquidity in the market place.
The Ghana Stock Exchange (GSE) also introduced a new method of calculating closing prices of equities on the market.Closing prices of listed equities from January 4, 2011 were calculated using the volume weighted average price of each equity for every given trading day.Hitherto, closing price was based on the last transaction price of listed equities.Two new indices were introduced on January 4, 2011 to replace the GSE All-Share Index which tracks price changes in the listed equities.The new indices were the GSE Composite Index (GSE-CI) and the GSE Financial Stocks Index (GSE-FSI).
The GSE operates a continuous auction market for frequently traded securities and a call auction for infrequently traded securities.During the pre-opening point, buy and sell orders are accumulated in the central order book without any transactions taking place and a price is displayed systematically.This price must maximize the number of stocks traded.If this price is not unique, the system chooses the one that minimizes the number of securities not served.
Despite the difficult year the Exchange still had crossed a watershed in 2009.The Automated Trading System (ATS) and the Electronic Clearing/Settlement both went live to complement the Depository system which went live in November 2008.With that move all the operations are fully automated now.Dealers now have access to trading from the Exchange's Trading Floor; the offices of Dealers; and through a secured internet facility at any location.Bi-lateral settlement between brokers ended with the introduction of Electronic Clearing and Settlement.Therefore trades are settled electronically on T+3 and the underlying securities also credited to the Depository accounts of buying investors.The GSE Securities Depository Company which became fully operational in November 2008 had 34,000 depository accounts opened at the end of 2009.Through resolutions and public education at the various AGMs, the Exchange's listed companies have amended their company regulations not to issue any paper share certificates.Indeed, under the automated environment, an investor (whether buying or selling) must have price when  Vol. 9, No. 11;

The Da
e year 2008 gr this also motiv e return in the y ture, investmen trading system e returns on eq the results wit ).
ver time, but sh .

Marke
For this st time as a p all assets a The marke 24million.
13.7millio investors r return in fi simply me of the mar market is i GSE.

Risk Fr
Risk-free r that the ris absolutely higher the The figure free rate a borrowing automation companies strongly hold the belief that raising capital and listing on the stock market is a sure way of success for companies operating in Ghana.
In 2008, UT Financial Services, SIC Insurance Co. Ltd and Golden Star resources were listed on the Exchange having successfully completed the offer for sale to the general public.The Bond market however did not record much activity.The value of listed Government of Ghana Bonds on the Exchange amounted to GH¢1,509.60 million while that for corporate bonds was GH¢6.40 million and GH¢35.00 for SCB Medium Term Notes showing a strong risk free exchange both before and after automation.

Conclusions
This paper analyzed the model behaviour of stock market variables on the GSE.The study extends previous research on changing trading mechanisms by focusing on the automation of an emerging market.Our findings provide support for the results of Theissen (2002) on the German Stock Market, and Hendershott and Moulton (2011) on the New York Stock Exchange.The introduction of the electronic trading system has increased the tempo of trading activities on the Ghanaian Stock Market.Automation have increased substantially the indices, market turnover, transparency, investor's confidence, foreign investments, liquidity, vibrancy of the market and prompt inter-broker money and stock settlement.

Recommendations
This study makes the following recommendations so that the Ghanaian Stock Market can enjoy fully the dividends of automation.


The Ghana Stock Exchange should accommodate all quoted securities.For now only equities are traded through the GSE: other securities such as government stocks, industrial loans, bonds and preference stocks are yet to be accommodated in the current automation process.


The transaction cycle should use of a single settlement automation bank as opposed to several settlement banks being used now.


The trading hours should be extended beyond the present closing time of 5:00 p.m. for the normal government working hours.


There should be a separation of functions between jobbers and dealers and finally the GSE should float new development stocks for capital projects instead of other forms of borrowing.


These recommendations should further advance the automation process thereby enhancing performance of listed companies on the Ghanaian Stock Exchange. Figur Figur Figur