This is the author’s version of a work that was submitted/accepted for pub-lication in the following source:
Teo, Pauline,Bridge, Adrian, & Love, Peter (2015)
Deriving optimal competition in infrastructure procurement. In
2015 International Conference on Construction and Real Estate Manage-ment (ICCREM), 11-12 August 2015, Luleå, Sweden.
This file was downloaded from: http://eprints.qut.edu.au/91095/
c
Copyright 2015 American Society of Civil Engineers
Notice: Changes introduced as a result of publishing processes such as copy-editing and formatting may not be reflected in this document. For a definitive version of this work, please refer to the published source:
Deriving optimal competition in infrastructure procurement Pauline TEO1, Adrian BRIDGE2 and Peter LOVE3
1
Research Fellow, Dept. of Civil Engineering, Curtin University, GPO BOX U1987, Perth WA 6845, Australia, Email: [email protected]
2
Associate Professor, School of Civil Engineering & Built Environment, Queensland University of Technology, QLD 4000, Australia; Email: [email protected]
3
John Curtin Distinguished Professor, Dept. of Civil Engineering, Curtin University, GPO BOX U1987, Perth WA 6845, Australia, Email: [email protected]
ABSTRACT
Typically only a limited number of consortiums are able to competitively bid for Public Private Partnership (PPP) projects. Consequently, this may lead to oligopoly pricing constraints and ineffective competition, thus engendering ex ante market failure. In addressing this issue, this paper aims to determine the optimal number of bidders required to ensure a healthy level of competition is available to procure major infrastructure projects. The theories of Structure-Conduct-Performance (SCP) paradigm; Game Theory and Auction Theory and Transaction Cost Economics are reviewed and discussed and used to produce an optimal level of competition for major infrastructure procurement, that prevents market failure ex ante (lack of competition) and market failure ex post (due to asymmetric lock-in).
INTRODUCTION
Public-Private Partnerships (PPPs) have been widely implemented on the premise of delivering value-for-money (VfM) in the procurement of major infrastructure. However, due to the financial and technical prequalification requirements, PPPs often operate in oligopolistic or monopolistic markets, in which there are a limited number of bidders (De Valence 2003, 2012; Thu & Akintoye 2007). In a review of studies on market share in the construction industry, De Valence (2012, p. 31) found that in countries such as Australia, South Korea, Japan, Hong Kong and the United Kingdom (UK), there were a plethora of small firms whereby their markets exhibited the features of perfect competition and a small number of large contractors, and in some instances accounted for up to 70% of turnover, which mimicked the characteristics of an oligopoly.
More specifically, De Valence (2012) concluded that large head contractors in the engineering construction and non-residential building sectors, and subcontractors
that supply lifts and building automation systems exhibit characteristics of oligopolistic competition. Furthermore, Thu and Akintoye (2007) examined the four-firm concentration ratio (CR4) present within the PPP/Private Finance Initiative (PFI) UK market as fifty percent or more in most subsectors such as transport and health sectors. The CR4 is a common tool for analysing the types of market structure and can be used as a measure of competition (Scherer & Ross 1990). The CR4 is computed as the percentage of market share held by the four largest firms in the industry. The results show that almost all sub-sectors within the UK PPP/PFI construction market can be considered medium concentration or an oligopoly market. As a result this may lead to oligopoly pricing constraints and competition. In accordance with Neoclassical Economics, non-competitive markets with limited competitors can lead to an unbalanced allocation of goods and services, which is referred to as Pareto inefficiency (Kaufman 2011). Competition drives down marginal costs and creates downward pressure on price, as well as facilitating innovations in design, based on outcome-based specifications that impinge on the time, cost and quality of the project, and which influence the overall performance of the project across its life-cycle. With lower competitive tension in the market, there is a lack of incentives for contractors to be innovative in order to be price competitive, and which can affect the performance across the whole-of-life of the asset. On the other hand, empirical evidence from construction procurement shows that a high level of competition or aggressive bidding is associated with opportunistic behaviour in complex projects or market failure post-contract (Sweeney 2009).
This paper aims to derive an optimal level of competition in the procurement of major infrastructure from the literature, that is, a lower limit of competition level that avoids ineffective competition ex ante and at the same time, an upper limit of competition that avoids negative opportunistic behaviour ex post. This is achieved by reviewing and integrating literature related to competition; the Structure-Conduct-Performance (SCP) paradigm; tendering perspectives based on Game Theory and Auction Theory; and negative opportunistic behaviour within the lens of Transaction Cost Economics.
STRUCTURE-CONDUCT-PERFORMANCE
The notion that competition can be driven by a small number of firms (oligopoly) originated from Chamberlin’s (1933) seminal work in formulating the theory of monopolistic competition. Also closely related to Chamberlin’s work, Clark (1940) developed the concept of “workable competition” or optimal competition that seeks to avoid the extreme forms of imperfect competition, which is either too weak or strong. Clark (1940) further describes workability as somewhere between “pure oligopoly, and the ruinously low prices likely to result from unlimited market chaos: more strongly competitive than the first, and more workable than the second” (p.253). The latter scenario often known as ‘destructive, ruinous, chaotic or cut-throat competition’ can occur in an oligopoly where the overhead costs are high and marginal costs are low. During cyclical or stand-by excess capacity in which demands fall below full-capacity levels, firms are willing to cut price well below an
efficient producer's total cost of production. This scenario can be a chronic issue especially in industries exhibiting characteristics of imperfect immobility, such as bituminous coal mining and agriculture, and Clark (1940) termed these industries as “sick industries”. On the other hand, Clark (1940) observed “some of the healthiest cases of workable competition in large-scale industry” in oligopolistic industry, such as automobile production (p. 256). This provides evidence that effective competition can exist amongst a relatively small number of firms.
Joe S. Bain (1950) takes Clark’s argument further to the industry level and explains the relationship between the patterns of competitive behaviour and oligopoly market structure in terms of conditions of entry and market concentration. Based on Bain’s prognosis (p. 46), oligopolies with moderate difficulties to entry and concentration of sellers, generally display the closest approximation to workability. These oligopolies display reasonable efficiency, with low or moderate prices and profits. Essentially, under difficult and moderate market entry conditions, established firms will set prices low enough to prevent new entry and maintain reasonable efficiency in scale and capacity in the long-run, whereas the prospect of long-run efficiency is less certain under easy entry (p. 42). In terms of market concentration, ceteris paribus , a large proportion of the market controlled by a “very few large firms and smaller firms are absent or small in number” (p. 43), that is high market concentration, do not display workability; whereas, a moderate proportion of the market controlled by a similar number of large firms, and a greater number of small to medium sized firms holding the balance of the market share, that is moderate market concentration may be argued to display quasi competitive behaviour, such as imperfect collusion or kinked demand curve conformations. Bain Joe Staten Bain (1951, p. 44) proposes eight-firm market concentration of 65 to 75 percent as a dividing line to differentiate between effective and ineffective competition in loose and tight oligopoly respectively.
Market structures can be classified based on the degree of competitiveness in the market; using information such as market share (Shepherd 1982). Shepherd (1990, p. 14) classifies six types of market structures into two categories of competition: 1) ineffective and 2) effective. Market structures under ineffective competition are characterised by high market concentration, such as: (1) monopoly (one firm has 100 percent market share); (2) dominant firm (one firm has 40 percent to 99 percent market share); and (3) tight oligopoly (four firms have over 60 percent market share). Under effective competition, firms are not able to control prices, little collusion and low profit rates (Shepherd 1982, p. 617). There are three market types under effective competition: (1) loose oligopoly (four firms have less than 40 percent market share): (2) monopolistic competition (many competitors each with a slight degree of market power); and (3) pure competition (many competitors, none of whom has market power). Based on the above, firms that operate in an oligopoly market can be in healthy competition. This has implications on infrastructure procurement, in terms of the extent of bundling and sizing of infrastructure projects, in order to generate effective competition within the limits of optimal competition.
Next, given that sealed bid auctions aim to increase competitive tension in the procurement process, the following section studies auction theory or game theory literature related to competitive bidding and the effects of the number of bidders on value of bids.
AUCTION THEORY AND GAME THEORY
In the study of auctions and bids, there are two dominant approaches based on the assumptions of maximising expected profits: (1) decision-theoretic approach (Carr 1982; Friedman 1956; Gates 1967; Skitmore 1991, 2013) and the (2) game theoretic approach (Engelbrecht-Wiggans, Milgrom, & Weber 1983; Milgrom 1989; Milgrom & Weber 1982; Riley & Samuelson 1981; Rothkopfm 1969; Vickrey 1961; Wilson 1977). The decision-theoretic approach determines a satisfactory balance between the probability of winning a contract and the profit generated as a result of winning the contract. It is assumed that the bidders compete by applying a strategic mark-up, that is, a percentage mark-up to the estimated true cost of the project to form the bid value. The focus is on forecasting the distribution of bids, and determining optimal bidding by “the product of the probability of winning the contract and the profit generated as a result of winning the contract (Skitmore & Pemberton 1994, p. 1263).” The other approach – game theoretic approach – is based on the study of the Bayesian-Nash equilibrium, by modelling sealed bid auctions as non-cooperative games with incomplete information, and derives equilibrium bidding strategies with the assumptions that all bidders bid optimally. Game theorists have shown that aggressive bidding is associated with increasing number of bidders and reducing the value of bids (e.g., Bajari 2001; Bajari, Houghton, & Tadelis 2014) and similarly, empirical studies on tendering in the construction industry show a correlation between a greater number of bidders and a reduction in the value of the lowest bid, and which is consistent with neoclassical economics (e.g., Brannman, Klein, & Weiss 1987; Domberger & Rimmer 1994; McCaffer 1979; Ngai, Drew, Lo, & Skitmore 2002; Skitmore 2002).
In terms of optimal competition, Chamberlin (1933) did not specify what is considered to be a “small” and “large” number of firms. Within the game theory literature however, the focus has been on determining the lower limit of competition, that is, the minimum number of firms needed to generate effective competition. Selten (1973) was first to demonstrate that five competitors represent the dividing line between few and many, when modelled as moves in a non-cooperative game. Subsequent to Selten’s influential work, Kagel and Levin’s (1986) experiments indicate that auctions with 6 to 7 bidders (many) generate greater competition than with 3 to 4 bidders (small). Based on a review of empirical research and experiments, Huck, Normann, and Oechssler (2004) report that 4 and more firms are never collusive (either competitive or at Nash equilibrium); 2 is highly collusive, and 3 has reached Nash equilibrium. The literature has indicated that a minimum of 5 competitors is enough to exert high competitive pressure in the market (e.g., Huck et al. 2004; Kahn 2006; Selten 1973). As there are increased opportunities for interactions amongst firms (multi-market contacts) with repeated games or auctions,
this may create opportunities for collusion. In other words, game theorists argue that cooperation or collusive behaviour is considered difficult to sustain at 5 or more competitors, even in repeated games (Huck et al. 2004, p. 436). As such, this demonstrates that when four or less firms compete in a market, it creates tight oligopoly market conditions and associated pricing constraints, along with ineffective competition (Beattie, Goodacre, & Fearnley 2003; Shepherd 1982). More importantly, five competitors or bidders are derived as the lower limit of optimal competition (Huck et al. 2004; Kagel & Levin 1986; Selten 1973).
On the other hand, game theorists have also indicated the idea of upper limit of optimal competition, that is, perfect competition or intense competition leads to the convergence of the true value of the object, even when bidders have incomplete information about its value (Milgrom 1979; Wilson 1977). Capen, Clapp, and Campbell (1971) explain that when the potential competition is large, the winning firm is likely to have underestimated the value and as a result suffers from negative profits. This phenomenon is known as the winner’s curse and mainly occurs in common-value settings 1 (such as bridge construction) (Bajari & Hortacsu 2003; Hong & Shum 2002; Kagel & Levin 1986; Thaler 1988; Thiel 1988). In common value settings, bidders only have estimates of the project, and rational firms will not bid as aggressively when there are many potential competitors (Barrus & Scott 2013, p. 6). In other words, having a large or excessive number of bidders does not necessarily lead to price reduction in common value auctions. As the final bid price is an optimal decision that maximises the bidder’s profits and the chance of winning the bid, a larger number of bidders may result in more aggressive bidding behaviour, but not at the expense of negative profits. As bidders learn out of “a trial and error survival process that is situationally specific” from previous tenders (Kagel & Levin 1986, p. 901), and gain sufficient experience and feedback regarding the outcomes of their previous decisions, bidders will learn to avoid the winner's curse. With adequate time, this disequilibrium will correct itself given period (Kagel & Levin 1986, p. 917). More importantly, game and auction theory have indicated an upper limit to competition, beyond which bid value reaches a constant.
TRANSACTION COST ECONOMICS
In addition to the lack of competition as a factor of market failure, Transaction Cost Economics (TCE) also envisages market failure occurring ex post due to post-contractual opportunistic behaviour. For instance, asset specificity in construction projects tends to be higher post-contract when the investment has been made, and the
1
Auctions are categorised into private value and common value auctions. In private value auctions, bidders know the value of the item (asphalt projects), whereas in common value auctions, the bidders only have the estimates of the value of the item (such as bridge projects) (De Silva, Jeitschko, & Kosmopoulou 2009; Hong & Shum 2002).
5
costs of replacing the main-contractor, subcontractor, or supplier are higher both financially and in terms of the implications for project progress. This is in contrast to pre-contract when there are plenty of alternative main-contractors, subcontractors, or suppliers in the market (Winch 2001). This change to a form of bilateral monopoly post-contract is known as ‘fundamental transformation’ (Williamson 1985). The more plentiful supply of bidders – the more stark the fundamental transition – and it is logical to expect, on average, more bidders in projects that are seen as attractive by contractors due to their assessment of the potential for variations, for example arising from incomplete documentation including lack of clarity in specifying requirements. Projects that are subjected to the dynamics of change and are locked into long-term contracts, such as an extremely complex PPP, create opportunities for post-contractual negative opportunistic behaviour. This can open up opportunities for hold-up ex post. For example, Sweeney examines the major infrastructure market in Australia and observes that contractors can use low bid tendering strategies to win contracts in an extremely competitive market, and adopt a claim strategy resulting in high variation claims ex post (Sweeney 2009, p. 141). Amaral, Saussier, and Yvrande-Billion (2013) also point out that bidders may bid more aggressively to win contracts with a high potential to be renegotiated post-contract, given the chance to renegotiate after winning the contract. This shows that the current contracting practice could have resulted in ex post market failure, as opportunistic behaviour is not uncommon in the construction industry, especially for complex projects associated with high uncertainty, and the market can perceive the project to be lucrative in terms of a potential for variation claims, and the opportunity to make superior profits post-contract (Sweeney 2009).
Furthermore, research in construction procurement has provided empirical evidence of an upper limit of competition. Gupta (2002) examined the effects of the number of bidders on bid value based on 1740 highway construction projects in the US over a five year period. The empirical results indicate that the absolute level of competition creates downward pressure on price. That is, the value of winning bids decreases as the number of bidders increases, but the effect on value becomes insignificant when the absolute number of bidders reaches a maximum number; in other words, a competitive threshold is reached. Gupta determines this competitive threshold to be near and up to 8 bidders in an open tender (or equivalent expressions of interest). Thus, Gupta (2002) has made the important contribution in terms of surfacing an optimal upper level of competition from a production costs and benefits perspective. Also highly relevant, Skitmore (2002) analysed ten data sets (representing 1,234 projects), this time in a different sector than Gupta and mainly from the building industries in various countries, including US, UK and Belgium. Skitmore’s (2002) findings are consistent with Gupta’s study, where the regression curves show that the value of the lowest bid decreases until eight bidders, and remains constant as the number of bidders or competition increases. Skitmore (2002, p.16) concludes that there is a tendency for the value of bid to be significantly higher than the pretender estimate when the competition is below eight bidders; conversely, the value of bid comes close to the pretender estimate or, in some instances, is lower than the pretender estimate when the competition is greater than eight bidders.
Since production costs and benefits ratio has been shown empirically to become negligible greater than eight bidders, then this upper limit can be used to assess the efficacy of the project’s specified requirements. In so far as, when tenders attract unusually strong interest beyond eight bidders, then this can be an indication of a lack of specified requirements and the likelihood of variations leading to ex post market failure. Moreover, Skitmore’s (2002) and Gupta’s (2002) empirical evidence establishes the validity of using competition as an indicator of testing the goodness of a model in terms of setting a project on a path towards delivering superior VfM. On this basis, a number of bidders in the range 5 to 8 inclusive is established as a suitable dependent variable and proxy for VfM; and an optimal response and reflective of the extent to which the procurement of the project has been optimised in relation to size, bundling and exchange relationship (Teo 2014).
In terms of a measure of competition, the extent of anticipated or expected competition, rather than the actual competition, is more likely to be a determinant of a firm’s bidding strategy in a sealed bid auction. As firms do not know the actual number of bidders at the time of preparation of bids in practice, studies on auctions have been criticised for assuming that the number of bidders is known (Amaral et al. 2013). Ngai et al. (2002, p. 475) propose that potential competition is dependent on market conditions; a reflection of supply capacity utilization in the industry; and can be measured as the “likely number of competitors for projects in the market”. This is affirmed by the empirical studies by De Silva et al. (2009) in road construction procurement and Amaral et al. (2013) in London bus transport; where their statistical analyses indicate that the expected number of bidders is more effective than the actual number of bidders in explaining the value of the bids; and bidders appear to be more aggressive when the number of expected number of bidders is large. Similarly, in terms of PPP procurement, and Teo (2014); Teo, Bridge, and Gray (2013) use the number of Expressions of Interest (EoI) to demonstrate the contractor’s willingness to bid as a more relevant measure of expected competition. As such, 5 to 8 EoI is derived as the optimal level of competition.
CONCLUSION
To conclude, a PPP project needs to be sufficiently large to justify leveraging private finance but, on the other hand, not too large so as to yield insufficient competition, or ex ante market failure arising from small bidding numbers, or conditions akin to monopoly supply. At the low level of potential competition or low EoI, SCP and game theory have indicated that, 4 or less firms demonstrating their willingness to bid for a project creates tight oligopoly conditions and associated pricing constraints, along with ineffective competition. As such, 4 or less EoI can be used as an early indication of market failure ex ante arising from a lack of competition associated with size or level of bundling of the project. At the same time, a balance needs to be struck, as too much competition becomes counter-productive and can be an indicator of market failure ex post. This is because the market can be signalling the prospect of gains arising from poorly specified requirements or a lack of predictable requirements, which leads to hold-up arising from variations in the
PPP long-term contract. TCE has indicated that high number of bidders or EoIs (over 8), can be seen as an early indication of market failure ex post arising from the prospect of lack of flexibility or possibility of renegotiation;
As such, 5 to 8 (inclusive) EoI is derived as an optimal level of competition, and an early indication of avoiding market failure ex ante arising from a lack of competition; and ex post arising from a lack of flexibility. It is expected that projects within 5-8 EoI are potentially on a path to superior VfM, whereas projects with sub-optimal EoI are expected to have room for VfM improvements in their procurement dimensions. 5 to 8 EoI is derived as a valid and early indicator of the optimal configuration of a project’s key procurement dimensions (including a PPP or non-PPP approach). Further quantitative and empirical research can be carried out to test the reliability of 5-8 EoIs in generating an optimal level of competition in the procurement of major infrastructure.
REFERENCES
Amaral, M., Saussier, S., & Yvrande-Billion, A. (2013). Expected Number of Bidders and Winning Bids: Evidence from the London Bus Tendering Model. Journal of Transport and Economic Policy, 47(1), 17-34.
Bain, J. S. (1950). Workable Competition in Oligopoly: Theoretical Considerations and Some Empirical Evidence. The American Economic Review, 40(2), 35-47. doi: 10.2307/1818021
Bain, J. S. (1951). Relation of Profit Rate to Industry Concentration: American Manufacturing, 1936-1940. The Quarterly Journal of Economics, 65(3), 293-324.
Bajari, P. (2001). Comparing competition and collusion: a numerical approach. Economic Theory, 18(1), 187-205.
Bajari, P., & Hortacsu, A. (2003). The Winner's Curse, Reserve Prices, and Endogenous Entry: Empirical Insights from eBay Auctions. The RAND Journal of Economics, 34(2), 329-355.
Bajari, P., Houghton, S., & Tadelis, S. (2014). Bidding for Incomplete Contracts: An Empirical Analysis of Adaptation Costs. American Economic Review, 104(4), 1288-1319.
Barrus, D., & Scott, F. (2013). Single Bidders and Tacit Collusion in Highway Procurement Auction. Retrieved 25 January 2015, 2015, from http://users.uoa.gr/~gdellis/Dimosies_Symvaseis/Single_Bidders_Tacit_Coll usion.pdf
Beattie, V., Goodacre, A., & Fearnley, S. (2003). And then there were four: A study of UK audit market concentration — causes, consequences and the scope for market adjustment Journal of Financial Regulation and Compliance, 11(3), 250-265.
Brannman, L., Klein, J. D., & Weiss, L. W. (1987). The price effects of increased competition in auction markets. The Review of Economics and Statistics, 69(1), 24-32.
Capen, E. C., Clapp, R. V., & Campbell, W. M. (1971). Competitive bidding in high-risk situations. Journal of Petroleum Technology, 23(1), 641-653.
Carr, R. I. (1982). General bidding model. Journal of the Construction Division, Proceedings of the American Society of Civil Engineers, 108(CO4), 639-650.
Chamberlin, E. (1933). The Theory of Monopolistic Competition: A 'Re-orientation of the Theory of Value' (8 ed.). Cambridge, Massachusetts: Harvard University Press.
Clark, J. M. (1940). Toward a Concept of Workable Competition. The American Economic Review, 30(2), 241-256. doi: 10.2307/1807048
De Silva, D. G., Jeitschko, T. D., & Kosmopoulou, G. (2009). Entry and Bidding in Common and Private Value Auctions with an Unknown Number of Rivals. Review of Industrial Organization, 35(1-2), 73-93. doi: 10.1007/s11151-009-9223-2
De Valence, G. (2003). Market Structure, Barriers to Entry and Competition in Construction Markets. Paper presented at the CIB W55/W65/W107 Conference.
De Valence, G. (2012). Project Finance Model for Small Contractors in USA. The Australian Journal of Construction Economics and Building, 7(1), 29-36. Domberger, S., & Rimmer, S. (1994). Competitive tendering and contracting in the
public sector: A survey. International Journal of the Economics of Business, 1(3), 439-453. doi: 10.1080/758536232
Engelbrecht-Wiggans, R., Milgrom, P., & Weber, R. (1983). Competitive bidding and proprietary information. Journal of Mathematical Economics, 11(1), 161-169.
Friedman, L. (1956). A Competitive-Bidding Strategy. Operations Research, 4(1), 104-112. doi: 10.2307/167522
Gates, M. (1967). Bidding strategies and probabilities. Journal of the Construction Division, Proceedings of the American Society of Civil Engineers, 93(CO1), 75-107.
Gupta, S. (2002). Competition and collusion in a government procurement auction market. Atlantic Economic Journal, 30(1), 13-25.
Hong, H., & Shum, M. (2002). Increasing competition and the winner's curse: Evidence from procurement. The Review of Economic Studies, 69(1), 871-898.
Huck, S., Normann, H.-T., & Oechssler, J. (2004). Two are few and four are many: number effects in experimental oligopolies. Journal of Economic Behavior & Organization, 53(4), 435-446. doi: 10.1016/j.jebo.2002.10.002
Kagel, J. H., & Levin, D. (1986). The Winner's Curse and Public Information in Common Value Auctions. The American Economic Review, 76(5), 894-920. Kahn, A. E. (2006). Telecommunications: The transition from regulation to antitrust.
Journal on Telecommunications and High Technology Law, 5(1), 159-188. Kaufman, B. E. (2011). Economic Analysis of Labor Markets and Labor Law: An
Institu tional/Industrial Relations Perspective. Andrew Young School of Policy Studies Research Paper No. 11-30.
McCaffer, R. (1979). Bidding behaviour. Quantity Surveying (New Zealand), 6-13. Milgrom, P. (1979). A convergence theorem for competitive bidding with differential
information. Econometrica, 47(3), 679-688.
Milgrom, P. (1989). Auctions and Bidding: A Primer. The Journal of Economic Perspectives, 3(3), 3-22.
Milgrom, P., & Weber, R. (1982). A Theory of Auctions and Competitive Bidding. Econometrica, 50(1), 1089-1122.
Ngai, S., Drew, D., Lo, H. P., & Skitmore, M. (2002). A theoretical framework for determining the minimum number of bidders in construction bidding competitions. Construction Management & Economics, 20(6), 473-482. Riley, J. G., & Samuelson, W. F. (1981). Optimal Auctions. The American Economic
Review, 71(3), 381-392.
Rothkopfm, H. (1969). A Model of Rational Competitive Bidding. Management Science, 15(7), 362-373.
Scherer, F. M., & Ross, D. (1990). Industrial Market Structure and Economic Performance (3 ed.). Boston: Houghton Mifflin.
Selten, B. (1973). A simple model of imperfect competition, where 4 are few and 6 are many. International journal of Game Theory, 2(1), 141-201.
Shepherd, W. G. (1982). Causes of Increased Competition in the U.S. Economy, 1939-1980. The Review of Economics and Statistics, 64(4), 613-626.
Shepherd, W. G. (1990). The economics of industrial organization (2 ed.). New Jersey: Prentice-Hall International.
Skitmore, M. (1991). The contract bidder homogeneity assumption: an empirical analysis. Construction Management and Economics, 9(5), 403-429.
Skitmore, M. (2002). Raftery curve construction for tender price forecasts. Construction Management and Economics, 20(1), 83-89.
Skitmore, M. (2013). kmth price sealed-bid auctions with general independent values and equilibrium linear mark-ups. Journal of the Operational Research Society, 65(12), 1864-1875. doi: 10.1057/jors.2013.163
Skitmore, M., & Pemberton, J. (1994). A multivariate approach to contract bidding mark-up strategies. Journal of the Operations Research Society, 45(11), 1263-1272.
Sweeney, S. (2009). Addressing market failure: Using transaction cost economics to improve the construction industry's performance. (PhD Thesis), University of Melbourne, Department of Civil and Environmental Engineering.
Teo, P. (2014). The effect of procurement on competition and flexibility : determining the suitability of public-private partnerships in major infrastructure projects. Queensland University of Technology.
Teo, P., Bridge, A., & Gray, J. (2013, 5-9 May 2013). A new first-order decision-making model for the procurement of public sector infrastructure: Procedures and Testing. Paper presented at the Proceedings of the 19th CIB World Building Congress, Brisbane 2013: Construction and Society, Queensland University of Technology, Brisbane, Australia.
Thaler, R. H. (1988). The Winner's Curse. The Journal of Economic Perspectives, 2(1), 191-202.
Thiel, S. E. (1988). Some Evidence on the Winner's Curse. The American Economic Review, 78(5), 884-895.
Thu, K., & Akintoye, A. (2007, 3-5 September 2007). Structure of the Public Private Partnership/Private Finance Initiative market in the UK construction industry. Paper presented at the 23rd Annual ARCOM Belfast, UK.
Vickrey, W. (1961). Counterspeculation, auctions, and competitive sealed tenders. The Journal of Finance, 16(1), 8-37. doi: 10.1111/j.1540-6261.1961.tb02789.x
Williamson, O. E. (1985). The economic institutions of capitalism: Firms, markets, relational contracting. New York, London: Free Press, Collier Macmillan. Wilson, R. (1977). A bidding model of perfect competition. The Review of Economic
Studies, 44(3), 511-518.
Winch, G. (2001). Governing the project process: a conceptual framework. Construction Management & Economics, 19(8), 799-808.
