On-Line Trading at the Singapore Hog Auction Market (HAM)

Neo Boon Siong, Nanyang Technological University, Singapore

and

Roger Clarke

Principal, Xamax Consultancy Pty Ltd, Canberra

Visiting Fellow, Department of Computer Science, Australian National University

Version of 25 July 1992

© Xamax Consultancy Pty Ltd, 1992

This document is at http://www.anu.edu.au/people/Roger.Clarke/EC/HAM.html


Abstract

Singapore's Hog Auction Market (HAM) has been operational since March 1990. It is a purpose-designed auction hall with an integrated on-line bidding system which enables the auction of over 2000 pigs per day. It is of especial interest for at least two reasons. One is its speed of operation, and the other the strategic purposes for which it was developed.

This case report is based on investigations undertaken by the authors in the period 1990-92, including site visits, observation of the market's operation, discussions with senior staff involved in its design, development, implementation, operation and regulation, and inspection of publicly available and relevant internal documents. This paper is intended to provide a basis for the preparation of a teaching case and to support case-based research into on-line trading systems and strategic information systems.


Contents

Background

Pigs in Singapore

Pig Trading in Singapore

Implementation

Form of the Market

Relevant Outcomes

Evaluation

Further Developments

Reference List


Background

Singapore is a small, disciplined and highly dynamic island state joined by bridge with the southern extremity of the Malay peninsula, and by sea and air with the world. Since gaining its independence in 1965, it has made enviable progress in improving its standard of living, its level of education and its industrial and social infrastructure. It has long been a major sea-port, and has vigorously pursued a strategy to maintain its role as the regional sea-trading hub, with the result that cargo passing through its port is three times its GDP. It has had some success in extending its hub-strategy into air-cargo, air-passenger travel and communications, and has consequently attracted many corporations to establish their Asian headquarters there. Singapore's 2.5 million population comprises a stable mix of people of 75% Chinese, 10% Malay and 10% Indian racial origins. The predominant language is English, but Mandarin, various Chinese dialects, Bahasa Malay? and Sanskrit? are also used.

Singapore is a small island, and long ago adopted the policy that it would not seek self-sufficiency, but rather specialise in areas in which it had economic advantage. In its early years it could offer a relatively cheap labour-force, but the price of success has been the disappearance of that advantage. During the last decade it has focussed on upgrading the educational standard of its population. One facet of this managed transition was to phase out agriculture and pastoral industries on the island, and move to full dependence on imported food. The adjacent Malaysian mainland is the primary source, but nearby islands of Indonesia and other South-East Asian countries also deliver produce to the island.


Pigs in Singapore

This most popular meat among Chinese Singaporeans is fresh pork. Each year, about 1.2 million pigs are imported live and slaughtered in the two abattoirs at Jurong (which handles all of the cattle and sheep slaughter and 60% of the pig slaughter) and at Kim Chuan (which handles the remainder of the pig slaughter). Both are operated by a government business enterprise, Primary Industries (Pte) Ltd, referred to in this case report as PIPL. PIPL is one of the operating subsidiaries of a holding company owned 100% by the Singaporean Ministry of National Development (MND).

The vast majority of pork is sold fresh, both directly to consumers in Singapore's many markets and through the very large restaurant and food centre trade. Due to the year-round high temperature and humidity, fresh foodstuffs have a very short shelf-life, and speed and efficiency in distribution are therefore critical. The health aspects of meat products are regulated by Singapore's Primary Production Department (PPD), and in particular its Animal Health Division.

There are many factors which determine the quality of pigs, including the breed, the source, the apparent lean and fat, and the size of the ham. Rather than sophisticated scientific measures, judgements about quality are based on buyers' experience. The most important market segments are:

Singapore's Muslim minority does not eat pork, but among the Chinese (non-Islamic) majority there is a very strong cultural bias towards pork which limits the impact of substitute products. Demand is fairly consistent throughout the year - being almost on the equator Singapore experiences no seasons other than the wet monsoon period. The abattoirs and hog markets operate six days per week (except Sunday) every week of the year, including all public holidays except the first three days of Chinese New Year. There are weekly peaks on Mondays and Fridays, each about 30% above the other days.

In the second quarter of 1992, 85% of the pigs consumed in Singapore were imported by truck from farms in Malaysia, some within 1-2 hours' drive in the adjacent Johor State, and others as far as 12-24 hours' drive away in Penang. The remainder came from nearby islands of Indonesia, or from Thailand. The PPD approves individual farms as suppliers, based on inspection against a set of criteria, among which hygiene plays an important role. PPD sets an annual export quota for each farm, based largely on the number of sows. An import permit costing $S150 is required for each truckload consigned to Singapore. Each consignment must be accompanied by health documents. Generally consignments are timed to arrive early each morning at the border, where documents are inspected, and off-loaded at the abattoirs, where the animals are inspected. The hogs are mostly sold and slaughtered the same day.


Pig Trading in Singapore

During the 1970s and 1980s, virtually all of the live pig trade was controlled by a group of about 20 importers, some Malaysian but most of them Singaporean enterprises and companies. The farmers exported the animals on consignment and without a price commitment. Property passed on delivery to the ultimate buyer, with deaths in transit being at the farmer's cost. The importer graded the animals, and sold them within three days. Malaysia precludes re-import of the animals back into that country, so return to the farm is not possible. Under a long-standing arrangement, the importers set the prices for the week in a syndicate meeting each Monday morning. Buyers could accept a load at one price, or select their own animals for a premium price. A model of the sector is presented at Exhibit 1.

[INSERT Exhibit 1: The Pig Industry Sector ABOUT HERE]

In 1982, PIPL investigated electronic trading systems used in several countries for pig-auctioning, particularly in Taiwan, some of whose large provincial livestock markets converted to electronic trading as early as 1979. It concluded that although the idea had promise, it would be premature to introduce such a system in Singapore. One reason was that the technology was not yet sufficiently mature. Another was because the domestic pig-industry was shortly to be phased out in favour of imports. This plan was implemented in 1988-89, with most remaining farmers quickly accepting the Government's compensation offer.

Taiwan was re-visited in 1988, and the features of the systems examined. PIPL decided to have an on-line auction system developed, with the objective of establishing a fair basis for the setting of pig-prices. This necessarily involved a significant reduction in the power of the importers. With the assistance of a firm of consultants, Price Waterhouse, a tender document was issued to three selected companies, a supplier of a system in Taiwan, a computer supplier (Hewlett Packard) and a subsidiary of the State-owned company Singapore Technologies, Singapore Electronic & Engineering Ltd (SEEL). SEEL was established as a government-owned company in 1969. It has expanded to about 700 staff, and since 1991 30% of the shares have been owned by shareholders other than the Government. SEEL won the tender, and in late 1988 entered into a fixed-price contract for the supply of an on-line system capable of auctioning up to 50% of the pigs sold in Singapore each year.

The tender documents were written at the level of a broad functional specification, with little operational detail. SEEL is a systems integrator, and commonly contracts on such a basis. Mr Chia King Young was appointed as Project Manager, with three teams of 2-4 people responsible for the development of the Auction Processing System (APS), the backend Management Information System (MIS) and the hardware. For its part, PIPL formed a subsidiary, Hog Auction Market Pte Ltd (HAM), and appointed Mr Paul Lu as Manager.

The specifications for the system were developed jointly by SEEL and HAM during the first half of 1989. During the second half of that year, the system was constructed by SEEL and a purpose-designed hall built adjacent to the Jurong abattoirs. The system was implemented in March 1990. Such very tight schedules are said to be normal in Singapore. The total cost including the building was reported to be about $S5 million, of which the auction system represented about $S1.2 million.


Implementation

HAM was introduced on 1 March 1990. PIPL announced that from mid-May only pigs bought through the HAM system would be slaughtered at the Jurong abattoirs. Considerable resistance was encountered, and during the parallel run pork merchants largely avoided its use, with pigs bought through HAM accounting for only 12% of Jurong abattoirs' throughput, or 7% of total pig sales. In April, the Malaysian pig-farmers made an appeal to the Singapore Cabinet Minister responsible for PIPL to review the implementation plans for HAM. On the day of cutover, Malaysian pig-farmers stopped pig supplies to Singapore for three days. Because 70% of supplies were coming from Malaysia at the time, the boycott caused havoc in the pork market and attracted much public attention. Pork prices jumped and many butchers' stalls were closed.

The pig importers and pork merchants' association publicly accused PIPL for the confusion in the market, and PIPL countered with press statements and conferences. The pig importers and pork merchants obtained an interim injunction preventing HAM's implementation. Further court action resulted in the injunction being first lifted and then restored. It appeared that the HAM system had been stopped in its tracks.

The Singapore Government was sympathetic to PIPL's endeavours to introduce HAM, and improve the fairness of the price-setting mechanism. With the injunction in place preventing HAM's operation, the Ministry of Trade and Industry provided active support through a 'price control order', under which PPD imposed a licensing requirement on importers. A key licensing requirement stipulated that an importer must sell a set minimum proportion of their pigs through an approved auction market. The proportion was set initially at 20%, but was to be progressively increased as importers and merchants became used to it. HAM was the only auction market which was approved, and although there was nothing to preclude the development of others, it was unlikely due to the cost and expertise necessary and the lead-time involved.

About three weeks after the new licence ruling came into force, the pig importers and pork merchants' association accepted the inevitable and withdrew their injunction. HAM quickly became accepted as the primary means of buying and selling pigs in Singapore. There is little evidence that buyers experienced any great difficulties in adapting to the HAM system's procedures; indeed, given that they subsequently requested that the system be speeded up, if anything the reverse may be true.


Form of the Market
* Hall Layout

The hall comprises a stage for running the pigs, and a steeply raked auditorium from which the bidders can view them. The stage also contains the auctioneer, the weighing station and its operator, and several employees who move the pigs along the run and stamp the Id of the winning Bidder on the animal. See Exhibit 2.

[INSERT Exhibit 2: The Hall Layout ABOUT HERE]

Above the stage, the Auction Display Panel shows the Hog Id and the animal's Weight, the Price currently bid and the Bidder Id of the current Bidder, and lamps indicating the state and the outcome of the transaction. The Price and the Bidder Id change during the course of the auction. See Exhibit 3.

[INSERT Exhibit 3: The Auction Display Panel ABOUT HERE]

Each seat in the auditorium has a Bidder's Station, which is a small panel attached to the right-hand arm-rest, containing an enable/disable switch, a lamp to indicate when the bidding position is enabled, and a bidding-button. Although the hall contains 187 bidding positions, there are commonly about 35-40 bidders at any one time, and a maximum of 100. See Exhibit 4.

[INSERT Exhibit 4: The Bidder's Station ABOUT HERE]

* Hog Logistics

A day's business comprises of the order of 30-40 consignments. Some consignments are truckloads of 85-100 animals, while others contain only between 15 and 50 animals. Prior to entering the run, the pigs are off-loaded from the delivery truck, put in holding pens, and sprayed with water. Apart from cleaning them, this is also intended to relieve the stress on the animals, which is a potential cause of death and lowers meat quality. They are inspected, and if unfit for consumption are condemned.

The sequence in which the consignments are put before the market is set each day by ballot. [This seems not to be borne out by the Auctioning Sequence Sheet for 13 July because the distributions of apparent importer lots vs. direct farmer-delivered lots, and of lot-size appear to be non-random. This needs checking]. Within each consignment, the sequence in which the animals are paraded is random.

The run comprises a series of sections separated by gates. Each pig has its Hog Id number painted clearly on its back, and is pushed into the first section, the Holding Bay. It is then released into the second bay, which is the Weighing Station. Because there are difficulties in weighing a moving animal, the computer integrated within the station uses an algorithm to compute the weight. The machine is calibrated annually by the Government's Weights and Scales Department to be accurate to within 0.1kg. The weighing station operator reviews and approves the computed weight, which is then displayed to the nearest 0.5kg, and used as the basis of the contract of sale.

From the Weighing Station, the animal is released into a run long enough for the bidders to observe the beast's breed, fat and musculature. The railings are sufficiently narrow that the bidders gain a clear view of the animal, without any risk of the beast escaping into the hall. Two further pig-length sections ensure an orderly exit of the animal out of the run. A Stamper's Panel beside the last bay displays the Hog Id and Buyer Id, enabling an employee to stamp the Buyer Id onto the beast's hide. The animal is returned to the holding pens. With few exceptions, beasts are slaughtered the same evening for carcase delivery the following morning.

The environment is very noisy due to the squeals of the pigs, which require considerable encouragement to go forward and not to turn back.

* Bidder Logistics

On arrival, each bidder proceeds to the Seat Allocation Office. There he views a screen which displays the auditorium seating including the positions already occupied, and selects a seat. The bidder also deposits funds with the Cashier, unless he has credit arrangements already in place and a sufficient credit-balance remaining. He then proceeds to his seat and enables the position by pressing the key in the bidding panel.

A bidder may disable his Bidder's Station at any time during the auction, but can only have it re-enabled (or another station assigned) by again registering with the Seat Allocation Office. The Station may also be disabled under program control, when, due to a purchase, the amount left on deposit falls below the set minimum. In this case, an indicator on the display board lights up at the same time as the lot is knocked down.

* Bidding Procedure

Once the animal has been weighed and has moved along the run, the auctioneer sets a starting price, generally about 5-15% below the likely sale price. The Bidding Display Panel shows the Hog Id, the Weight, and the starting Price.

Bidders press their bidding button to register their interest. If there is a single bidder, then the starting price is confirmed. If there are two or more bidders, the price proceeds upwards until only one button is still depressed. As soon as there is only one bidder in the market, the count-down commences, and the Awarded-lamp blinks three times. At the third blink, the lot is knocked down, the Awarded lamp remains lit and the Id of the successful bidder is displayed. The auctioneer announces the Price and Buyer Id using the microphone at his workstation. If the sale has reduced the successful bidder's deposit below the minimum requirement, the 'Deposit Limit' lamp also lights up and the Bidding Station is disabled under program control.

If there is no bidder active at the time the auction commences [within a pre-set period after the auction commences?], the price falls. If a bid is made, the normal logic re-commences from the new starting-point. If no bid is made when the reserve-price, if any, or zero is reached, the lot is passed in, and the Abort indicator lights up. This is unusual.

The speed of the auctioning process is extremely high. During initial operations it functioned at 100 hogs per hour, but, under pressure from the bidders, it has been progressively increased to 400 hogs per hour, which is equivalent to 9 seconds per lot. This achieved using the parameter settings in Exhibit 5.

Exhibit 5: Auction Parameter Settings

*	?period before the auction commences?	?0.3 seconds?
*	price-increment	$0.02
*	price-increment delay-time	0.2 seconds
*	price-decrement delay-time	0.2 seconds
*	delay before award	0.3 seconds

A typical auction proceeds with about 3-5 seconds of bidding, 3 seconds of result-display, and 1-3 seconds set-up time for the next animal. The settings would appear to challenge the bounds of human capabilities [we need to look up a text on ergonomics!], but are accepted by bidders as the desirable compromise between bidding precision and speed of auction. There are generally four pigs on the stage at one time, one entering the holding bay, one in the weighing station, one in the run, and one being taken out of the stamping bay.

The speed enables the hall to put through the typical 2000 beasts in 5 hours plus occasional delays, and the auction runs from 9am to between 2 and 3pm each day, although sometimes until 4.30pm on peak days, Mondays and Fridays. The primary hall has therefore been able to support the auctioning of most of the pigs available to the Jurong abattoirs, and about 50% of the total volume of pigs sold in Singapore. As a result, the adjacent, identical hall on the Jurong site is seldom used.

* Auction management

The system parameters (see Exhibit 5 above) are set at the beginning of operations, and cannot be changed without re-initiating the system. The auctioneer's capabilities are restricted to, for each beast, selecting the starting price, initiating the auction, cancelling a lot, and, if necessary, manually changing the sequence of Hog Ids. The initiation of the next auction is not possible until the data relating to the previous sale has been updated to both data-servers, which is generally about 1 second after knock-down of the previous animal. The weighing station operator's capabilities are restricted to approving the weight registered or requiring a re-weigh.

Statistics both for each auction and over time are generated for HAM management. Of particular concern is whether any bidder establishes dominance in a market segment. During the second quarter of 1992, the largest few bidders have accounted for at most 7% each of the market in any one day.

* IT Infrastructure

The hardware configuration is depicted in Exhibit 6. It comprises a number of on-line components (the frontend or Auction Process Subsystem - APS) and a backend system (the Management Information Subsystem - MIS).

[INSERT Exhibit 6: The Hardware Configuration ABOUT HERE]

The MIS is an HP3000-922LX midrange machine running a mix of packaged accounting software, additional report programs, and custom-built software, written in Powerhouse over the Image CODASYL DBMS. The MIS and APS are interconnected by a System Administrator's Station (an MS-DOS 3.3 286-based PC-compatible), which is connected to the Ethernet backbone.

The central component of the APS is SEEL's proprietary 'Interative OutStation' (IOS), which is a customisable 68000-based product sold in the security and automation markets. The hardware required expansion, and custom-built programs were written, in C and cross-compiled to 68000 Assembler, to perform the following functions:

Directly connected to the IOS are:

Connected to the IOS via Ethernet are:

The application software for each component was written in [what?].

All components except the Ethernet are duplicated, the majority across the two halls, others within the appropriate office or the machine-room. The data is also duplicated, such that failure of the primary data-server results in suspension of the current auction and in delay in switching data-servers, but no other loss of data or auctioning capability. There is also a backup power supply.

The system has been very stable since it was implemented. The MIS has been further developed, and the weighing scale replaced by a more robust alternative, with consequential software changes. SEEL has an ongoing contract with HAM to provide maintenance and enhancement services, but the day-to-day system administration is performed by HAM staff.

* Lot-Size

It is unlikely that the current design can be speeded up any further, because the weighing activity is the current limitation on throughput. Another approach entirely would be to increase the lot size. This was set at one animal per lot, however, in order to enable prices to reflect demand for the specific qualities of each beast, as judged by the bidders. Pigs differ sufficiently that they could not economically be grouped into lots of sufficiently similar beasts. In addition, the bidders' judgement is dependent on the beast's movement, such that fat and musculature can be assessed.

* HAM's Role

HAM, the company, provides specific services in return for specific fees. It charges a service fee of 3% of each beast's sale-value (assuming an average of about $S240 per pig this is equivalent to a little over $S7 per pig). This covers unloading, inspection, washing, handling, auction, storage during the day of arrival/sale/slaughter and payment to the producer within three days. HAM recompenses the producer for deaths after arrival, at the day's average sale-rate for that class of animal [what classes are recognised - we need to check that], which provides an incentive for it to ensure appropriate handling and yardage. It charges extra where an animal stays overnight.

HAM management compares its fees favourably with that of other on-line auction service providers; for example the Dutch flower auction markets charge 5% of sale-value. Taiwan operators charge somewhat less than half as much as HAM, but do so on higher volume and higher sale-prices, and do not take responsibility for pig mortality.

HAM offered some interim arrangements during the first 2-3 years of the scheme, and in particular it was still paying the import licence fee of $150 per consignment 2 years after commencement of operations. It makes average prices for each class of animal available by phone [again - what classes?], and provides a weekly report to producers and bidders.


Relevant Outcomes
* The Nature of the Market

Bidders can make very quick, binding judgements about the value of a beast to themselves or their principal, despite being at a distance from the animal. They can also express that judgement by depressing and releasing a button on their chair-arm, with sufficient precision, even though the price is escalating by about 4% per second.

Much greater fairness has been achieved in relation to the price-setting mechanism. In addition, the weighing of each animal is subject to public scrutiny, and the yield can be readily calculated for each animal.

* The Implementation Phase

The introduction of HAM was marked by very active resistance to change, not only by the importers, who clearly stood to lose, but also by other actors, who either perceived or were encouraged to perceive that the change would be disadvantageous for them too. While the anger lasted, and particularly with the success of the application for an injunction against HAM, change was stymied.

When the licensing provisions were imposed, however, with its implicit requirement that importers use HAM, opposition wilted in a matter of weeks. The proportion of pigs passing through HAM climbed to 35% within a couple of months of the licensing provisions being imposed in May 1990, and after two years had reached about 50%, or an average of over 2000 animals per day. The proportion of pigs which each importer must sell on HAM has never been increased from the originally set 20%, because it has not been necessary to do so.

* Changes in the Sector's Structure and Process

Two years after HAM's introduction, the majority of consignments were being sent directly to the market by the producer, and the number of importers had reduced from 20 to about a dozen, and only about six of these were very active.

Little structural change had occurred in the classes of buyers. There did, however, appear to have been an increase in both the number of buying agents and the sophistication of buying agent services. Some 60% of sales were being made to specialists purchasing on behalf of particular classes of customer, for fees of the order of $3-5 per pig. The remaining 40% of the market was made up of direct purchase by:

* Contribution to Product Quality

Although this objective has been less prominent than fairness of the price-setting and weight-measuring mechanisms, there are several ways in which HAM has contributed to improvements in the quality of pigs and meat, and the closeness of fit between market needs and what the farmer is seeking to produce. Prices are sensitive to demand, and visible. The price-difference between upper and lower quality product is 40-60 cents per kg or $30-$50 per animal. Although detailed data on carcases is not gathered, analysed and published, producers come to the market not only from nearby Johor, but also from as far away as Penang, and hence are able to gather the information they need by observation and word-of-mouth.


Evaluation

This section assesses HAM from the perspective of the young and turbulent body of knowledge generally referred to as 'strategic information systems theory'. The analysis adopts the provisional structure used in Clarke [1992a]. This identifies in turn the motivations underlying the system's conception, design and exploitation; the nature of the processes of development and of implementation; the degree of intended change in the industry sector's structure and processes; and the management process which resulted in the system's emergence.

* Motivations

Most existing literature associates the motivation for strategic systems with the competitive framework of Porter [1980, 1985]; see also Parsons [1983], McFarlan [1984], Ives & Learmonth [1984], Porter & Miller [1985] and Keen [1986]. In particular, it focusses on the notion of competitive advantage, on the sustainability of advantage [Clemons 1986], and recently on leveraging on the organisation's existing characteristics and advantages [Beath & Ives 1986, Clemons & Row 1987, Ives & Vitale 1988, Hopper 1990] and strategic alignment of IT with corporate objectives [Henderson & Venkatraman 1989, Earl 1989, Broadbent & Weill 1991]. A secondary theme which has emerged is the implementation of IT as a competitive necessity, in order to neutralise a competitor's advantage [Vitale 1986, Warner 1987, Brousseau 1990]. A special case of this theme is the notion of second-mover advantage.

PIPL's objectives were quite expressly to change the structure of the pork industry in Singapore, and in particular to wrest control of the price-setting mechanism away from a cartel of importers: "[The objective is] to bring together both buyers and suppliers of live pigs for direct transaction, utilising modern computerised auctioning system to provide fast, fair and competitive bidding in an open manner ... Malpractices on prices, weight manipulations and losses through mortality are eliminated. These problems were not uncommon in the traditional method of sale" (HAM promotional brochure). The primary benefit sought was fairness in pricing, and the avoidance of monopoly profits being gained by the importers. This strategy must be seen in the context of the control over the supply of hogs into Singapore exercised by PPD.

A secondary benefit was greater sensitivity of suppliers to the patterns of demand. This is achieved through price differentials arising between different classes of pig, and improved information flows about those price differentials. Equity for Malaysian pig-farmers does not appear to have been a primary aim, although HAM does guarantee payment within 3 days. There appears to have been no revenue objective, beyond the need for PIPL's operations to cover their operational costs.

Competition among buyers to set the price for a pig was the strategic means whereby PIPL achieved its objective of fair and adaptive price-setting. HAM was thus not a competitive weapon, but rather a tool for a government agency to achieve a social and political purpose.

No alternative hog auctioning scheme has emerged, and as long as HAM continues to function in a manner perceived by participants to be efficient and fair, it appears that the investment necessary and the uncertainties relating to penetration of the market and licensing by PPD may be sufficiently high barriers to entry that no competitor will emerge. HAM, the company, intends to operate the two HAM systems at Jurong and Kim Chuan in an integrated manner, rather than encouraging them to compete. It therefore appears likely that the price-setting mechanism will remain a monopoly operated by a government business enterprise for the foreseeable future.

Competition-based strategic information systems theory does not provide a satisfactory framework for describing and explaining HAM's intentions. An interpretation of 'strategic information systems' is needed which treats competition as an element of a broader theory. For example, Clarke [1992b] identified the strategic motive for multi-organisational IT applications as being either a threat to competitors, a defence against intruders, or a facilitator of improvements at a level higher than that of a single participant. The last of these is clearly applicable to HAM.

* Development and Implementation

The processes of development and of implementation of a strategic information system are generally characterised as competitive, collusive among nominally competing suppliers, or collaborative within an alliance of complementary enterprises [Barrett & Konsynski 1982, Gummesson 1987, Johnston & Vitale 1988, Rockart & Short 1989, Wiseman 1989].

HAM was developed by PIPL, a government business enterprise which had a legislated monopoly in abattoirs, and which extended the scope of its operations to embrace the price-setting mechanism for the animals it slaughtered. It is feasible to interpret this as the entrance of a new player into an established industry sector, and an exercise of market power in order to drive the existing participants out. It is more realistic, however, to recognise PIPL's actions as being of a regulatory nature, supra to the competitive realm. Once again, it is apparent that the conventional competition-based theory requires extension in order to achieve a satisfactory level of explanatory and predictive power .

* Degree of Intended Effect

In order to achieve its objectives, PIPL chose not to merely automate the existing procedures. Its intention was to achieve fairness in the setting of prices, and this necessarily involved re-structuring of the industry sector, and hence revolutionary change. It did not undertake gradual change, but rather judged that a severe jolt to existing market practices was necessary.

In a mixed-market economy such as Singapore's, government business enterprises (GBEs) have multiple objectives. Their operational style must be consistent with private enterprise management principles, and must make a reasonable return on funds employed. On the other hand, they must assist in, and may even be responsible for, the implementation of relevant government policy. Important government policies which influenced PIPL's strategy were the controlled supply of pigs into Singapore (exercised by PPD), the need for reliability in the supply of fresh pork, and the desire for fairness and responsiveness in the price-setting mechanism.

* The Management Process

The literature has generally presumed that strategic systems arise from semi-plannable management processes [Rackoff et al 1985, Earl 1986, Vitale et al 1986, Lederer & Mendelow 1988]. Recently, the serendipitous origins of many of the key cases has been more closely examined, and approaches sought whereby inventiveness arising at all levels and locations in an enterprise can be exploited [Mintzberg 1990, Ciborra 1991].

HAM was a calculated, strategic manoeuvre. Although a range of negative examples existed, in the form of failed attempts to automate pig sales in various parts of North America, prior successful on-line pig trading schemes existed in Taiwan, and PIPL's executives took full advantage of the experience gained, and conceived and developed a related scheme suitable for the conditions prevailing in Singapore. Existing strategic information systems theory appears satisfactory as a basis for describing and evaluating the process whereby HAM came into existence.

* Conclusions

The application of strategic information systems theory was of value in assessing the HAM system and its intentions and impacts. It is apparent, however, that the existing, competition-based approach needs to be complemented by alternative perspectives and in time subsumed by a broader theory.


Further Developments

HAM signed a contract with SEEL in January 1992 for the delivery of a system for the second abattoirs at Kim Chuan, which slaughters the remaining 40% of Singapore's pig supply. The new system is planned for implementation in December 1992, and includes a straight rather than a curved run and greater control over the physical flow of bidders.

Because the equpment used in the original system has been superseded, it has been necessary to re-construct the hardware and software. Despite this opportunity, little change is being made to the system's functions, apart from some improvements to the user interface for the auctioneer, provision for multiple representatives per bidder, integration forward to the slaughter-process, changes to farm and importer code-structures, and some rationalisation of the backend MIS in order to run both markets on a single machine (a more powerful HP3000-937), and integrate the resulting management information. The new system features are to be progressively built back into the existing system at Jurong.

HAM's stated intention is to sell 100% of Singapore's pigs through on-line auction, and eliminating the secondary market. The reason given for this is that it will provide all participants with a fair price-setting and weight-determining mechanism, and orderly and assured payment arrangements. The question remains as to whether it may be sufficiently fair and orderly, and more efficient, to permit market-players to decide whether to accept the previous day's average prices as a basis for negotiating sales of lots of pigs. This applies in particular to the 20% of the market which deals in lower quality and therefore lower value animals and meat, the bulk of which is handled by the wholesalers.

It is possible that the experience with HAM may be applied to other agricultural and pastoral produce, especially fruit and vegetables. It is also possible that SEEL may market the new version of the product to pig markets in other countries, and perhaps also to other livestock or produce markets.


Acknowledgements

The authors express their gratitude for the considerable assistance provided by Mr Paul Lu, General Manager of the Meats Department of Primary Industries Pte Ltd and also by Mr Ng Eng Hup, Accountant and Company Secretary of Hog Auction Market (Pte) Ltd and Mr Chia King Young, Project Manager of Singapore Electronic and Engineering Limited and several members of his staff. [The companies have confirmed the factual data contained in the report, and indicated that they do not oppose the publication of this report, but] all interpretations and conclusions are the responsibility of the authors alone.


Reference List

Barrett S. & Konsynski B. (1982) 'Inter-Organisational Information Sharing Systems' MIS Qtly Special Issue (Dec 1982) 92-105

Beath C.M. & Ives B. (1986) 'Competitive Information Systems in Support of Pricing' MIS Qtly 10,1 (Mar 1986) 85-96

Brousseau E. (1990) 'Information Technologies and Inter-Firm Relationships: The Spread of Interorganisational Telematic Systems and Its Impacts on Economic Structures' Proc. 8th Int'l Telecomms. Conf., Venice, March 1990

Ciborra C.U. (1991) 'From Thinking to Tinkering: The Grassroots of Strategic Information Systems' Proc. 12th Int'l Conf. Inf. Sys., New York, December 1991, DeGross J.I. et al (Eds.), Association for Computing Machinery, New York

Clarke R.A. (1992a) 'Strategic Information Systems: Retrospect and Prospect' Proc. Int'l Conf. Inf Sys and Org, Bled, Slovenia, 28-31 August 1992

Clarke R.A. (1992b) 'A Contingency Model of EDI's Impact on Industry Sectors' J. Strat. Inf. Sys. 1,3 (June 1992)

Clemons E.K. (1986) 'Information Systems for Sustainable Competitive Advantage' Information & Management (November 1986) 131-136

Clemons E.K. & Row M. (1987) 'Structural Differences Among Firms: A Potential Source of Competitive Advantage in the Application of Information Technology' Proc. Int'l Conf. Info. Sys. ICIS'86, Pittsburgh, December 1987

Earl M.J. (Ed.) (1986) 'Formulating Information Technology Strategies' Oxford. U. P., Oxford, 1986

Earl M.J. (1989) 'Management Strategies for Information Technology' Prentice-Hall, 1989

Gummesson E. (1987) 'The New Marketing: Developing Long-Term Interactive Relationships' Long Range Planning 20 (1987) 10-20

Henderson J.C. & Venkatraman N. (1989) 'Strategic Alignment: A`Process Model for Integrating Information Technology and Business Strategies' MIT, Sloan Working Paper, 1989

Hopper M.D. (1990) 'Rattling SABRE - New Ways to Compete on Information' Harvard Business Review (1990)

Ives B. & Learmonth C.P. (1984) 'The Information System as a Competitive Weapon' Commun. ACM 27,12 (December 1984) 1193-1201

Ives B. & Vitale M.R. (1988) 'After the Sale: Leveraging Maintenance with Information Technology' MIS Qtly 12,1 (March 1988) 6-21

Johnston H.R. & Vitale M.R. (1988) 'Creating Competitive Advantage with Interorganisational Systems' MIS Qtly 12,2 (June 1988) 153-65

Keen P.G.W. (1986) 'Competing in Time: Using Telecommunications for Competitive Advantage' Cambridge Mass, Ballinger, 1986

Lederer A.L. & Mendelow A.L. (1988) 'Convincing Top Management of the Strategic Potential of Information Systems' MIS Qtly 12,4 (December 1988)

McFarlan F.W. (1984) 'Information Technology Changes the Way You Compete' Harvard Business Review (May/Jun 1984) 98-105

Mintzberg H. (1990) 'The Design School: Reconsidering the Basic Premises of Strategic Management' Strat. Mngt J. 11 (1990) 171-195

Parsons G.L. (1983) 'Information Technology: A New Competitive Weapon' Sloan Mgt Rev 25,1 (Fall 1983)

Porter M.E. (1980) 'Competitive Strategy' The Free Press, New York, 1980

Porter M.E. (1985) 'Competitive Advantage' The Free Press, New York, 1985

Porter M.E. & Millar V.E. (1985) 'How Information Gives You Competitive Advantage' Harvard Business Review 63,4 (1985) 149-160

Rackoff N., Wiseman C. & Ullrich A. (1985) 'Information Systems for Competitive Advantage: Implementation of a Planning Process' MIS Qtly 11,4 (December 1985)

Rockart J.F. & Short J.E. (1989) 'IT in the 1990s: Managing Organisational Interdependence' Sloan Mngt Rev. (Winter 1989)

Vitale M. (1986) 'The Growing Risks of Information Systems Success' MIS Qtly 10,4 (Dec 1986) 327-334

Vitale M., Ives B. & Beath C. (1986) 'Identifying Strategic Information Systems' Proc. 7th Int'l Conf. Inf. Sys., San Diego, December 1986, pp.265-276

Warner T.N. (1987) 'Information Technology as a Competitive Burden' Sloan Mngt Rev. 29,1 (Fall 1987)

Wiseman C. (1989) 'Strategic Information Systems' Irwin, 1988


Navigation

Go to Roger's Home Page.

Go to the contents-page for this segment.

Send an email to Roger

Created: 22 November 1997

Last Amended: 22 November 1997


These community service pages are a joint offering of the Australian National University (which provides the infrastructure), and Roger Clarke (who provides the content).
The Australian National University
Visiting Fellow, Faculty of
Engineering and Information Technology,
Information Sciences Building Room 211
Xamax Consultancy Pty Ltd, ACN: 002 360 456
78 Sidaway St
Chapman ACT 2611 AUSTRALIA
Tel: +61 6 288 6916 Fax: +61 6 288 1472