Electronic Reverse Auction and The Public Sector: Factors of Success Moshe E. Shalev & Stee Asbjorensen

JOURNAL OF PUBLIC PROCUREMENT, VOLUME 10, ISSUE 3, 428-452 FALL 2010

ELECTRONIC REVERSE AUCTIONS AND THE PUBLIC SECTOR:

FACTORS OF SUCCESS
Moshe E. Shalev and Stee Asbjornsen*

ABSTRACT. Electronic reverse auctions (e-RAs) provide purchasers with an

opportunity for greater savings, transparency and administrative benefits.
This paper reviews the literature, identifying the factors relevant to public
sector e-RA success, and then tests these factors through a statistical
analysis of a large sample of e-RAs, with a view towards providing guidance
to public sector purchasers looking to conduct e-RAs. Our research found no
correlation between e-RA success and the number of bidders or auction
volume, but did find correlation between e-RA success and (1) competition
among suppliers, (2) purchase complexity, and (3) purchase specification.
The number of bidders was found to be correlated with competition among
suppliers, specification and complexity. Auction volume was found to be
correlated with none of the factors. We discuss the possible explanations for
our results and draw conclusions for public sector procurement managers to
consider when deciding whether to conduct an e-RA for a particular
purchase.

INTRODUCTION
Introduced during the 1990s, electronic reverse auctions (e-RAs)
have grown popular in the business-to-business (B2B) and
government procurement communities. The use of e-RAs has
increased in both relative and absolute terms over time, and
practitioners expect this trend to continue (Jap, 2002; Beall et al.,
2003; Kaufmann & Carter, 2004; Schoenherr & Mabert, 2007).
Developed country governments, including Australia, Canada,
France, Korea, Portugal, the United Kingdom and the United States,
-------------------------------
* Moshe E. Shalev, LLM, MBA, and Stee Asbjornsen, BA, LLB, are the
Contracts Officer, and an Associate Procurement Officer, respectively,
United Nations Secretariat Procurement Division, New York.

Copyright © 2010 by PrAcademics Press

ELECTRONIC REVERSE AUCTIONS AND THE PUBLIC SECTOR: FACTORS OF SUCCESS 429

and many developing country governments such as Brazil, Chile,

China, Colombia, Ecuador, Estonia, Panama, Paraguay and Peru,
among others, as well as thousands of private sector organizations
such as General Electric, BMW, Nestlé and Kimberly-Clarke, already
make extensive use of e-RAs. For example, the United-Kingdom
Office of Government Commerce [OGC] reported savings of £50
million in IT hardware e-RAs over four years, and anticipating total
savings of £270 million through all e-RA purchases by the end of
fiscal year 2011/2012 (OGC, 2010), while General Electric claimed to
have saved $600 million, or 8% of their spending, through e-RAs in
2001 (Sawhney, 2003).
In an e-RA, suppliers compete dynamically online against each
other to offer the lowest price, resulting in greater savings to the
buyer. Since the entire process happens online, it is also often faster
and more transparent than traditional procurement methods.
However, not all e-RAs are successful in achieving savings for buyers,
and suppliers have also complained that the e-RAs ignore product
innovation and quality.
The existing literature on e-RAs focuses primarily on the private
sector, and little research has been conducted on the factors that
lead to successful e-RAs in the public sector. This is despite the fact
that the public sector as a whole is the biggest single buyer in any
country (Lindskog, 2008), and is estimated to constitute
approximately 15% of worldwide GDP on average and about 20% in
some OECD countries (Organisation for Economic Co-operation and
Development, 2008). Wagner and Schwab (2004) noted that the
majority of existing research is conceptual, prescriptive and
descriptive in nature, and based on qualitative interviews and case
studies. They also noted that even when surveys were conducted,
the sample sizes were small. We have found that most papers that
do focus on the factors that affect e-RAs success in the public sector
rely primarily on surveys and case studies to support their
conclusions.
Following a review of the literature, we identified three broad
categories of factors that may affect the savings achieved by buyers,
or the “success” of e-RAs. The first category of factors relates to
market conditions: the number of bidders, the total auction volume
and the competition among suppliers. The second category relates to
the purchased product or service: its complexity, and how well it was
430 SHALEV & ASBJORNSEN

specified. The final category relates to factors that resulted from

changing suppliers, including the break in trust with longstanding
suppliers and switching costs.
The purpose of this paper is to contribute a statistically-supported
evaluation of these factors using a relatively large sample of e-RAs to
determine which affect the success of public sector e-RAs. We then
draw conclusions for public sector procurement managers to consider
when deciding whether to conduct an e-RA for a particular purchase.

OVERVIEW
Electronic Reverse Auctions
While auctions provide suppliers the opportunity to find the best
price among interested buyers, reverse auctions give buyers a chance
to find the lowest-price supplier. In e-RAs, the buyer is able to gather
all interested suppliers online, irrespective of their geographic
location, at a relatively low cost (Jap, 2002). During the e-RA,
suppliers submit multiple offers, usually as a response to competing
suppliers’ offers, bidding down the price of a good or service to the
lowest price they are willing to offer. By revealing the competing bids
in real time to each participating supplier, e-RAs promote
“information transparency”. This, coupled with the dynamic bidding
process, improves the chances of reaching the fair market value of
the purchase (Sashi & O'Leary, 2002; Mabert & Skeels, 2002;
Smeltzer & Carr, 2003; Beall at al., 2003; Schoenherr & Mabert,
2007).
The Advantages of e-RAs
e-RAs are promoted as providing purchasers with a number of
benefits. The foremost is that e-RAs often provide buyers with
significant price reductions. Savings from e-RAs generally average
5% to 30% when compared with traditional bidding methods, with
15% being more typical (Sashi & O'Leary, 2002; Jap, 2002; Beall et
al., 2003; Major, 2007), although exceptional savings as high as 90%
have been achieved (Beall et al., 2003). However, more recently,
Singer, Konstantinidis, Roubik, & Beffermann (2009) questioned
these purportedly high savings rates, suggesting that savings of
around 4% are more realistic.

Second, e-RAs provide a number of administrative benefits. They

are quick; avoid the need for subsequent price negotiations; allow for
easier information delivery through email and the Internet; and offer
quicker procurement cycle times, thereby allowing for lower inventory
levels (Mabert & Skeels, 2002; Jap, 2002; Smeltzer & Carr, 2003;
Beall et al., 2003; NSW, 2006; Singer et al., 2009). Collectively, such
benefits lead to better outcomes in less time, resulting in a more
efficient procurement process for both suppliers and the buyer.
Third, e-RAs provide for enhanced transparency by disclosing all
bids in real time and by reducing the need for person-to-person
contacts among suppliers, and with buyers. The electronic record
permits a comprehensive audit trail and compliance capabilities
(Turley, 2002; Walker & Harland, 2008; Schapper, 2008; Alilovic,
2008). As transparency and compliance become increasingly
important to public sector practitioners, these administrative benefits
may become even more important than any cost savings achieved.

The Disadvantages of e-RAs

e-RAs, when conducted in their most basic form, restrict suppliers
to compete on price alone. This threatened to damage longstanding
supplier relationships and lead to reduced supplier innovation and
quality, as anything above the minimum specifications was not
rewarded by simple e-RA designs. By ignoring past supplier
performance and loyalty, the introduction of e-RAs caused alarm
among incumbent suppliers, leading some to refuse to participate in
e-RA-based procurement exercises (Smeltzer & Carr, 2003; Emiliani &
Stec, 2005). In a situation known as “winner’s curse”, some
suppliers caught up in the auction environment bid below their costs
in order to win (Sashi & O'Leary, 2002). Finally, researchers
expressed concerns that e-RAs might be coercive and unethical
(Emiliani & Stec, 2002; Carter et al., 2004).

LITERATURE REVIEW
The literature identified a wide range of potentially relevant
factors which may affect the success of e-RAs in the private sector.
The most commonly mentioned were (1) the number of bidders, (2)
auction volume, (3) competition among suppliers, (4) purchase
complexity, (5) purchase specification, (6) the importance of the
supplier-buyer relationship, and (7) switching costs (Wagner &

Schwab, 2004; NSW, 2006; Crane, 2008; Gattiker, 2008). The first
three factors addressed market conditions, and sought to measure
how much interest suppliers had in competing to provide a product or
service, including auction volume, the number of bidders and
competition among suppliers. The following two factors consider the
product or service in question: how complex the product or service is,
and how well the buyer specified it. The final two factors relate to
changing suppliers: the implications of potentially changing suppliers
through an e-RA on the quality of the product and the purchase price,
and the effect of the costs of switching to a new supplier.

Market Factors
As with any auction, favorable market conditions improve the
likelihood of auction success. The number of bidders participating in
e-RAs was identified as a crude indicator of supplier interest. Auction
volume is thought to be an important means to attract suppliers to
participate in e-RAs and to allow suppliers to accept lower margins
based on higher total profit potential. Competition among suppliers is
a measure of how strongly participating suppliers are competing with
each other to win the e-RA in the product-particular market.

Number of Bidders
The literature on e-RAs emphasizes the importance of the number
of bidders in ensuring that the lowest possible prices are achieved. In
a sealed-bid auction, each supplier has only a single opportunity to
bid. The difference when using e-RAs is that bidders dynamically drive
the bid price lower only when beaten by another bidder.
Consequently, there must be at least two suppliers, and preferably
more, to achieve savings.
Singer et al. (2009) found that e-RAs with only one supplier lead
to detrimental outcomes, that those with two suppliers had mixed
outcomes, and that most of e-RA savings can be attributed to
attracting more bidders. The literature also provides examples of
successful e-RAs with only two suppliers (Turley, 2002). However, in
such cases there is a risk that anonymity will be lost, and that
suppliers will be able to identify each other (Beall et al., 2003; Yukins
& Wallace, 2005). This may also create a greater risk of collusion,
whereby suppliers may agree not to compete and divide the auctions
among themselves (NSW, 2006).

Most researchers have suggested that at least three to five viable

suppliers are necessary for an effective e-RA (Mabert & Skeels, 2002;
Beall et al., 2003; Kaufmann & Carter, 2004; Schoenherr & Mabert,
2007; Major, 2007). More generally, there is widespread support in
the literature (De Boer, Harink, & Heijboer, 2002; Jap, 2002; Mabert
& Skeels, 2002; Sashi & O'Leary, 2002; Crane, 2008) that “[f]irms
should not conduct e-RAs with only few suppliers” (Wagner & Schwab,
2004, p. 23) and that “[g]enerally, the more participants, the more
competitive the bidding” (Mabert & Skeels, 2002, p. 76). Singer, et
al. (2009) found clear evidence that an increase in the number of
bidders led to greater e-RA savings.
While there is widespread support for having high numbers of
bidders, Wagner & Schwab (2004) found no correlation between the
number of bidders and the success of the e-RA, although they did find
a strong correlation between the number of bidders and competition
among suppliers, and so recommended that the number of bidders
be considered.

Auction Volume
In order to compensate for lower prices and lower margins,
suppliers seek higher volumes to maintain or increase the total
revenue. Buyers, in turn, benefit from the lower transaction costs and
economies of scale that result from larger volumes (Smeltzer & Carr,
2003; Beall et al., 2003; Carter et al., 2004; Kaufmann & Carter,
2004). As well, buyers will seek to offset the additional e-RA costs,
normally resulting from third-party e-RA service-provider fees. This
cost depends on the scope of the service to be provided by the third
party provider, and may range between 0.25%-5% of procurement
volume, occasionally subject to minimum and maximum amounts
(Mabert & Skeels, 2002; Turley, 2002).
In part as a result, numerous studies have indicated that the
procurement volume must be sufficiently high to provide sufficient
profits to attract enough suppliers, and provide buyers with enough
savings to cover their additional costs (Jap, 2002; Sashi & O’Leary,
2002; De-Boer et al., 2002; Beall et al., 2003; Kaufmann & Carter,
2004). The Australian government of New South Wales [NSW]
recommended to public procurement buyers that “[g]enerally, the
higher the volume of the procurement, the greater the potential for
realizing savings” (NSW, 2006, p. 5).

Despite this, the literature cites examples of smaller auctions that

did attract suppliers. For example, the United States Department of
Defense (DoD) was able to attract bids for auctions of only $10,000
(Turley, 2002). Yet, Turley then cites the DoD as suggesting a
minimum auction volume of $500,000 for a successful e-RA. The
Office for Government Commerce of the United Kingdom has
suggested £80,000 as a minimum amount (OGC, 2005), while Beall
et al., (2003) found that even $1,000,000 is too low for an e-RA to
succeed. However, Schoenherr & Mabert (2008) noted that while the
auction volume should be high enough to enhance the e-RA’s success
and to attract more suppliers, it should not be so high as to strain
suppliers’ capabilities and eliminate small and medium businesses,
which comprise a significant proportion of suppliers in most markets.

Competition among Suppliers

While the number of bidders measures how many suppliers are
participating, competition among suppliers attempts to measure the
degree to which the suppliers are competing with each other. It
stands to reason that a large number of bidders is of no benefit if only
one of them is willing to make concessions to win the bid, while the
others are unwilling to reduce their prices. Similarly, a few fiercely
competitive bidders may result in great savings.
There is wide agreement in the literature on the importance of
competition to e-RA success (Beall et al., 2003; NSW, 2006; Major,
2007; Crane, 2008). Jap (2002, p. 515) notes that “competition
among suppliers forms the bedrock of successful reverse auctions”,
while Wagner & Schwab (2004, p. 22) went so far as to say that “one
can summarize the findings of our study as: ‘all that counts is
competition’. The importance that a competitive situation exists can
never be over-emphasized.”

Purchase Complexity and Specification

Although e-RAs may be used to purchase anything, they are
normally recommended only for simple purchases, such as
commodities. Since e-RAs do not permit subsequent rounds of
negotiation to address any issues that may arise, it becomes crucial
that the purchase be very well specified before the auction. The
difficulty of accurately specifying complex purchases leads some
researchers to recommend only simple purchases for e-RAs.

Purchase Complexity
Simple e-RAs restrict suppliers to compete on the price alone.
Where other factors such as intangible concerns or non-price factors
are involved, direct negotiations with suppliers are thought to be
necessary to discuss all aspects of the purchase. It has therefore
been argued that e-RAs are best suited for simple purchases or
commodities, where the only differentiating factor is price (Jap, 2002;
Beall et al., 2003; Yukins & Wallace, 2005; NSW, 2006). Schoenherr
and Mabert (2008) also found that increased complexity in purchases
appeared to reduce success.
In addition, since simple e-RAs focus primarily on price
competition and fail to consider product quality or innovation,
suppliers may end up focusing on price to the exclusion of other
considerations, leading to decreased product quality and fewer
product innovations. This leads some observers to suggest that e-
RAs should be limited to standardized or commoditized purchases
(Smeltzer & Carr, 2003).
As there are often more suppliers for simple, commoditized
goods, some researchers have recommended that e-RAs be limited to
such goods, as the larger number of bidders would likely result in
increased competition and hence greater price savings (Beall et al.,
2003; Wagner & Schwab, 2004). Schoenherr and Mabert (2008)
found that the greater the complexity, the fewer bidders participated.
Furthermore, suppliers of complex goods are often also strategic
suppliers and have longstanding relationships with the buyer, which
the buyer is reluctant to risk. Smeltzer and Carr (2003) noted that, as
a result, buyers often fear destroying such suppliers’ trust, making
complex product or service procurement through e-RA too risky,
although they called for further research in this area.
While there are a number of arguments in favor of using only
simple products, some researchers noted that complexity may not
affect e-RA success. Wagner and Schwab (2004) found no
correlation between complexity and e-RA success, although they
noted that in all of the e-RAs surveyed the complexity was “rather
low”. Moreover, others observed that simple products may have
lower margins, particularly when they have been commoditized and a
large number of suppliers are in the market. In contrast, complex
products tend to have higher margins, and as a result offer greater

potential savings (Smeltzer & Carr, 2003; Beall et al., 2003;

Kaufmann & Carter, 2004).
In order to use e-RAs for complex goods and services and give
some comfort to suppliers, some observers have suggested a “two-
stage” solution. The first stage involves a technical evaluation,
producing a technical score. The second stage involves obtaining a
score based on the relative bid price through an e-RA. The technical
score can either be integrated dynamically into the e-RA by
automatically modifying the bid rank in real time, or combined with
the bid scores of each vendor after the e-RA is completed (Turley,
2002; SAP, 2006; NSW, 2006; Schoenherr & Mabert, 2007).

Purchase Specification
Purchase specification faces similar concerns to those relating to
purchase complexity. As e-RAs normally limit suppliers to compete
only on price, it becomes imperative that all suppliers bid on the
same specifications. As a result, researchers have recommended
that e-RAs be used only for very clearly defined products (De-Boer et
al., 2002; Smeltzer & Carr, 2003; Beall et al., 2003; Kaufmann &
Carter, 2004; NSW, 2006). One commentator expressed concern
that poorly specified products, combined with a lack of subsequent
rounds of negotiations, may create uncertainty in bidders and lead to
higher prices for buyers (Crane, 2008).
Wagner and Schwab (2004) found a strong positive logistic
correlation between “easy to specify demand” and “success”, but
noted that they did not find a correlation using linear regression.
Gattiker (2008) found that specification was the most important
factor in the opinion of buyers, echoing the comment “[i]f you can
spec it, you can bid it” (Beall et al., 2003, p. 51). As long as
purchases can be specified, e-RAs can be used for even complex
purchases (Kaufmann & Carter, 2004). However, Schoenherr and
Mabert (2008) found that difficulty in specifying demand has no
direct impact, either negative or positive, on e-RA success, although
they did find an indirect effect, as the difficulty to specify demand was
found to negatively correlate with the number of bidders.

Changing Suppliers
Some of the most common criticisms of e-RAs, particularly in
popular press articles, relate to concerns about buyers using e-RAs to
change longstanding suppliers, merely because of price. However,
we conclude that these issues do not apply in the public sector, first
because the nature of public sector procurement requires that most
purchases be made through a competitive bidding process,
regardless of form, and second because these arguments do not
apply to e-RAs any more than they would to more traditional auctions,
such as sealed bid auctions.

Relationship with the Supplier

The literature on e-RAs places significant concern on the erosion
of trust between long-standing strategic suppliers and purchasers
following the implementation of e-RAs. Emiliani and Stec (2005)
observed that e-RAs cause divisiveness between suppliers and the
buyer, and that the buyer must take a long-term view of purchasing,
rather than focusing only on short-term savings. Others note that e-
RAs cause a breakdown in trust by implying that the buyer’s
expectations are no longer being met, and therefore the purchase will
be “shopped around” (Smeltzer & Carr, 2003). Moreover, some
suppliers claim that e-RAs are only a tool for the buyer to conduct a
market-survey or to obtain further price concessions from the
incumbent supplier, with no real intention to award the business to
the winning supplier (Jap, 2002). As a result of all these concerns,
some suppliers may refuse to participate, reducing the number of
bidders in the auction, while other suppliers may no longer be willing
to invest in improving the quality of the goods provided to the buyer
(Jap, 2002; Beall et al., 2003).
Perhaps the concern which underlies all of these arguments is
that existing suppliers are naturally reluctant to risk their margins by
entering into dynamic competitive bidding arrangements (Sashi &
O'Leary, 2002).
Furthermore, some researchers appear to show that these
concerns about trust may be exaggerated. Jap (2002) and Carter et
al. (2004) observed that about two-thirds of buyers surveyed claimed
improved buyer-suppliers relationships, greater cost savings, better
performance, delivery and product excellence. Beall et al. (2003, p.

60) found “little or no evidence that e-RAs are driving suppliers into
non-sustainable relationships with the buyers. In fact, there is
evidence to suggest the opposite,” and Schoenherr and Mabert
(2007) observed that this was how e-RAs were perceived only in the
very beginning, and they suggested some strategies in order to
overcome this prejudice. Recent research has also noted that buyers
perceived the relationship to be suffering, but that suppliers largely
felt that their relationship with the buyers remained unaffected by e-
RAs (Loesch & Lambert, 2007).
One circumstance where e-RAs, and indeed any auction process,
is not appropriate, is with strategic suppliers. Strategic suppliers are
those that provide to the buyer a crucial product or service not easily
replaced in the market. In such circumstances a competitive auction
process is not appropriate, since replacing the supplier is not a viable
option for the buyer (Mabert & Skeels, 2002). Hartely, Lane and
Duplaga (2006) found that most buyers were able to distinguish
between strategic and non-strategic suppliers and applied e-RAs in
appropriate situations.

Switching Costs
Sometimes, the cost to a buyer of switching suppliers may be very
high. Wagner and Schwab (2004, p. 16) observed that “[i]n general,
the lower the switching costs, the more a buyer can benefit from
conducting a reverse auction because switching costs would be offset
by the potential savings. Therefore, several researchers have pointed
out that the lower the expense of switching suppliers, the more
suitable e-RAs would be”.
Wagner and Schwab (2004) also found a correlation between
switching costs and the number of bidders willing to bid, since
competing suppliers are unlikely to bid when they suspect that the
buyer is not going to change suppliers due to high switching costs.
Further, the incumbent supplier is likely to resist an e-RA in these
circumstances, as they may perceive it as just a gambit to reduce the
supplier’s margin (Smeltzer & Carr, 2003; Emiliani & Stec, 2005). In
certain circumstances, the costs of switching suppliers may exceed
the savings available from a procurement exercise, and alternative
solutions should be considered.

Relevance to the Public Sector

We did not include the relationship with the supplier or switching
costs in our research for the following reasons. First, in the public
sector, buyers are almost always obliged to engage in an auction
process as contracts expire. As a result, the introduction of e-RAs
would not create new tensions in the relationship with suppliers
(Major, 2007). The highly regulated auction process in the public
sector, combined with the open and transparent nature of e-RAs, may
even enhance supplier trust. Similarly, given that periodic auctions
cannot be avoided in the public sector, switching costs are always
incurred, and so switching costs would not be a relevant
consideration.
Second, the very nature of any auction process, whether e-RA or
any other traditional method, means that neither risks to the
relationship with the supplier nor switching costs can be avoided.
The only alternative would be to conduct a non-competitive,
negotiated process, which in most circumstances is not permitted for
public sector buyers. As a result, public sector suppliers will always
risk being replaced, and switching costs will be incurred in each case,
regardless of the form of auction. Since we are seeking to evaluate e-
RAs as an alternative to traditional public sector auction methods, we
have not evaluated these factors, as they apply equally to any form of
auction.

RESEARCH
Research Methodology
This paper reviewed the primarily private sector-focused literature
to identify factors that affect the success of e-RAs, and then identified
those most relevant to public sector procurement by considering the
structural difference between public and private sector procurement.
Data on these factors was gathered from 139 e-RAs conducted by
four organizations operating in the United States and the EMEA. The
139 e-RAs were used to purchase a wide range of products and
services, from simple commodities to electronics, to a variety of
services ranging from gardening to call-centers. Telephone interviews
were subsequently held with purchasing managers from these
organizations. The e-RA data was examined through statistical
regression analysis, with a view to evaluating which of these factors

correlated with e-RA success for public sector buyers, and to what
extent. We concluded by offering recommendations for public sectors
buyers considering whether to conduct an e-RA for specific products
or services.

Success
We considered e-RAs to be successful when the price achieved
was lower than the price the buyer expected to pay, based on past
auction experience and market surveys. The variable “success” was
defined by the ratio between the expected price and the price
achieved, and ranged from a high of 36.36% savings to a loss of
28.6%, with an overall average savings of 12.1%. Only four e-RAs
failed to achieve any savings. However, two organizations did not
provide data for e-RAs which had only one bidder, since such bids
were automatically cancelled.
Our study did not consider the additional administrative costs of
hosting e-RAs, and did not evaluate whether the purchase price
savings exceeded such costs. As well, it was not possible to place a
financial value on the administrative benefits which resulted from
implementing a more streamlined and efficient online process,
compared to paper-based procurement exercises. Due to the
difficulties of measuring the additional costs of conducting e-RAs
against any efficiency gains (Wagner & Schwab, 2004), the variable
success is limited to the difference between the expected and actual
costs of purchases.

Number of Bidders
The e-RAs reviewed attracted an average of 6.7 competitors, from
a low of 1 to a high of 26. While we evaluated the effect that the
number of bidders has on success, we did not seek to identify an
optimal number of bidders.

Competition among Suppliers

Buyers were asked to estimate the competition among suppliers
using three factors from Porter’s model for industry analysis to
measure the degree of rivalry between suppliers (Porter, 1980). The
buyers were asked to rank on a scale of one to three each of the
following factors in relation to the competitive nature of the purchase-

specific supplier base: the market share concentration of suppliers,

the barriers to entry and exit in their market, and suppliers’ sensitivity
to competing supplier actions. Each sub-factor score was then
aggregated to provide a score that approximated the degree of
competition. For example, a market with one very dominant supplier,
with very high barriers to entry, and complete indifference to the
actions of its competitors would rank one, one and one, for an
aggregate score of three.

Complexity and Specification

Buyers were asked to estimate the complexity of each product or
service from a scale of one to five, with one representing a simple
product or service, and five being very complex. Finally, buyers
estimated how accurately the e-RA described the product or service
on a scale of one to five, with one representing a poorly defined
specification, and five representing a very well-defined specification.

Auction Volume
The auction volume was defined as the final price achieved, and
ranged from $1,528 to $16.5 million, with an average of $771,492.

Research Findings
The bivariate correlations between the different variables of our
research are summarized in Table 1. The main findings of our
research are as follows.

TABLE 1
Correlation Matrix
Competition Number of Auction Complexity Specification
Bidders Volume
Success 0.4015* 0.0970 . -0.0034 -0.3224* 0.3411*
Competition - 0.5654** 0.0476 -0.5028** 0.5288**
Number of - - -0.0514 -0.4235** 0.3744*
Bidders
Auction - - - 0.0231 0.1447
Volume
Complexity - - - - -0.7209**
Notes: ** Significant at the 0.01 level.
* Significant at the 0.05 level.

Success
Surprisingly, we found that success did not correlate with either
auction volume or the number of bidders, despite widespread support
for their importance in the literature. However, we found that success
was correlated with competition (r=0.401, p<0.05) (the higher the
competition, the greater the success), specification (r=0.341,
p<0.05) (the better defined the purchase, the greater the success)
and complexity (r=-0.322, p<0.05) (the less complex the purchase,
the greater the success).

Competition
Competition was also found to correlate strongly with the number
of bidders (r=0.565, p<0.01), specification (r=0.529, p<0.01) and
complexity (r=-0.503, p<0.01), but not with auction volume.

Number of Bidders
We found that the number of bidders correlated with competition
(r=0.565, p<0.01), specification (r=0.374, p<0.05) and complexity
(r=-0.423, p<0.01). No correlation was found between the number of
bidders and either success or auction volume.

Specification
In addition to success, competition and number of bidders,
specification was also strongly correlated with complexity (r=0.721,
p<0.01), but not with auction volume.

Complexity
Complexity correlated negatively with all other variables except
auction volume.

Auction Volume
Auction volume did not correlate with any other variable.

DISCUSSION AND RECOMMENDATIONS

Following our literature review, we identified seven factors that
are commonly said to affect the success of e-RAs, and we considered
five to be particularly relevant to public sector buyers. Our review of

139 e-RAs found a number of similarities with the private sector

literature, but also some important differences.

Success
Success was found to be correlated with competition, complexity
and specification, and not correlated with the number of bidders or
auction volume.

Competition
Success correlated most closely with competition. The literature
widely supported the importance of competition as the primary driver
of e-RA success. Our findings appear to support that argument, as
competition correlated strongly with three out of the other four
variables measured, save auction volume. As e-RAs are designed to
leverage the competitiveness of suppliers to provide greater savings,
this conclusion is not surprising.

Specification
We note that Schoenherr and Mabert (2008) found no correlation
between success and specification, but did find it with complexity,
while Wagner and Schwab (2004) found the exact opposite. Our
findings, however, found a correlation between success and both
specification and complexity.
Some of the buyers interviewed suggested that specification
affects only the degree of competition among suppliers, and so any
effect on success would be indirect, if it exists at all. These buyers
believed that competition was the only causal factor affecting
success, while specification had either an indirect or spurious effect.
However, others noted that without clear specifications, some
suppliers would either hesitate to participate in e-RAs or would
provide themselves with additional margin in order to compensate for
this uncertainty, thereby directly affecting success.

Complexity
Success was found to be negatively correlated with complexity.
This finding may support the argument that complex purchases
require further clarifications and negotiations to discuss all aspects of
a purchase. We also observe that the same buyers that believed

specification had either an indirect or spurious effect on success also

believed the same about complexity.
Our finding that e-RAs achieve the best savings for simple
products appears to undermine those who suggest the higher
margins in complex purchases might allow for greater savings. We
note, however, that our data included relatively few examples of
complex and very well-specified products. Complexity and
specification are two separate measures of supplier uncertainty,
strongly correlated with each other, and it may be that success is
greater when both measures are satisfied. Further review of e-RAs
for a larger number of complex and very well-specified products may
demonstrate higher success, in part from the greater potential for
margin reduction as compared to simple, commoditized products,
and ,indirectly, by attracting more suppliers by being well-specified.
Nonetheless, our findings clearly indicate that public sector buyers
should have confidence in using e-RAs for simple, easily specified
purchases.

Number of Bidders
Our findings did not show a correlation between success and
number of bidders. Unlike the recommendations in the literature
against conducting e-RAs with few bidders, we found no evidence that
public sector buyers should be concerned with the number of bidders.
Some buyers even observed that strong competition among two
suppliers may be all that is needed to hold a successful e-RA.
We also note however that the number of bidders is strongly
correlated with competition, and may be a factor which affects the
degree of competition among suppliers in the market.

Auction Volume
Auction volume was not found to correlate with any variable,
including success. The literature suggested that higher auction
volume would directly affect success, with higher volume leading to
higher savings through economies of scale and lower transaction
costs. However, our findings suggested that such savings, if any, were
not being passed on to buyers. Some buyers speculated that in some
cases, experienced suppliers were able to calculate their internal

costs across differing auction volume amounts accurately and

maintained their profit margins accordingly.

Competition
Competition was found to be correlated with all factors except
auction volume.

Number of Bidders
As widely observed in the literature, our findings showed that
competition correlated most strongly with the number of bidders,
although causation is unclear. While it was assumed that more
suppliers would lead to higher competition and greater success, no
direct correlation was found between the number of bidders and
success.

Specification and Complexity

The correlation with specification and complexity is thought to
result from the greater certainty given to suppliers by better quality
specifications or lower differentiation. As a result, suppliers are
better able to assess their margins, and better able to react to bids
from their competitors.

Auction Volume
Auction volume did not correlate with competition, suggesting
that auction volume does not promote additional competition. Our
data included a wide range of products, and the degree of
competition in each market varied significantly. Further research on
this issue should be conducted to determine whether these findings
remain the same when purchasing the same product for both small
and high volumes. Keeping competitive factors constant, increasing
auction volume may further increase competition.

Number of Bidders
The number of bidders was found to be correlated with
competition, specification and complexity, but not with auction
volume or success.

Complexity
As anticipated by the literature, our findings appear to indicate
that simple purchases had more participating suppliers. However, it
is unclear whether simpler purchases cause more suppliers to attend,
whether there are simply more suppliers in the marketplace for
simple purchases, or whether both are true.
In order to address this issue, it may be useful to conduct further
research, such as measuring the proportion of the number of
suppliers participating against the total suppliers in the market for
each purchase. For example, it may be that very complex purchases
have four out of five suppliers in the market bidding, while for
commodities, ten out of fifty suppliers participate. While this appears
to show that simple purchases attract more bidders, in relative terms,
complex purchases may have very high participation rates among
very few total suppliers. Such research will assist buyers in
determining how changes in complexity affect the number of
participating bidders. We note, however, that since complexity is also
negatively correlated with competition, and competition is correlated
with number of bidders, it may be that the absolute number of
bidders, rather than the proportion of bidders participating, has the
greatest effect on e-RA success.

Specification
As expected, improving the specification appeared to increase the
number of participating suppliers. The literature suggests that this is
a result of reduced supplier uncertainty, which enabled them to better
assess their lowest price and gave them more confidence to
participate.

Auction Volume
Perhaps most surprising is that no correlation was found between
auction volume and the number of bidders. The literature
recommends increasing auction volume to attract additional
suppliers; however, the relationship is either very weak, or perhaps
indirect. Buyers noted that differing auction volumes may attract
different suppliers. For example, smaller auctions may attract
smaller businesses, whereas very large auctions may be dominated
by large, multinational companies. As a result, the number of bidders

does not correlate with an increase in auction volume. Buyers also

noted that in some cases, where there were a limited number of
bidders in the market, all suppliers participated, regardless of the
amount. This was particularly likely with complex purchases
containing a high degree of intellectual property.

Specification and Complexity

Specification was found to strongly correlate with complexity. This
suggests that simple products are more easily accurately specified,
whereas complex purchases were more difficult to specify well. In our
interviews we noted that buyers felt that simple purchases were
almost always well-specified.
In our research we measured the effect of specification against
complexity based on the buyer’s view of how well he or she (or their
requisitioner) specified the purchase. We noted that one defense
department buyer indicated that his extensive procurement
experience in defense-related purchases ensured that he was always
able to well-specify both simple and complex purchases. Future
researchers may wish to survey suppliers for their views on
specification and complexity, first to verify the strong correlation of
these factors, and second, to determine how their perception of the
specification and complexity affected their willingness to participate
and their pricing strategy.

CONCLUSIONS
This paper examined the factors which could affect the success of
e-RAs in the public sector. Our findings indicated that competition
among suppliers, the complexity of the purchase, and how well the
purchase was specified were the factors which were most strongly
correlated with e-RA success. We also note that neither the number of
bidders nor the auction volume should be a determining factor for
whether public sector buyers should use e-RAs. Our findings
indicated that success was possible regardless of the number of
bidders, provided that other factors such as competition, complexity
and specification were favorable. Similarly, the auction volume in
itself had no apparent effect on success, or on any other factor. Our
findings on the number of bidders and auction volume differ from the
general recommendations in the literature against conducting e-RAs
with few suppliers or with low auction volume.

Competition among suppliers is the single most important factor

in achieving successful e-RAs. Buyers would be well-advised to
assess the competition among suppliers in each purchase-specific
market when considering whether to hold an e-RA.
Our findings indicate that the more well-specified the purchase,
the greater the success will be. While simple purchases were found
to be better specified, and correlated with greater success, we also
note that this may be more a function of the lack of well-specified,
complex purchases. Nonetheless, buyers new to e-RAs may wish to
start with simple, easily specified purchases initially, as both the
literature and our findings support such an approach.
Public sector buyers should also note that concerns about e-RAs
creating distrust between buyers and suppliers and the occurrence of
switching costs are not relevant considerations.
Several limitations in our research are worth noting. First, the
sample data reviewed in this study was taken from large public sector
organizations with significant purchasing power. Future studies may
wish to consider whether differences exist among smaller
government agencies, public corporations, and whether the nature of
the buyer affects the relative importance of the factors studied.
Second, the heterogeneity of purchases may mask whether certain
goods or services are particularly ill-suited for e-RAs. Finally, our
statistical model emphasized success as defined by cost savings
alone. However, public sector buyers are often required to consider
many other factors when making their purchasing decisions.
Measuring savings alone does not adequately capture product
quality, procurement cycle times, or audit or compliance possibilities,
all of which may be more important than cost alone. Subsequent
research may seek to measure these variables against the factors
described herein.

NOTE
The views presented in this paper are those of the authors and do
not necessarily reflect the views of the United Nations or any other
organisation.

REFERENCES
Alilovic, M. (2008). “E-Procurement Impact on Corruption: E-GP
Scenario/Model in Indonesia.” In Asian Development
Bank/Organisation for Economic Co-operation and Development
(Ed.), Fighting Bribery in Public Procurement in Asia and the
Pacific (Vol. 1, pp. 129-135). Manila, Philippines: Asian
Development Bank.
Beall, S. et al. (2003). “The Role of Reverse Auctions in Strategic
Sourcing.” [On-line]. Available at http://lilgerry.com/
beall2003ecom.pdf. [Retrieved March 23, 2010]
Carter, C. R., Kaufmann, L., Beall, S., Carter, P. L., Hendrick, T. E., &
Petersen, K. J. (2004). “Reverse auctions — Grounded Theory
from the Buyer and Supplier Perspective.” Transportation
Research, 40 (3): 229−254.
Crane, L. S. (2008). “Seven Questions to Ask Before Running Your
Next Reverse Auction.” Purchasing. [On-line]. Available at
www.purchasing.com/article/CA6537987.html?q=reverse+auctio
n. [Retrieved March 23, 2010]
De Boer, L., Harink, J., & Heijboer, G. (2002). “A Conceptual Model for
Assessing the Impact of Electronic Procurement.” European
Journal of Purchasing and Supply Management, 8 (1): 25–33.
Emiliani, M. L., & Stec, D. J. (2002). “Squaring Online Reverse
Auctions with the Caux Round Table Principles for Business.”
Supply Chain Management, 7 (2): 92-100.
Emiliani, M. L., & Stec, D. J. (2005). “Commentary on ‘Reverse
Auction for Relationship Marketers’ by Daly and Nath.” Industrial
Marketing Management, 34 (2): 167-171.
“Fighting Cartels in Public Procurement.” (2008). OECD Policy Brief.
[On-line]. Available at www.oecd.org/dataoecd/45/63/
41505296.pdf. [Retrieved May 24, 2010]
Gattiker, T. F. (2008). “Internet Reverse Auctions: Listening to the
Voices of the Non-Adopters.” In D. Parente (Ed.), Best Practices
for Online Procurement Auctions (Vol. 1, pp. 43-56). Harrisburg,
PA: Idea Group.
Hartely, L. J., Lane, D. M., & Duplaga, E. (2006). “Exploring the
Barriers to the Adoption of e-auctions for Sourcing.” International

Journal of Operations & Production Management, 26 (2): 202-

221.
Jap, S. D. (2002). “Online Reverse Auctions: Issues, Themes, and
Prospects in the Future.” Journal of the Academy of Marketing
Science, 30 (4): 506-525.
Kaufmann, L., & Carter, C. R. (2004). “Deciding on the Mode of
Negotiation: To Auction or Not to Auction Electronically.” The
Journal of Supply Chain Management, 40 (2): 15-26.
Lindskog, H. (2008). “E-procurement of Telecom Services for the
Public Sector.” International Journal of Knowledge Management
Studies, 2 (1): 17-28.
Loesch, A., & Lambert, J. S. (2007). “E-Reverse Auctions Revisited: An
Analysis of Their Context, Buyer-Supplier Relations, and
Information Behaviour.” Journal of Supply Chain Management,
43(4): 47–63.
Mabert, V. A., & Skeels, J. A. (2002). “Internet Reverse Auctions:
Valuable Tool in Experienced Hands.” Business Horizons, 45 (4):
70–76.
Major, C. (2007). “Reverse Auctions – A Suitable Procurement Tool
for the WA Public Sector?” [On-line]. Available at
www.dtf.wa.gov.au/cms/uploadedFiles/CorneliaMajor_Reverse_A
uction-A_Suitable_Procurement_Tool_for_the_WA_Public_Sector.
[Retrieved March 23, 2010].
“NSW Government Procurement Guidelines - Reverse- Auctions”
(2006). [On-line]. Available at www.nswprocurement.com.au/
PDF/Policy/Reverse-Auctions.aspx. [Retrieved March 23, 2010].
Porter, M. E. (1980). Competitive Strategy: Techniques for Analyzing
Industries and Competitors. Free Press, New York.
“Reverse Auction Best Practices: Practical Approaches to Ensure
Successful Electronic Reverse Auction Events.” (2006). SAP White
Paper. [On-line]. Available at www.sdn.sap.com/irj/scn/go/
portal/prtroot/docs/library/uuid/b018b90a-b9e5-2910-c6bd-
f8f7bc3a34a9?QuickLink=index&overridelayout=true. [Retrieved
March 23, 2010].

Sashi, C. M., & O'Leary, B. (2002). “The Role of Internet Auctions in

the Expansion of B2B Markets.” Industrial Marketing
Management, 31 (2): 103-110.
Sawheny, M. (2003). “Reverse Auctions Cutting Costs.” [On-line].
Available at http://www.cio.com/article/29717/Reverse_
Auctions_Cutting_Costs. [Retrieved March 23, 2010].
Schapper, P. (2008). “The Impact of e-Procurement on Corruption –
The Potential of e-Procurement for Curbing Corruption Risks.” In
Asian Development Bank/Organisation for Economic Co-operation
and Development (Ed.), Fighting Bribery in Public Procurement in
Asia and the Pacific (Vol. 1, pp. 113-127). Manila, Philippines:
Asian Development Bank.
Schoenherr, T., & Mabert, V. A. (2007). “Online Reverse Auctions:
Common Myths Versus Evolving Reality.” Business Horizons, 50:
373–384.
Schoenherr, T., & Mabert, V. A. (2008). “The Use of Bundling in B2B
Online Reverse Auctions.” Journal of Operations Management,
26: 81-95.
Singer, M., Konstantinidis, G., Roubik, E., & Beffermann, E. (2009).
“Does e-Procurement Save the State Money?” Journal of Public
Procurement, 9 (1): 58-78.
Smeltzer, L. R., & Carr, A. (2003). “Electronic Reverse Auctions-
Promises, Risks and Conditions for Success.” Industrial Marketing
Management, 32 (6): 481-488.
Turley, S. L. (2002). “Wielding the Virtual Gavel - DOD Moves Forward
With Reverse Auctions.” Military Law Review, 173: 1-67.
United-Kingdom Office of Government Commerce (2005). OGC
Desktop Guide to e-Procurement, Part 3: What Are e-Auctions....
and How to Get Started. [On-line]. Available at www.localtgov.org.
uk/webfiles/NePP/Guidance/8.0%20e-Sourcing/8.1.2.pdf. [Re-
trieved March 23, 2010].
United-Kingdom Office of Government Commerce (2010, March).
OGC announces e-auction strategy to save £270m. [Press
Release] [On-line]. Available at www.ogc.gov.uk/About_OGC_
news_10098.asp. [Retrieved March 23, 2010].

Wagner, S. M., & Schwab, P. A. (2004). “Setting the Stage for

Successful Electronic Reverse Auctions.” Journal of Purchasing &
Supply Management, 10 (1): 11–26.
Walker, H., & Harland, C. (2008). “E-Procurement in the United
Nations: Influences, Issues and Impact.” International Journal of
Operations & Production Management, 28 (9): 831-857.
Yukins, R. C., & Wallace, D. (2005). “UNCITRAL Considers Electronic
Reverse Auctions.” Public Procurement Law Review, 4: 183-202.

Electronic Reverse Auction and The Public Sector: Factors of Success Moshe E. Shalev & Stee Asbjorensen

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Electronic Reverse Auction and The Public Sector: Factors of Success Moshe E. Shalev & Stee Asbjorensen

Uploaded by

Copyright:

Available Formats

JOURNAL OF PUBLIC PROCUREMENT, VOLUME 10, ISSUE 3, 428-452 FALL 2010

ELECTRONIC REVERSE AUCTIONS AND THE PUBLIC SECTOR:

ABSTRACT. Electronic reverse auctions (e-RAs) provide purchasers with an

Copyright © 2010 by PrAcademics Press

and many developing country governments such as Brazil, Chile,

specified. The final category relates to factors that resulted from

Second, e-RAs provide a number of administrative benefits. They

The Disadvantages of e-RAs

Most researchers have suggested that at least three to five viable

Despite this, the literature cites examples of smaller auctions that

Competition among Suppliers

Purchase Complexity and Specification

potential savings (Smeltzer & Carr, 2003; Beall et al., 2003;

Relationship with the Supplier

Relevance to the Public Sector

Competition among Suppliers

specific supplier base: the market share concentration of suppliers,

Complexity and Specification

DISCUSSION AND RECOMMENDATIONS

139 e-RAs found a number of similarities with the private sector

specification had either an indirect or spurious effect on success also

costs across differing auction volume amounts accurately and

Specification and Complexity

does not correlate with an increase in auction volume. Buyers also

Specification and Complexity

Competition among suppliers is the single most important factor

Journal of Operations & Production Management, 26 (2): 202-

Sashi, C. M., & O'Leary, B. (2002). “The Role of Internet Auctions in

Wagner, S. M., & Schwab, P. A. (2004). “Setting the Stage for

You might also like